Optical Properties of Wood: Measurement Methods and Result Evaluations (Smart Sensors, Measurement and Instrumentation, 45) 3031469054, 9783031469053

This book describes all optical properties of wooden materials, including definitions and measurement methods of optical

113 1 14MB

English Pages 323 [319] Year 2023

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Foreword
Preface
Acknowledgements
Contents
Abbreviations
1 Measurement Methods and Characterisation of the Optical Parameters of Wood
1.1 Introduction
1.2 Wood–Photon Interaction
1.3 Reflectance Measurement
1.3.1 Fundaments of Reflectance Measurement
1.3.2 Visible and Ultraviolet (UV) Spectrum of Wood
1.3.3 Near Infrared (NIR) Spectrum of Wood
1.3.4 Analytical (Middle) Infrared (IR) Spectrum of Wood
References
2 Measurement and Data Evaluation of Wood Colour and Gloss
2.1 Introduction
2.2 Colour and Its Measurement
2.2.1 Anomalies in the CIE Lab Colour System Regarding to the Colour Determination of Wood
2.3 Gloss and Its Measurement
2.3.1 Regularities Related to the Gloss Parameters of Natural and Treated Wood Surfaces
References
3 Applications of Colour Measurement in Wood Research
3.1 Introduction
3.2 Colour Modification by Dry Thermal Treatment
3.3 Steaming as a Colour Modification Process for Wood
3.3.1 Steaming Properties of Black Locust
3.3.2 Steaming Properties of Beech and Turkey Oak
3.3.3 Steaming Properties of Poplar
3.3.4 Steaming Properties of Scots Pine and Spruce
3.3.5 Steaming Properties of Larch and Sugi
3.4 Effect of Wetting and Finishing on Wood Colour
References
4 Monitoring of Wood Photodegradation by Colour Measurement
4.1 Introduction
4.2 Outdoor Weathering Test of Wood
4.3 Colour Change of Wood During Photodegradation
4.4 Species Dependence of Colour Change Caused by Photodegradation
4.5 Temperature Dependence of Colour Change During Photodegradation
4.6 Air Humidity Dependence of Colour Change During Photodegradation
4.7 Effect of Water Leaching for Photodegraded Wood
4.8 Photodegradation Properties of Thermally Modified Wood
References
5 Applications of IR Spectrum Measurement in Wood Research
5.1 Introduction
5.2 Monitoring the Chemical Changes Caused by Steaming
5.3 Examination of Chemical Changes Generated by Photodegradation
5.3.1 Basic Chemical Changes During Photodegradation
5.3.2 Effect of Temperature
5.3.3 Effect of Air Humidity Content
5.3.4 Effect of Water Leaching
5.3.5 Photodegradation Properties of Steamed Wood.
References
Index
Recommend Papers

Optical Properties of Wood: Measurement Methods and Result Evaluations (Smart Sensors, Measurement and Instrumentation, 45)
 3031469054, 9783031469053

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Smart Sensors, Measurement and Instrumentation 45

László Tolvaj

Optical Properties of Wood Measurement Methods and Result Evaluations

Smart Sensors, Measurement and Instrumentation Volume 45

Series Editor Subhas Chandra Mukhopadhyay, School of Engineering, Macquarie University, Sydney, NSW, Australia

The Smart Sensors, Measurement and Instrumentation series (SSMI) publishes new developments and advancements in the fields of Sensors, Instrumentation and Measurement technologies. The series focuses on all aspects of design, development, implementation, operation and applications of intelligent and smart sensors, sensor network, instrumentation and measurement methodologies. The intent is to cover all the technical contents, applications, and multidisciplinary aspects of the field, embedded in the areas of Electrical and Electronic Engineering, Robotics, Control, Mechatronics, Mechanical Engineering, Computer Science, and Life Sciences, as well as the methodologies behind them. Within the scope of the series are monographs, lecture notes, selected contributions from specialized conferences and workshops, special contribution from international experts, as well as selected PhD theses. Indexed by SCOPUS and Google Scholar.

László Tolvaj

Optical Properties of Wood Measurement Methods and Result Evaluations

László Tolvaj Institute of Wood Technology and Technical Sciences University of Sopron Sopron, Hungary

ISSN 2194-8402 ISSN 2194-8410 (electronic) Smart Sensors, Measurement and Instrumentation ISBN 978-3-031-46905-3 ISBN 978-3-031-46906-0 (eBook) https://doi.org/10.1007/978-3-031-46906-0 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

Foreword

Wood products have been widely used since ancient times for making homes, utensils and military equipment. After the first settlements and towns appeared, the basic furniture was developed using available local wood species. Wood materials have very favourable mechanical, thermal and aesthetic properties, which have ensured their popularity from ancient times to the present day. In particular, their aesthetic properties, such as colour and gloss, are unsurpassed in quality furniture manufacturing. Therefore, proper scientific design and production of the wood surface have become an indisputable necessity. In the world literature, there are many contributions to this subject, but no wellorganised summary book has yet been published. Reviewing, correlating, evaluating and compiling all the major results require a long period of research and a deep understanding of physical phenomena. This need is further strengthened by the fact that the colour determination is not free from some uncertainties in its definition and evaluation. Therefore, the direct examination of the reflection spectrum requires much more attention than at present. This book is the first attempt to summarise our existing knowledges in this important field of science. The author, Professor of Physics, has 40 years of continuous experience in the field of colour measurement and evaluation, colour modification and colour degradation caused by various influencing factors. Particularly important are his contributions to explore causal relationships related to colour degradations including chemical changes in the surface layer. He has introduced a number of new ways to develop and demonstrate the combined effect of several influencing factors on colour degradation. As an experienced researcher and teacher, he gives important advices for young researchers to avoid misinterpretations. I hope that this book will help and contribute to the practical application of the knowledge gained mainly over the last four decades and will promote new and successful ways for further development of this important field of wood science. Sopron, Hungary

György Sitkei

v

Preface

Wood has many excellent material properties, and therefore it is widely used for different applications. The wonderful aesthetic appearance is one of its advantageous properties. The colour inhomogeneity of wood is one of the most beautiful colour harmonies created by the nature. Attractive colour of solid wood is sensitive to the different biotic and abiotic effects. Measurement of optical properties can help to monitor and understand the changes generated by the different treatments. The optical properties of wood are of practical and scientific interest. The colour of wood is one of the most important parameters of wood products today. The appearance of wooden artefacts provides the viewer with a warm image. The little colour deviation between earlywood and latewood generates unique surface texture. The exceptional beauty and splendour of the Biedermeier cabinet are given by the appropriate combination of different shades. Machined surfaces of different wood species include lots of micromirrors aligned parallel to the grain. Gloss from these micromirrors provides a more elegant, soft, natural and beautiful texture than that of plastics or metal. The colour of the wood surface is sensitive to heat and light irradiations. Studying the optical properties of wood helps to understand the degradation mechanisms and provides information on the protection of the unique wooden surfaces. Colour measurement is becoming more and more important not only in the wood industry but in the food industry for quality control as well as in quality and moisture content determination in soil monitoring. Microscopes have limited magnification. The individual organic chemical groups (carbonyl, carboxyl, methyl, acetyl, hydroxyl) cannot be seen under a microscope. Photons are able to bring information from these groups based on their absorption properties. Spectroscopy helps us to determine the structures of the organic compounds, and it is a good analytical tool to follow the chemical changes of organic molecules during different treatments. Optical properties of wood are physical properties. Consequently, the physical aspects of the presented topics are dominant in the book. It is well-known that most physical changes have strong chemical background. This chemical background is discussed partly but not in depth in the book. That could be a topic for another book. vii

viii

Preface

The basis of this book is the work of myself and my research group, as well as the experience gained while reading plenty of papers dealing with our research topics. 40 years of research experience has brought a lot of observations and technical courses that may also be of interest to research colleagues. This book contains only the experimental results of my research group. There are no imported figures. My goal is to make suggestions for Ph.D. students and for young scientists in order to perform correct measurements. The content of the book is intended for graduate and postgraduate students, Ph.D. students, researchers, timber engineers, furniture restorers and application engineers involved in the design and manufacture of wood products. The book contains five main chapters. Chapter 1 deals with the measurement methods and characterisation of optical parameters. The very beginning of the book discusses the physical backgrounds of the photon-wood interaction. The chapter reviews the basics of reflectance measurement in the ultraviolet, visible, near- and middle-infrared radiation ranges. Several suggestions and recommendations are given for correct measurement and data processing. Chapter 2 presents the definitions and measurement methods of colour and gloss parameters. Although the L*, a*, b* colour parameters are independent variables, linear correlation was found between lightness and hue values in all discussed colour modification processes. Anomalies of dark wood samples in CIE L*a*b* system are also discussed. The chapter provides new, extended experimental results in the field of gloss measurement. The basic regularities of the gloss of natural and treated wood surfaces were established. Chapter 3 discusses the applications of colour measurement in wood research dealing with the colour modification effects of steaming, dry thermal treatment and surface wetting. The chapter contains detailed information regarding the steaming properties of eight wood species. Chapter 4 shows how effective colour measurement is for monitoring the sensitivity of different wood species to photodegradation. Colour has proven to be an excellent parameter for monitoring the effect of air temperature and humidity during photodegradation. Colour measurement was found to be a proper method for determining the photodegradation properties of thermally modified wood. Chapter 5 examines the applications of infrared spectrum measurements in wood research. These spectra provide information about the chemical changes generated by different treatments. A detailed discussion of the chemical changes generated by steaming and by photodegradation can be found in this chapter. The effect of influencing parameters during photodegradation, such as temperature, air relative humidity and leaching effect of rain, is also discussed. Finally, some recommendation for young scientists: It is important to know that most spectroscopic measurement methods and data manipulation techniques have validity limits and can generate anomalies in some cases. Scientists should keep these in mind and take care to avoid misinterpretations. The use of an incorrect method is not validated because it is present in the peer-reviewed literature. Never copy a method used by other researchers without scepticism, whether it is correct or not in your case.

Preface

ix

The author is especially grateful to Prof. G. Sitkei for reading the finished manuscript and giving many useful suggestions, for instance, for the development of generally applicable solutions. Special thanks to D. Varga Ph.D. for the valuable scientific and grammatical corrections in the manuscript. The author is also sincerely grateful to the staff of Springer Nature for their excellent cooperation. Sopron, Hungary

László Tolvaj

Acknowledgements

The author is grateful to Elsevier for the following permissions: Reprint from Journal of Photochemistry and Photobiology A: Chemistry, Volume: 329, Robert Nemeth, Laszlo Tolvaj, Miklos Bak, Tibor Alpar, Colour stability of oilheat treated black locust and poplar wood during short-term UV radiation. Pages: 287–292 (2016), with permission from Elsevier. Reprint from Journal of Photochemistry and Photobiology A: Chemistry, Volume: 348, Denes Varga, Laszlo Tolvaj, Satoru Tsuchikawa, Laszlo Bejo, Edina Preklet, Temperature dependence of wood photodegradation monitored by infrared spectroscopy. Pages: 219–225 (2017), with permission from Elsevier. Reprint from Journal of Photochemistry and Photobiology A: Chemistry, Volume: 356, Edina Preklet, Laszlo Tolvaj, Laszlo Bejo, Denes Varga, Temperature dependence of wood photodegradation. Part 2: Evaluation by Arrhenius law. Pages: 329–333 (2018), with permission from Elsevier. The author is grateful to Wood Research journal for the permission of reusing the scientific content of the following papers: Tolvaj, L., Molnar, S., Nemeth, R., Varga, D. (2010) Color modification of black locust depending on the steaming parameters. Wood Research 55(2):81–88. Tolvaj, L., Persze, L., Lang, E. (2013) Correlation between hue angle and lightness of wood species grown in Hungary. Wood Res 58:141–145. Kannar, A., Tolvaj, L., Magoss, E. (2018) Colour change of photodegraded spruce wood by water leaching. Wood Research 63(6):935–946. Preklet, E., Tolvaj, L., Banadics, E.A., Alpar, T., Varga, D. (2019) Colour modification and homogenisation of larch wood by steaming. Wood Research 64(5):811–820. Varga, D., Tolvaj, L., Preklet. E. (2021) Colour stability of steamed black locust, beech and spruce timbers during short-term photodegradation. Wood Res 66(4):544– 555.

xi

xii

Acknowledgements

The author is grateful to Acta Silvatica et Lignaria Hungarica journal for the permission of reusing the scientific content of the following papers: Tolvaj, L., Molnár, S. (2006) Colour homogenisation of hardwood species by steaming. Acta Silvatica et Lignaria Hungarica 2:105–112. Preklet, E., Tolvaj, L., Tsuchikawa, S., Varga, D. (2021) Photodegradation properties of earlywood and latewood spruce timber surfaces. Acta Silvatica et Lignaria Hungarica 17:9–21. Preklet, E., Tolvaj, L., Tsuchikawa, S., Varga, D. (2023) Colour modification of wood by dry thermal treatment between 90 °C and 200 °C. Acta Silvatica et Lignaria Hungarica 19(1):9–23. The author is grateful to Holztechnologie journal for the permission of reusing the scientific content of the following paper: Molnar, S., Tolvaj, L., Nemeth, R. (2006) Holzqualität und Homogenisierung der Farbe von Zerreiche (Quercus cerris L.) mit Dämpfung. Holztechnologie 47(5): 20– 23. The author is grateful to Springer Nature for the permission of reusing the scientific content of the following paper and 16 figures of the following book (Chap. 3): Tolvaj, L., Tsuchikawa, S., Inagaki, T., Varga, D. (2015) Combined effects of UV light and elevated temperatures on wood discoloration. Wood Sci Technol 49:1225–1237. Csanady, E., Magoss, E., Tolvaj, L. Quality of machined wood surfaces. 2015 Springer

Contents

1 Measurement Methods and Characterisation of the Optical Parameters of Wood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Wood–Photon Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Reflectance Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.1 Fundaments of Reflectance Measurement . . . . . . . . . . . . . . . . 1.3.2 Visible and Ultraviolet (UV) Spectrum of Wood . . . . . . . . . . 1.3.3 Near Infrared (NIR) Spectrum of Wood . . . . . . . . . . . . . . . . . 1.3.4 Analytical (Middle) Infrared (IR) Spectrum of Wood . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 2 6 6 19 27 33 46

2 Measurement and Data Evaluation of Wood Colour and Gloss . . . . . . 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Colour and Its Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Anomalies in the CIE Lab Colour System Regarding to the Colour Determination of Wood . . . . . . . . . . . . . . . . . . . 2.3 Gloss and Its Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Regularities Related to the Gloss Parameters of Natural and Treated Wood Surfaces . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51 51 52

3 Applications of Colour Measurement in Wood Research . . . . . . . . . . . 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Colour Modification by Dry Thermal Treatment . . . . . . . . . . . . . . . . 3.3 Steaming as a Colour Modification Process for Wood . . . . . . . . . . . . 3.3.1 Steaming Properties of Black Locust . . . . . . . . . . . . . . . . . . . . 3.3.2 Steaming Properties of Beech and Turkey Oak . . . . . . . . . . . 3.3.3 Steaming Properties of Poplar . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.4 Steaming Properties of Scots Pine and Spruce . . . . . . . . . . . . 3.3.5 Steaming Properties of Larch and Sugi . . . . . . . . . . . . . . . . . . 3.4 Effect of Wetting and Finishing on Wood Colour . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

91 91 92 105 108 117 125 130 133 141 153

68 76 78 88

xiii

xiv

Contents

4 Monitoring of Wood Photodegradation by Colour Measurement . . . . 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Outdoor Weathering Test of Wood . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Colour Change of Wood During Photodegradation . . . . . . . . . . . . . . 4.4 Species Dependence of Colour Change Caused by Photodegradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Temperature Dependence of Colour Change During Photodegradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Air Humidity Dependence of Colour Change During Photodegradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7 Effect of Water Leaching for Photodegraded Wood . . . . . . . . . . . . . . 4.8 Photodegradation Properties of Thermally Modified Wood . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

157 157 158 162

5 Applications of IR Spectrum Measurement in Wood Research . . . . . . 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Monitoring the Chemical Changes Caused by Steaming . . . . . . . . . . 5.3 Examination of Chemical Changes Generated by Photodegradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.1 Basic Chemical Changes During Photodegradation . . . . . . . 5.3.2 Effect of Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.3 Effect of Air Humidity Content . . . . . . . . . . . . . . . . . . . . . . . . 5.3.4 Effect of Water Leaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.5 Photodegradation Properties of Steamed Wood. . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

223 223 226

174 178 188 191 197 218

231 232 248 262 267 287 296

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

Abbreviations

2D A ATR C* CIE DR DRIFT E FT FTIR GU H h* IR KBr K-M L*, a*, b* LVL MC NIR OHT R R2 RH Rk Rm Rpk Rz S SD T

Two-dimensional Absorbance Attenuated total reflection Chroma Commission Internationale de l’Éclairage Diffuse reflection Diffuse reflectance infrared fourier transform Earlywood Fourier transformation Fourier transform infrared Gloss unit Heartwood Hue Infrared Potassium bromide Kubelka–Munk CIE colour coordinates Laminated veneer lumber Moisture content Near-infrared Oil-heat treated Reflectance Coefficient of determination Relative humidity Roughness parameter (vertical difference in the core section) Physical lightness Roughness parameter (average height of protruding peaks) Roughness parameter (average peak-to-valley height) Saturation Standard deviation Transmittance xv

xvi

TR UV X, Y, Z ΔE* φ λ0 II

Abbreviations

Transmission Ultraviolet Tristimulus values Total colour difference Radiation flux Intersection wavelength Parallel Perpendicular

Chapter 1

Measurement Methods and Characterisation of the Optical Parameters of Wood

Abstract The chapter deals with the measurement methods and characterisation of basic optical parameters. In the first section the physical backgrounds of the photonwood interaction are discussed. Definitions of basic optical parameters are introduced and the fundaments of reflectance measurement are presented. It also introduces the methods in regards to the determination of the absorbance data from the measured reflectance parameters. The chapter reviews the basics of reflectance measurement in the ultraviolet, visible, near- and middle infrared radiation ranges. Several suggestions and recommendations are given for correct measurement and data processing. Drawbacks of attenuated total reflectance measurement method for porous samples like wood is presented, together with the usual mistakes of data processing. Difference spectrum method is introduced as an excellent method to follow the chemical changes during different treatments in the infrared radiation region. It is also demonstrated how the position of maximum of overlapping absorption bands can alter if the intensity of the individual bands change a different extent. Keywords Wood · Diffuse reflection · Total reflection · Infrared spectrum · Visible spectrum · Difference spectrum

1.1 Introduction Determination of the optical properties of solid materials has old history. Newton presented his light and colour theory in Royal Society in 1672 generating extremely huge quarrel. His collected research results in the field of optics were published much later in 1704. In terms of wood science, Schramm’s (1906a, b) publications together with Wislicenus’s (1910) were among the earliest scientific papers dealing with the colour change of wood. Despite all these early scientific efforts, regular and comfortable usage of spectrophotometers and colorimeters started only after the spreading of computer usage. Although, the Michelson interferometer was designed to replace the prism of spectrophotometer at the beginning of the twentieth century, the first results in wood science measured using FTIR (Fourier transformed infrared)

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 L. Tolvaj, Optical Properties of Wood, Smart Sensors, Measurement and Instrumentation 45, https://doi.org/10.1007/978-3-031-46906-0_1

1

2

1 Measurement Methods and Characterisation of the Optical Parameters …

technique were published only in the last part of the century. The reason for this delay is that the Fourier transformation requires an appropriate computing capacity. Similar delay can be observed in the case of colour measurement as well. Computer supported portable spectrophotometers and colorimeters are available nowadays. Optical properties include the different interactions between a material and the photons of light. Light has both corpuscular and wave properties, therefore the basic knowledge of wave and corpuscular physics is essential to follow and understand the wood–photon interactions. The second section of this chapter summarise this basic information. Absorption properties can deliver information about the chemical structure of the studied material. As wood is an opaque material, reflectance properties can be mainly utilised to get information regarding the optical properties. Consequently, absorption properties of wood can be determined based on the reflectance measurement. Unfortunately, the intensity of the measurable reflected light is poor because of the diffuse reflection. For most of the twentieth century, the potassium bromide pellet method was used to measure the absorption properties of wood. Scientific results from reflectance measurements were only published in the field of wood science in the last part of the twentieth century. The use of the Michelson interferometer and computers enabled the measurement of a high-quality reflectance spectrum. This chapter provides a detailed overview of the definition and the measurement methods of optical parameters used in scientific and practical work. Suggestions are given for the correct measurement and examples are also presented showing common mistakes.

1.2 Wood–Photon Interaction Photon is a type of elementary particle. It is stable, massless and has no electric charge. Photons are the smallest possible packets of electromagnetic energy. Albert Einstein explained, light behaves as both a particle and a wave, with the energy of each particle of light corresponding to the frequency of the wave. Light is a flow of photons. The energy of these photons is the measure of their oscillation frequency, and the intensity of the light corresponds to the number of photons. Each photon has a wavelength and a frequency. The wavelength (λ, lambda) is defined as the distance between two peaks of the electric field with the same vector. The frequency (ν, nu) of a photon is defined by the oscillation frequency of the electric and magnetic field. Wavelength and frequency are related by the following equation: λ=

c ν

(1.1)

where: c is the velocity of propagation of light within the material. All photons travel at a constant speed (300,000 km/s) in vacuum. Photons as packets of energy can be transmitted over vast distances with no decay in energy or speed. Material objects

1.2 Wood–Photon Interaction

3

emit and absorb photons. Electrons of the material are involved into the emission and absorption of photons. As the energy of the electron changes, photon is emitted or absorbed at energies corresponding to the energy of the change. The energy of a free photon is determined by: E = hν

(1.2)

where: h is Planck’s constant (6.6 × 10–34 Js). This equation establishes a connection between the wave properties and the particle properties of light. Optical properties characterize the response of materials to incident electromagnetic radiation. This chapter provides an overview on important principle of optical properties of wood which involves the phenomena in physics of the interaction between light and material. Description of the light-matter interactions and how light behaves are illustrated by considering the five key factors: scattering, absorption, transmission, reflection, and refraction. As the light proceeds from one material into another several interactions can be observed, depending on the material properties. Some of the light radiation may be transmitted through the receiving material, some will be absorbed and scattered, and some will be reflected at the boundary between the two materials. Compared to the whole electromagnetic radiation range, only a small wavelength interval of the range containing infrared (IR), visible and ultraviolet (UV) radiations will be involved the discussion. The term of light will be used for UV and IR radiation as well because the physical properties of these radiations are similar as the properties of visible light. (Only the visible radiation is called light in most cases.) In general, materials could be categorized into three groups depending on the light-material interactions. Transparent materials are capable of transmitting light with relatively little absorption and scattering. Translucent materials can transmit light diffusely. In this case light is scattered within the material to some degree. Materials that are impervious, that is do not transmit light, are called opaque. Wood belongs to the third group. It absorbs and scatters the incident light heavily. The penetration depth of light into wood depends on the wavelength. This phenomenon will be discussed later. Optical properties that define the material response to the incident radiation can be described as transmissivity, reflectivity and absorptivity. The transmissivity of wood is extremely low. Even a very thin wood sample is translucent only. Translucent wood sample must be so thin that its thickness is usually smaller than the diameter of the largest cavities of the wood tissue. It means that these thin samples have holes where light can go through without contacting wood material. The main light absorber components of wood are the lignin macromolecules and the extractives. Removing lignin and extractives, wood becomes translucent (Yaddanapudi et al. 2017; Li et al. 2018, 2019; Mi et al. 2019; Rao et al. 2019; Bisht et al 2021). An atom can absorb or emit one photon when an electron makes a transition from one stationary state, or energy level, to another. Law of energy conservation determines the energy of the photon and thus the frequency of the emitted or absorbed light. The frequencies of absorbed light are well defined for individual atoms, but the

4

1 Measurement Methods and Characterisation of the Optical Parameters …

situation is much more complicated for macromolecules because of the perturbing effects of the surrounding atoms. The absorbance (A) and the transmittance (T) values can be measured and calculated easily by the following equations for transparent materials. A = log

I0 I and T (%) = 100 I I0

(1.3)

where: I is the transmitted intensity, and I 0 is the incident intensity. According to the Lambert law the absorbance is proportionate to the thickness of the sample. A = kl

(1.4)

where k is the absorption coefficient and l is the thickness of the sample. The Beer law states that the concentration of a chemical is directly proportional to its absorption coefficient. The united Beer-Lambert law gives the possibility to calculate the concentration of a solution by measuring its absorbance. Photons of light travelling within wood material can be absorbed or deviated from the straight trajectory by the atoms of the material. This second effect is called scattering. When radiation is scattered only by one localized scattering centre, this is called single scattering. Scattering in wood is usually multiple scattering. The scattering coefficient (s) was introduced into the theoretical description of diffuse reflection as a semi-empirical parameter to account for the internal scattering processes. The scattering coefficient is determined by particle size and refractive index of the sample. The scattering coefficient is slightly wavelength dependent. Shorter wavelengths are scattered more. Scattered photons do not follow the law of reflectance during “collusion”. Wonderful examples of light scattering are the blue sky and the red sunset. Light coming from the sun is made of all the colours of the rainbow. As this light hits the particles of air in the atmosphere, it is scattered in all directions. Blue light has a shorter wavelength than red light, so it is scattered much more than red light. When looking at the sky, one can see the multiple scattered blue light coming from all directions. The sunlight travels much longer distance in the atmosphere before sunset that at midday. Most of shorter wavelengths such as blue lights are already scattered out and only red light reaches our eye. Reflectance is an optical property of material, which describes how much light is reflected from the material in relation to an amount of light incident on the material. The reflection occurs always on the surface of the material, but for translucent materials multiple scattered photons in the volume of the material also join to the reflected photons to some extent. Photons reflected on the surface follow the law of reflectance. It means that the angle of the incident light beam is equal but opposite to the angle of the reflected beam. Both angles are measured to the normal vector (perpendicular to the surface) of the surface. On a smooth (polished) surface the reflection is called “specular” reflection or mirror reflection. Rough surface reflects the photons diffusely. Diffusely reflected

1.2 Wood–Photon Interaction

5

photons follow the law of reflection as well, however, the normal vector of the surface is changing from location to location in case of rough surface. Highly rough surfaces reflect the light in all directions. Reflectance (R) is defined as the ratio of the radiant flux (φr ) reflected from the surface or body of a material to the incident radiant flux (φi ). The radiant flux (power density of the radiation) is defined as the ratio of passed energy (ΔE) to the passing time (Δt).) φ=

φr ΔE and R(%) = 100 Δt φi

(1.5)

Scientists are often interested in the concentration (or mainly the change of concentration) of different chemical substances within the investigated material. The absorption of a material can usually be characterized by the absorption spectrum, which demonstrates the absorption values as a function of the wavelength, frequency or wavenumber. (The wavenumber gives the number of waves within one centimetre. Wavenumber is used in IR spectroscopy instead of wavelength.) Wood is an opaque material consequently, the Beer-Lambert law is not suitable to determine the absorption properties of its chemical components. The Kubelka–Munk (K–M) theory gives the possibility to determine the quotient of absorption coefficient (k) and scattering coefficient (s) by measuring the reflectance (R) value. The Kubelka-Munk equation is the follow. (The K-M function is presented in Fig. 1.5.) k (1 − R)2 = s 2R

(1.6)

Fortunately, the scattering coefficient hardly depends on the wavelength, thus the shape of the K–M function shows the wavelength dependence of the absorption coefficient. The Kubelka-Munk theory (Kubelka and Munk 1931; Kubelka 1948) was created to calculate the absorption properties of the surface layer using the reflected light. This theory was used by Kubelka and Munk for poorly absorbing materials. Wood is a good light absorber in some wavenumber intervals, non the less the K-M theory is widely applied to determine its light absorption. Only a few research papers recognised anomalous results generated using the K-M theory. These anomalous results were found in that positions where the absorption was high. Hembree and Smyrl reported the distortions of the K-M spectrum around intensive absorption bands of calcium carbonate and caffeine (Hembree and Smyrl 1989). Faix and Németh appointed that the DRIFT spectra evaluation of photodegraded wood is very difficult because the surface properties strongly influence the quality of the DRIFT spectra (Faix and Németh 1988). Zavarin et al. (1990) reported deviation in the intensity of the band at 1166 cm−1 measured in directions parallel and perpendicular to the grain. Tolvaj and Mitsui (2004) found that the intensity of the peak around 1173 cm−1 changed depending on the angle between the fibre direction and the incident IR light. The 950–1200 cm−1 IR region is where the wood has relatively

6

1 Measurement Methods and Characterisation of the Optical Parameters …

high absorption. Validity limits of the K-M theory in this region will be discussed later. Electron elevated to an excited stage try to return to a lower energy level as soon as possible. When the electrons jump back to the lower energy level and leave the excited state, energy difference will be emitted in the form of photon. The rapid emission is called fluorescence and the retarded emission is phosphorescence. Reflectance and absorbance spectra of wood will be demonstrated and discussed in this study.

1.3 Reflectance Measurement All liquid and solid materials reflect a fraction of incident radiation. In the IR wavelength region, electric conductors and non-conductors behave in a different manner. Conductors reflect the major part of the incident radiation without wavelength dependence provided that the surface is pure. Oxidized conductor surface shows wavelength dependent absorption, however. Non-conductors on the other hand absorb the major part of incoming radiation and this absorption is highly wavelength dependent. Most of conductors reflect the light equally in all visible wavelength interval. A surface with the same reflectivity for all wavelengths is called a grey surface. Surface of some conductors is close to gray surface. The surface of slate presents usually good gray surface. It is important to know that not the surface itself but a thin layer is responsible for the absorption. The reflected intensity consists of the photons reflected by the atoms of the surface and the scattered photons coming from a thin surface layer. Both reflectivity and absorptivity of a body depend on the surface temperature.

1.3.1 Fundaments of Reflectance Measurement As wood is an opaque material, traditional determination of absorbance using transmittance measurement would not be successful. Reflectance spectrum, however, can be measured and used for determining the absorbance spectrum. Although, the reflected light from each individual incident photons follow the law of reflection, material roughness effects that each individual photon meets a surface that has a different orientation. Thus, the normal line at the point of incidence varies from point to point depending on the surface properties causing an individual reflection for each and every photon. Subsequently, when the photons reflect off over the rough surface according to the law of reflection, they scatter in different directions. The result is that the concentrated bundle of incident rays of light is shown diffused reflection. Clearly, roughness properties of the reflecting surface strongly determine the concentration and angle dependence of the reflected light. This phenomenon is important for designing the device which collects the reflected light for the detector. One possible diffuse reflectance unit is presented in Fig. 1.1. This type of diffuse

1.3 Reflectance Measurement

7

Fig. 1.1 Schematic presentation of a possible arrangement of the diffuse reflectance collecting unit

reflectance unit built in the spectrophotometer is unable to collect all the reflected photons. The collection of reflected photons occurs only within a well-defined square angle which is determined by the size and the position of the spherical mirror. Any changes in the surface properties modify the number of the mirror-collected photons modulating the measured reflectance values. When using this type of unit, it has to be considered that the optimum collection can be assured only if the surface of the sample is located on the same position where the top of the sample holder is. Figure 1.1 illustrates that the spherical mirror collects reflected light only from a mirror determined spherical angle. There is a need for a device that can uniformly collect all reflected light from a sample. The integrating sphere with its highly reflective enclosure solves this need (Fig. 1.2). The light reflected by the sample enters the sphere, bounces around the highly reflective surface of the sphere wall and after multiple reflections hits the detector. The integrating sphere needs to be coated with the highest possible reflective surface for the desired wavelength region. The coating must be uniform and close to being a perfect Lambertian scatterer. Specially designed integrating sphere can exclude specular reflection by blocking it. The modern semiconductor industry can produce powerful and highly sensitive detectors working in the UV, visible and near infrared (NIR) wavelength regions. These detectors are sensitive to different 10 nm wavelength intervals. A properly built-in series of detectors in integrating sphere can measure the reflectance spectrum directly. Colorimeters use this arrangement as well. Traditional dispersive spectrophotometers can disperse the white light of the light source by prism or optical grid. The disadvantage of a dispersive spectrophotometer is that it applies a narrow slit to achieve high spectral resolution. This narrow slit reduces extremely the intensity of the passed light. Applying 8 cm−1 resolution in IR spectroscopy, only 8/3600th part (0.2%) of the incident radiation reaches the detector (Michell 1988). That is why it is not possible to measure correct reflectance spectrum by dispersive spectrophotometer.

8

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.2 Schematic presentation of an integrating sphere

The Michelson interferometer and the computer jointly solved this problem. The prism (or grid) was replaced by the Michelson interferometer to generate powerful signal for reflectance measurement. The moving mirror of the interferometer generates an interferogram during some second movement. This interferogram undergoes Fourier transformation (FT) by the computer and the result will be the actual spectrum. This powerful technique is suitable to determine the reflectance spectrum of wood. Although, the Michelson interferometer existed as early as the beginning of the twentieth century, the first results measured by FTIR technique were published in the last part of the century in the field of wood science. The reason was the need of a computer for Fourier transformation. (The working system of these spectrophotometers is not discussed in this study only the special attachments are presented.) Traditional good quality spectrophotometers use the double beam system (reference and sample beams). The reference and the sample modified intensities are measured parallel in this system. This system does not need extremely stable light source and detector. Since the modern electronics systems can guarantee the stability

1.3 Reflectance Measurement

9

of the light sources and the detectors, one beam systems can assure correct measurement. Nevertheless, proper background spectrum is needed for the good quality measurement in case of one beam system. The background spectrum contains information in terms of the emission intensity distribution of the light source together with the wavelength sensitivity of the detector, moreover it incorporates important absorption or reflection characteristics of all materials where the light beam travels through, including the air. Therefore, background spectrum must be measured in advance of sample measurement. For transmittance or ATR (attenuated total reflection) method, empty sample holder is used for background measurement. The situation is more complicated for diffuse reflectance measurement, because a diffusely reflecting material must be applied (without any absorption and emission in the investigated wavelength interval) for background spectrum determination. KBr (potassium bromide) powder is proper material for IR background measurement. KBr powder has no absorption in the IR wavelength region and it has diffusely reflective surface. Disadvantage of KBr powder is that it is heavy water absorber. Therefore, only the freshly dried KBr powder is appropriate as a background material. Rough metal surface can be a good background material as well. The same configuration of parameters (wavelength interval, resolution, number of scans) must be used for both background and sample measurement. Rough surface is preferable. The mirror reflection must be minimalized. The spectrophotometer displays the measured reflected intensities. These intensity values can be used for background material selection. Better material results in greater intensity. Poorly reflecting background material produces noisy background spectrum and the noises will transform the spectrum of the sample as well. Measured sample reflection values above 100% show that the chosen background material was not the proper one. These values represent that the sample has better reflectivity than the background material does. Recording new background is recommended before all measurement series. Another measurement technique is the attenuated total reflection method (ATR). When a light beam travels from a medium of high refractive index (crystal) to a medium of low refractive index (sample) the beam is reflected totally back from the sample to the crystal if the incident angle of the beam is high enough. The light penetrates a small distance beyond the surface of the sample and the intensity of beam is partly reduced according to the absorption properties of the sample. This method requires excellent contact between the crystal and the measured material and is therefore an excellent method for liquids or soft, easily deformable solids. Wood having high porosity does not fulfil this requirement. ATR technique is only suitable to measure the spectrum of small wooden chips. The ATR spectrum differs from a classic transmission spectrum because the light penetrates the sample at a depth proportional to the wavelength. Consequently, the absorbance in an ATR-measured spectrum at longer wavelengths are proportionally greater then at shorter wavelengths when compared to a classic transmission spectrum generated absorbance. This problem can be eliminated by mathematical manipulation. Most of the spectrophotometer programs offer this manipulation. The photoacoustic method is also suitable to determine the IR absorption spectrum of wood. Basic theories of the photoacoustic effect and photoacoustic signal

10

1 Measurement Methods and Characterisation of the Optical Parameters …

generation with solids have been given by Rosencwaig (1975). This method is hardly used in wood science. It is mentioned here only as a possibility to determine the IR spectrum. The short description of photoacoustic method is the follow. A pulsed IR beam is partly absorbed by the sample, depending on the absorption properties. The frequency of pulsed IR beam must be within the acoustic frequency range. The absorbed energy heats the sample resulting in pressure waves. These signals are detected by a sensitive microphone generating the absorption spectrum. The photoacoustic method does not need special sample preparation. Even samples with rough surfaces can be easily measured, since the photoacoustic signal is proportional to the absorbed energy (Kuo et al. 1988; Pandey and Theagrajan 1997; Ohkoshi 2002; Yamauchi et al. 2004; Noupponen et al. 2004; Pandey and Vuorinen 2008). The presentation of absorbance spectrum is not standardised in field of wood science. The wavelength is used in UV and visible light spectroscopy. The wavelength and wavenumber are parallel applied in NIR spectroscopy. In the analytical IR region, the wavenumber is used exclusively nowadays, while both wavelength and wavenumber were utilised simultaneously 30–50 years ago (Tolvaj 2022). The wavenumber itself is acceptable because it represents one kind of frequency, only the use of the unit 1/cm does not fit to the International System of Units (SI). In the case of wavelength either nanometre (nm) or micrometre (micron, μm) is SI unit. The correlation between wavenumber f (cm−1 ) and wavelength λ(nm) is determined by the following equation: ) ( f cm−1 · λ(nm) = 10, 000, 000

(1.7)

The real question for wood scientists is which parameter represents better the IR spectrum of wood. Figure 1.3 shows the K-M spectrum of poplar (Populus x euramericana cv. Pannonia) measured by the diffuse reflectance method as function of wavenumber, while Fig. 1.4 presents the same spectrum as function of wavelength. The two spectra contain the same K-M values, however, their appearance is different. The fingerprint region (right side) is compressed, and the hydroxyl-methyl region (left side) is enlarged in Fig. 1.3. Exactly the opposite can be observed in Fig. 1.4, where the fingerprint region is expanded, and the hydroxyl-methyl region is compressed. These appearances are generated by the reciprocal relationship between wavenumber and wavelength. The fingerprint region contains plenty of overlapped absorption bands. This region occupies 29% of the horizontal axis if the independent variable is wavenumber. In contrast, the fingerprint region takes up 73% of the horizontal axis if the independent variable is wavelength. Meanwhile the horizontal interval of hydroxyl-methyl region decreases from 36 to 13%. The expansion is beneficial for a more detailed investigation of the fingerprint region, while 13% of the scale is a sufficient proportion for representing the hydroxyl-methyl region. The main advantage of presentation by wavelength is that the most important fingerprint region is enlarged. Although Figs. 1.3 and 1.4 do not display the whole analytical IR region, the introduced interval contains all important absorption bands of wood. It can be concluded that the presentation of the IR absorption spectrum of wood is much more useful by wavelength than by wavenumber provided that the whole

1.3 Reflectance Measurement

11

wavelength interval is shown in one diagram. On the other hand, both representation (by wavelength or by wavenumber) could give similarly detailed information if a zoomed (such as fingerprint) region is demonstrated. Most of the measurement methods and data manipulation techniques have validity limits, and in some cases, they can generate even anomalies. Scientist must keep in mind them and take care of avoiding wrong interpretations.

Fig. 1.3 Absorption spectrum of poplar in the 1000–3800 cm−1 wavenumber interval

Fig. 1.4 Absorption spectrum of poplar in the 2500–10,000 nm (or 2.5–10 μm) wavelength interval

12

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.5 The curves of K-M, 1/R and Log(1/R) functions

The challenge of diffuse reflectance type measurement is how to convert the reflectance values to absorbance values. There is a hyperbolic (reciprocal) relationship between the reflectance values and absorbance values suggesting that the reciprocal of reflectance (1/R) values gives the absorbance values. The other possibility to calculate the absorbance spectrum is the application of the K-M equation presented in Sect. 1.1. (Eq. 1.6). The curves of 1/R, Log(1/R) and K-M functions are presented in Fig. 1.5. The form of the curves is similar. Vertical value of starting points is different at the right corner. This spot represents 100% reflectance and zero absorbance values. The Log(1/R) and the K-M function provide it perfectly. However, the 1/R function gives one unit value instead of zero. The distance between the curves is increasing with decreasing reflectance values. These distances between 1/R and K-M values are 3.4, 6 and 31 (this last value is not visible in Fig. 1.5) units at 20, 10 and 2.5% reflectance values, respectively. This phenomenon modifies the internal differences between the two types of absorbance values. It means that the absorbance values calculated by 1/R and K-M equation will be close to each other at high absorbance values and fare at low absorbance values. This effect can be seen in Fig. 1.6 where the absorbance values of poplar are presented using the 1/R and the K-M functions. The form of the K-M function is close to linear in the 0.12–1 reflectance interval (12–100%) resulting that the K-M transformation does not generate distortion in this reflectance interval. However, the calculated absorbance values start to increase rapidly if the reflectance value is below 10%. Small reflectance intensity change can lead to a great absorbance intensity change. Surface roughness increase for example can result in reflectance value decrease which will then be transformed into absorbance increase by the K-M function, obviously without any real absorption increase. Therefore, Fig. 1.5 shows clearly that 10% reflectance value is the validity limit of the K-M function. This limit is 20% reflectance value for the 1/R function.

1.3 Reflectance Measurement

13

Fig. 1.6 Absorbance spectra of poplar calculated using K-M and 1/R functions

The Log(1/R) function is close to linear in the whole reflectance interval suggesting that this function could be used to calculate the absorbance values properly. Figure 1.6 shows clearly that the 1/R function provides the same places for all maxima as the K-M function does, however, the intensity relations are different. These results confirm that the 1/R function is not suitable to determine the absorbance values of wood material. Figure 1.7 shows the comparison of absorbances calculated by the K-M and the Log(1/R) functions. Calculated values of the Log(1/R) spectrum were smaller than the values of the K-M spectrum. Comparing the two spectra, the methyl band at 3443 nm was the position where the Log(1/R) was normalised (multiplied 1.5 times) to the K-M spectrum for correct comparison. It is visible that the Log(1/R) function gives the same places for all maxima as the K-M function does, however, the intensities are different. The values of the K-M spectrum are higher at those places where the absorbance values are high comparing to the values of the Log(1/R) spectrum. These are the positions where the reflectance values are low. The reflectance values around 2890 and 8500 nm (3460 and 1176 cm−1 ) are close to 20%. Looking at the K-M function in Fig. 1.5, it is already slightly curved around 20% reflectance value. This phenomenon suggests that the K-M transformation can generate distortion even at 20% reflectance values. This is a valuable information if we need the absolute values of absorption. However, it can be neglected if the changes are important generated by a treatment. The difference spectrum is not sensitive to such kind of distortion. It is important to keep in mind that the decision is always in the researcher’s hand regarding which measurement method to apply for determining the absorbance spectrum of a wood sample. Spectrophotometer devices offer different data presenting possibilities (transmittance, reflectance, absorbance and K-M values). It is important to choose the proper one. Naturally, choosing any of the available options, the

14

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.7 Absorbance spectra of poplar calculated by K-M and Log(1/R) functions

computer calculates a spectrum using the measured intensities without considering whether it is reasonable or not, and the spectrum will be named according to the chosen type. Of course, it may as well lead to a wrong interpretation of the optical properties or procedures. For example, there are papers presenting the transmittance spectrum of block type wood sample which by nature does not make any sense since wood is not transparent for light. Here, it should be emphasized that computers do not doubt the methodology, but only complete the program; no other than the scientist should rise the question if the principle applied is reasonable. Spectrophotometers usually use the Log(1/R) function to calculate the absorbance values. Nevertheless, it is worth to check the computer applied calculation when using a spectrophotometer for the first time. The absorbance spectrum is actually calculated according to the Beer-Lambert law, which employs the transmittance values. Wood is an opaque material, consequently a very thin layer of wood is a proper sample for transmittance measurement. A 0.07 mm thin sugi (Cryptomeria japonica D. Don) sample was prepared to measure the transmittance (TR) spectrum. The highest transmittance value was 5% in the 4000–5500 nm (2500–1818 cm−1 ) interval and the lowest was almost zero at the places of high absorbances. The measured (TR) absorbance spectrum is demonstrated in Fig. 1.8 along with the absorbance spectrum of block type sugi sample measured by diffuse reflectance (DR) measurement method and calculated by K-M equation. Both spectra were two-point baseline corrected at 2600 and 5200 nm wavelengths (3846 and 1923 cm−1 ). The comparison of these two spectra can validate the usefulness of K-M transformation for determining the absorption spectrum of wood material. The spectra run close to each other in most part of the presented wavelength interval. The only exception is the 9000–11,000 nm (1111–909 cm−1 ) interval. Two high peaks are visible at 9430 and 9680 nm (1060 and 1033 cm−1 ) for

1.3 Reflectance Measurement

15

the absorption spectrum measured by TR method. These peaks are hardly visible on the absorption spectrum measured by DR method. This anomaly is generated by the DR measurement method not by K-M transformation. This statement is confirmed by Fig. 1.9 as well.

Fig. 1.8 Absorbance spectra of sugi measured by transmittance (TR) method and diffuse reflectance (DR) method. The DR spectrum was calculated by K-M transformation

Fig. 1.9 Absorbance spectra of sugi measured by transmittance (TR) method and diffuse reflectance (DR) method. The DR spectrum was calculated by Log(1/R) function

16

1 Measurement Methods and Characterisation of the Optical Parameters …

Figure 1.9 reveals the absorption spectra of sugi measured by TR and DR methods. The DR spectrum was calculated by Log(1/R) function. Both spectra were twopoint baseline corrected at 2600 and 5200 nm wavelengths (3846 and 1923 cm−1 ). The calculated values of Log(1/R) spectrum were smaller than the values of TR absorbance spectrum. Comparing the two spectra, the methyl band at 3443 nm (2904 cm−1 ) was the position where the Log(1/R) was normalised (multiplied 1.5 times) to TR absorbance spectrum for correct comparison. Investigating side by side these two spectra, the usefulness of the Log(1/R) function for determining the absorption spectrum of wood can be validated. The spectra are located close to each other in most part of the presented wavelength interval. The places of maxima are the same, however, the internal differences among the peak intensities are visible. The Log(1/ R) generated DR spectrum has smaller internal differences. The only exception is the 9000–11,000 nm (1111–909 cm−1 ) interval similarly to the K-M transformation generated spectrum (Fig. 1.8). Figures 1.8 and 1.9 together demonstrate that the anomaly in the 9000–11,000 nm interval is generated by the reflectance measurement. Fortunately, this region is not crucial when analysing the main components of wood. This anomaly of DR measurement was recognised by many scientists earlier (Zavarin et al.1990; Anderson et al. 1991a, b, c; Michell 1991; Faix and Böttcher 1992; Pandey and Theagrajan 1997). It can be concluded that both K-M equation and Log(1/R) function are suitable for determining the absorbance spectrum of wood materials, however, it is important to keep in mind the validity limits and the anomalies generated by these methods. The problems of ATR measurement technique were discussed in a recent paper (Tolvaj 2022). The ATR measurement method requires perfect contact between the crystal and the measured material. It is an excellent method for determining the absorption properties of liquids and soft, compressible solid materials. Wooden blocks (as samples) are not soft enough so that the close contact between the crystal and the sample cannot be ensured by pressure. The only usable crystal which could tolerate high pressure is diamond, however, the active surface of the diamond crystal is not larger than 1 mm2 . Consequently, if the surface of the sample is considerably larger than that of the crystal, the increasing pressure does not generate closer contact between the crystal and the sample. This is because the sample holder around the crystal “absorbs” the pressure, thus the contact remains poor. Using a small wooden chip, the ATR technique gives better spectrum than for solid wood sample. Figure 1.10 shows the absorbance spectra of poplar wood in both block and chip format (the diffuse reflectance FTIR spectrum of the same block sample is presented in Fig. 1.4). The chip was 3 mm in diameter and 0.5 mm thick. The surface of the block was 30 × 10 (mm2 ) and it was 5 mm thick. The active surface of the diamond crystal was less than 1 mm2 . The spectrum of block type sample gives all information visible in chip type spectrum, but the intensities are extremely low in case of the spectrum of the block. Low intensities are because of the poor contact between the sample and the crystal. The results represented in Fig. 1.10 demonstrate that the ATR technique is not suitable to determine the absorption properties of block type wood sample.

1.3 Reflectance Measurement

17

Fig. 1.10 ATR spectra of poplar chip and block sample (the wavelength dependence of the absorption was corrected for both spectra)

Another interesting detail in Fig. 1.10 is that there is a small negative peak visible close to 4500 nm. The negative band is abnormal in case of an absorption spectrum. The reason is the background spectrum measurement failure. If the surface of crystal is not clean during the background spectrum measurement the recorded background spectrum will contain absorbance information according to the pollution. These absorbance values are missing during sample spectrum measurement, hence generating negative absorbance values. The penetration of the radiation into the sample during total reflection is wavelength dependent. Longer wavelengths penetrate deeper than shorter wavelengths. The penetration decrease reduces the absorption value. This phenomenon can be seen in Fig. 1.11 (not corrected spectrum). The absorbance value of hydroxyl groups at 2890 nm is low although the concentration of hydroxyl groups is relatively the highest in wood. Consequently, the absorbance at 2890 nm should be the highest. Both diffuse reflectance and KBr pellet method deliver this requirement (Zavarin et al. 1990; Anderson et al. 1991c; Michell 1988; Tolvaj and Faix 1995; Zanuttini et al. 1998; Huang et al. 2008; Tolvaj et al. 2014) (see also in Fig. 1.4). This distortion of ATR measurement can be minimalized by wavelength dependence correction which can be performed by any FTIR spectrophotometers. Figure 1.11 shows how the correction eliminates the distortion. Although, this wavelength correction of ATR spectra is important to get correct intensity ratios among the bands, many scientists fail to do it. Comparing the IR spectra of a poplar block sample measured by both the ATR and diffuse reflectance technique, some anomalies can be seen. Figure 1.12 shows these spectra. The presented ATR spectrum was subjected to wavelength dependence correction, and the DR spectrum was baseline corrected. The measured values of the ATR spectrum were much smaller (approximately 9 times) than the values of DR

18

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.11 ATR spectra of poplar chip before and after wavelength dependence correction

spectrum. Comparing the two spectra, the methyl band at 3435 nm was the only position where the intensity of ATR spectrum was high enough and the place of maximum was equal to that of the DR spectrum. This place was used to multiply the ATR spectrum to the DR spectrum for correct comparison. Both spectra give the bands and their position correctly in the 5500–7800 nm wavelength (1818–1282 cm−1 ) interval, but the intensity differences within the ATR spectrum are extremely small. The DR spectrum shows much more differentiated intensities. The left part of the hydroxyl band below 3000 nm (3333 cm−1 ) is missing in the ATR spectrum, whereas a huge peak is visible at 9710 nm (1030 cm-1 ). This latter band modifies the bands located in the 8400–9500 nm (1190–1053 cm−1 ) interval. Two strong characteristic bands of wood are completely missing at 8540 and 8850 nm (1171 and 1130 cm−1 ), representing the absorption of ether linkages. This is an anomaly generated by the ATR measurement. Unfortunately, the DR measurement method also generates an anomaly in this region. The absorption band of C-O stretching at 9710 nm (1030 cm−1 ) is completely missing in the spectrum measured by DR measurement method. (In contrast, it is the highest band of ATR spectrum.) Fortunately, this region is not crucial for the analysis of the main components of wood. The magnification of the ATR spectrum raised the values in the 4000–5500 nm (2500–1818 cm−1 ) interval where there is not real absorption. The same happened below 2700 nm (3704 cm−1 ) as well. Figure 1.12 shows that the DR technique provides a much more detailed spectrum in terms of the main region of the analytical IR interval compared to the ATR technique. The small measured area (≈ 1 mm2 ) is a disadvantage of the ATR technique if an inhomogeneous material (such as wood) surface is measured. It is hardly possible to measure exactly the same surface area before and after treatment. The ATR technique featuring a diamond crystal generates an additional anomaly. Diamond has a wide and strong absorption band in the middle of the analytical IR region. This absorption can be visualised by measuring the background spectra of the

1.3 Reflectance Measurement

19

Fig. 1.12 Absorbance spectrum of block type poplar sample measured by ATR and DR technique

FTIR spectrophotometer with an empty sample compartment than with a diamond ATR crystal. These two spectra are visible in Fig. 1.13. The background intensities measured with a diamond crystal are approximately 20 times smaller than the background intensities measured with an empty sample compartment. The background values of diamond were multiplied to generate equal values at the 2680 cm−1 wavenumber (3731 nm) for the correct comparison. Diamond shows strong absorption between 1600 and 2600 cm−1 (6250 and 3846 nm). This absorption reduces the sensitivity of the ATR measurement in this region considerably. Fortunately, wood hardly has any absorption in this region, only the unconjugated and conjugated carbonyl region is affected between 1600 and 1800 cm−1 (6250 and 5556 nm). The sensitivity decrease in the absorption region of diamond is well demonstrated by the absorption band of CO2 at 2360 cm−1 (4237 nm). The background measured with an empty sample compartment shows high absorption caused by CO2 . The background measured in presence of the diamond crystal exhibits only a small negative peak representing the absorption of CO2 . (The two background measurements were performed after each other.)

1.3.2 Visible and Ultraviolet (UV) Spectrum of Wood Electromagnetic radiation in the 180–2500 nm wavelength interval is characterised with uniform physical properties. Only the human perception can divide this interval into three individual regions. The visible interval is located between 380 and 750 nm. These boundaries are not sharply defined and may vary depending on the individuals. The perception is week close to the boundaries. That is why the 400–700 nm interval

20

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.13 Background spectra of the FTIR spectrophotometer measured with an empty sample compartment (Air) and with a diamond ATR crystal. The ATR diamond crystal provided background was multiplied by 20 to fit to the air background at 2680 cm−1

is defined in everyday usage as visible interval. Radiation with shorter wavelengths (180–400 nm) belong to the ultraviolet (UV) rang while those with longer wavelengths (700–2500) can be referred to as near infrared (NIR) light or radiation. (The real border of UV radiation is 10 nm but the 10–180 nm interval can be investigated only in vacuum because of the strong absorption by air.) The spectroscopically usable UV and visible wavelength interval is only 570 nm wide (between 180 and 750 nm). This short wavelength interval has limited spectroscopical utility. The absorption spectrum of an atom absorbing in the UV and visible region consists of very sharp lines at well-defined wavelengths. The ground and excited states of atoms are well-defined. For molecules, however, the UV and visible absorption occurs over a wide range of wavelengths, because they have extended ground and excited states. This system allows many possible transitions, each differing from the others only slightly in terms of the wavelength of radiation exciting the molecules. Radiation at these wavelengths generate a band of absorption centred at the wavelength of the major transition. Reflectance spectrum of wood does not show separate absorption peaks in the visible region. Wood reflects most of the photons belonging to the red–orange region and absorbs most of them in the blue-violet region of the visible light. Considerable differences have been found among the wood species regarding their reflectance spectrum in the visible radiation range (Fig. 1.14). Light species (spruce and ash) reflects much more photons than the dark species (Karri, Eucalyptus diversicolor and Scots pine heartwood). Dark species absorb more light than the light species, mainly in the blue-green region. Also, the form of the reflectance curve refers to the colour of the measured sample as the surface colour is

1.3 Reflectance Measurement

21

Fig. 1.14 Reflectance spectra of different species plotted in the visible region (EH; earlywood of heartwood, LH; latewood of heartwood)

determined by the reflected light. Therefore, the connection between the reflectance spectrum and the colour will be explained in detail hereinafter. Reflectance curves can be described with acceptable accuracy in the following general equation (Sitkei 2013): R = a + bλn

(1.8)

where R is the reflectance, λ is the wavelength and a, b and n are constants characterizing a given wood species. The evaluation of this equation is presented here according to Csanady et al. (2015, Sect. 3.4). It is interesting to mention that a surface whose absorptivity or reflectivity is the same for all wavelength is called a gray surface. In this case the exponent n in Eq. (1.8) is equal to zero. Concerning the existing different wood species, the exponent n varies in a wide range between 0.4 and 4.0. Considering the spectral reflectance curve of different wood species presented in Fig. 1.14, the following general rule can be observed. Starting from a true gray colour, the refection is continuously decreasing mostly in the 400–500 nm range and the colour is also changing from gray-yellow to yellow and further to yellow, yellow–brown, brown, brown–red and finally to red. At the same time, the exponent n varies from 0.5 (gray–yellow) to 3.0 (brown–red). The use of Eq. (1.8) allows a convenient mathematical treatment for revealing some basic relationships between the shape of the reflection curve and the colour hue belonging to it. The area under the reflection curves defines the true lightness referred to as physical lightness Rm . The value of perceived lightness (defined by CIE L*) is always smaller than the value of physical lightness because of the selective

22

1 Measurement Methods and Characterisation of the Optical Parameters …

wavelength sensitivity of the human eye. The physical lightness is given by integration of the reflection curve (Eq. 1.8) in the visible region from 400 to 700 nm wavelength (Eq. 1.9). [ Rm = a · λ +

b λn+1 n+1

]700 (1.9) 400

It is worth to put / the origin of the coordinate system to the wavelength of 400 nm and to take x = λ 100. Due to the specific shape of the curves, the average reflectance value R, as a horizontal line, intersects the reflection curve at a wavelength λ0 uniquely characterizing the spectral distribution of reflectance values. Figure 1.15 demonstrates the meaning of λ0 , R and Rm . The value of physical lightness Rm is given by the area of a rectangle determined by R and the 400–700 wavelength interval. The point of intersection has special meanings. The reflectance value of this point (R) determines the value of physical lightness (Rm = 300 R) and it determines the λ0 which corresponds to the colour hue. Now the main target is to determine the λ0 direct from the data of the reflectance spectrum. The point of intersection λ0 can be calculated as ( λ0 = 100

Rm − a b

)1/n + 400

(1.10)

The intersection point does not depend on the initial intersection of the reflection curve at 400 nm and therefore, a simpler equation can be derived (Sitkei 2013)

Fig. 1.15 Schematic presentation of λ0 , R and Rm . The dotted line is the reflectance spectrum

1.3 Reflectance Measurement

23

Fig. 1.16 Characteristic wavelength λ0 as a function of exponent n and the corresponding colour hues of wood samples. (Reproduced with the written permission from Springer Nature)

λ0 =

300 + 400 (n + 1)1/n

(1.10a)

Equation (1.10a) shows that the value of λ0 is uniquely determined by the exponent n and the relationship is plotted in Fig. 1.16. In practice, the exponent n varies from zero to 5. The ideal grey body with constant reflection independently of wavelength has the exponent n = 0 and a deep red coloured body has exponent up to 5. The lowest value of λ0 is 510 nm and the maximum value is 610 nm. Analysing wood samples of different colour, a correspondence could be established between λ0 values and colour hues given also in Fig. 1.16. The next problem is the determination of the exponent n from the spectral reflectance curves for each wood species. Using the reflectance data, the curves can be fitted according to Eq. (1.8) with standard procedures. The plot of reflection curves on double logarithmic scale has revealed that the exponent n cannot always be considered as a constant, however. Furthermore, to perform the approximation of very large data system is a time-consuming procedure. In order to solve this problem, another characteristic number, a shape factor (K1 or K2 ) was searched which is uniquely related to the exponent n. One possible choice was the following number composed of distinct ordinates of the reflection spectrum: K1 =

R550 − R450 R650 − R550

(1.11)

where the subscripts refer to the given wavelength. This number is independent of the initial section of the reflection curve (a) because the following equality holds: R550 − R450 (R550 − a) − (R450 − a) = R650 − R550 (R650 − a) − (R550 − a)

24

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.17 Theoretical relationship between K and λ0 with colour hue ranges. (Reproduced with the written permission from Springer Nature)

Therefore, the characteristic number K 1 is/the only function of the exponent n. Using Eq. (1.8) and taking a = 0 and x = λ 100, Eq. (1.11) can be expressed in the following form: K1 =

1.5n − 0.5n 2.5n − 1.5n

(1.11a)

Both K and λ0 are the function of the exponent n only, therefore, a uniquely defined relationship between K 1 and λ0 can be established, which is given in Fig. 1.17. The proof of usability of this relationship will be given later using and plotting the corresponding measurement results in the same coordinate system. The characteristic number K can be calculated in a somewhat different manner as well. The ordinates R450 , R550 and R650 are the central ones in the three sections of the reflection spectrum. Assuming a linear spectral distribution, the ordinate R450 is numerically equal to the average reflection between 400 and 500 nm, that is the integrated area under the reflection curve in this section (Rm1 ). In other cases, when n /= 1, the central ordinates do not accurately coincide with the average reflection values. Therefore, another possibility is to express the characteristic number K with the corresponding average reflections. The average reflections in the three sections can be calculated quite similar to Eq. (1.9), integrating Eq. (1.8) between the pertinent upper and lower limits. Replacing the ordinates R450 , R550 and R650 with the corresponding average reflections Rm1 , Rm2 and Rm3 , the following theoretical relationship is obtained K2 =

2n+1 − 2 3n+1 − 2n+2 + 1

(1.12)

1.3 Reflectance Measurement

25

Fig. 1.18 Plot of experimental results in a K − λ0 coordinate system. (Reproduced with the written permission from Springer Nature)

which is plotted also in Fig. 1.17. The course of the curve in the central part is just the same as for K 1 . Some deviations can be seen first of all in the upper part of the curve compared to K 1 . . The comparison of this theoretical relationship with experimentally obtained values will be given in the following. (see also Fig. 1.18) In the practice a spectrophotometer measures and supplies the reflection ordinates for every 10 nm in the visible region. The range from 400 to 700 nm contains a total of 31 data. Numerical integrations between the appropriate upper and lower limits give the average partial reflections Rm1 , Rm2 , Rm3 and Rm for the entire range. A simple numerical procedure uses the subsequent ordinates, for example Rm1 =

R400 + R410 + · · · + R500 11

A slightly better accuracy can be achieved by using the average values of two neighbouring ordinates, for example: Rm1 =

(R400 +R410 ) 2

+ ··· + 10

(R490 +R500 ) 2

In order to determine the characteristic numbers K 1 or K 2 based on experimental results, the following equations can be used: K1 =

Rm2 − Rm1 R550 − R450 and K 2 = R650 − R550 Rm3 − Rm2

26

1 Measurement Methods and Characterisation of the Optical Parameters …

The experimental value of λ0 will be given by searching the intersection point between the horizontal line of Rm and the spectral curve (see Fig. 1.15). Knowing the experimentally obtained K 1 and λ0 or K 2 and λ0 coherent values, they will be plotted in a coordinate system similar to Fig. 1.17. Through the comparison of the agreement between the theoretically derived curve and the experimentally obtained data points, the usability of this method can be assessed for a quick colour estimation. In the case of a good agreement namely, the K 1 or K 2 value alone can determine the expected value of λ0 together with the colour hue using Fig. 1.18. The results based on the measured spectral reflectance curves for a lot of different wood species are presented on Fig. 1.18. The measured points follow the theoretical curve with acceptable accuracy, the relative standard deviation is low enough to determine the colour hue with absolute certainty and accuracy. The horizontal axis below indicates also the range of colour hues for quick estimation. The resolution of colour coordinates enables to differentiate within a colour range. On the upper side of the diagram the corresponding standardized colour scale (h*) is also given for comparison and control. The h* is defined here as a quotient of b* and a* colour coordinates (without arc tan). The reflectance spectra of wood species (Fig. 1.14) do not show separate and characteristic absorption bands in visible domain. The spectrum of most wood species presents intensive absorption only in the blue region and this tendency seems to be the same in the UV region as well. Therefore, the united absorbance spectrum of different wood species in the UV and visible region is presented in Fig. 1.19. Black locust and Scots pine were chosen because of their different types of extractives and poplar because of its low extractive content. All spectra have their maximum below 300 nm. This is the absorption band of lignin. The benzene ring has three absorption bands in the UV region. The primary bands are located at 184 and 202 nm, and one secondary band can be found at 255 nm. The additional groups of phenylpropane units shift the absorption locations of benzene towards longer wavelengths. The secondary peak is located for isolated syringyl and guaiacyl lignin at 273 and 281 nm, respectively. Peak position of lignin in Fig. 1.19 is at 282 nm for black locust and poplar. and 290 nm for Scots pine. The shifting of peak position comparing to the isolated lignin was generated by the perturbing effect of the surrounding chemical compounds. The spectrum of Scots pine has an additional maximum at 334 nm and black locust shows two shoulders at 326 and 400 nm. These absorption bands are generated by different extractives in black locust and Scots pine. Poplar does not present these absorption bands because of its low extractive content. There are many absorption bands close to each other in the visible region. This abundance of bands generates a peak-less absorption curve. Absorption bands of chromophore molecules can be found here comprising conjugated double chemical systems. These systems are located mainly in extractives and in lignin. The aromatic ketones absorb at around 330 nm. Hydroperoxides and carboxylic groups absorb at about 210 and 235 nm (Hon 1979). Visible light is mainly absorbed by lignin below 500 nm and by phenolic extractives, such as tannins, flavonoids, stilbenes and quinones above 500 nm (Hon 2001). In the red region, absorption of the fourth overtone for C-H stretching vibration can be found as well.

1.3 Reflectance Measurement

27

Fig. 1.19 Absorbance spectrum of different wood species in the UV and visible radiation range

Basically, those interactions should be taken into account where the absorbed photon splits a chemical bond. Moreover, if a photon can split a chemical bond, all other photons with shorter wavelength have enough energy to split the same bond. Thus, there are well defined upper photon wavelength limits to split chemical linkages. These limits are for O–O, C–C, C-O, C-H and O–H bonds 440, 360, 343, 295 and 255 nm, respectively.

1.3.3 Near Infrared (NIR) Spectrum of Wood The NIR wavelength interval is located between 750 and 2500 nm (13,333 and 4000 cm−1 if expressed in wavenumber). Similarly to the neighbouring visible region, shorter wavelength (750–1000 nm) part of the NIR region does not show individual absorption bands. Scientist usually present the spectrum above 1000 nm (10,000 cm−1 ). The NIR spectrum of wood contains overtone and combination bands of the main chemical components. Overtones can be thought of as harmonics. So, every fundamental vibration will produce a series of absorptions at multiples of the basic frequency. The C-H associated vibrational information is repeated three times in the NIR region. The fundamental vibration of the C-H group at 3400 nm (2941 cm−1 ) is accompanied with overtones at 1700, 1130 and 850 nm (5882, 8850 and 11,765 cm−1 ). (The wavelength of overtones is about one-half, one-third and onequarter of the wavelength of fundamental vibration.) Similar list of overtones can be calculated for O–H stretching vibrations as well. Although, the intensity of overtones is decreasing quickly with their increasing sequential number. The presence of several overtones helps the chemical analysis.

28

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.20 Absorbance spectrum of black locust, poplar and Scots pine specimens in the NIR region

The NIR spectrum is dominated by hydrogen. The overtone and combination bands of C–H, O–H and C=O stretching vibrations generates the visible absorption peaks of wood in the NIR region (Fig. 1.20). Absorption intensities of these overtones are much smaller than the intensities of the fundamental bands in the analytical IR region. Fortunately, modern and sensitive spectrophotometers can detect these subtle intensities as well. Photons in the NIR domain have more energy than those belonging to the analytical region. This energy gives the possibility to lift up a molecule more than one energy level. Similar transitions are forbidden by the selection rules of quantum mechanics. The consequence is that the absorptivity in the NIR region is typically small. The NIR absorption bands are very broad. It is difficult to assign particular features to specific chemical components. Multivariate (using multiple variables) calibration techniques (e.g., principal components analysis, or partial least squares method PLS) are often employed to specify the desired chemical information. Absorptions of combination bands are rather more complex. NIR absorptions are at a higher state of excitement. They require more energy than a fundamental absorption. Combinations arise from the sharing of NIR energy between two or more fundamental absorptions. The number of possible overtones belonging to a fundamental absorption in a molecule is limited. In contrast, very large number of combinations can be observed. The effect of all these absorptions makes the NIR spectra to consist of only a few rather broad peaks. It is important to realise that all of these broad absorption bands are generated by multiple overlapped absorptions. NIR spectra are much more complex than they appear. The most common combination bands result from stretch and bend combinations in the same group. Absorption can be seen due to the combination of O–H stretch with O–H bend and C–H stretch with C–H bend and these occur in different positions in the spectrum. It is difficult to assign single bands in NIR spectrum because of the high number of bands and the great overlapping. The second derivative spectra and multivariate data

1.3 Reflectance Measurement

29

analysis can help to differentiate the bands. Schwanninger et al. (2011) published a band assignment review containing 136 references. Table 1.1 contains the main absorption bands represented by Fig. 1.20. The advantage of NIR absorption measurement is that NIR light can penetrate deeper into a sample than analytical infrared radiation. Because of the low absorption coefficient of NIR light, relatively thick specimens with high moisture content can be measured. NIR spectroscopy is important in quality control and quality assurance processes where the key goal is to accurately determine the chemical and physical properties of the investigated material. NIR spectroscopy has high priority as online measurement technique during different process controls. Tsuchikawa and his coworkers collected and reviewed more than 350 papers dealing with NIR research and applications (Tsuchikawa 2007; Tsuchikawa and Schwanninger 2013; Tsuchikawa and Kobori 2015). Extremely large number of scientific topics and applications can be found in the above cited review papers. Here only two main topics are discussed. Water is a strong NIR radiation absorber, thus NIR measurements play an important role in moisture quantification applications for wood. Many researchers Table 1.1 Assignment of absorption bands for NIR spectrum of wood (Mitsui et al. 2008; Mehrotra et al. 2010) Wavelength (nm)

Wavenumber (cm−1 )

Chemical group

Location

1195

8368

CH stretching second overtone

Lignin

1350

7407

CH stretching and CH deform

Cellulose

1428

7003

OH stretching first overtone

Cellulose, amorphous region

1460

6849

OH stretching first overtone

Water

1487

6722

OH stretching first overtone

Cellulose, semi-crystalline region

1547

6460

OH stretching first overtone

Cellulose, crystalline region

1672

5981

CH stretching first overtone

Lignin aromatic

1724

5800

CH stretching first overtone

Hemicellulose furanose/ pyranose ring

1930

5181

OH stretching and OH deform

Water

2066

4840

OH stretching and CH deform

Cellulose

2258

4429

CH stretching and CH deform

CH3 / cellulose

2329

4294

CH stretching and CH deform

Hemicellulose

30

1 Measurement Methods and Characterisation of the Optical Parameters …

presented the usefulness of NIR spectroscopy in moisture content determination both with scientific and practical purpose (Tsuchikawa et al. 1996; Tsuchikawa and Tsutsumi 1998; Thygesen and Lundqvist 2000; Defo et al. 2007; Mora et al. 2011; Kobori et al 2013). Water has two strong typical absorption bands with peaks around 1930 nm (5181 cm−1 ) and 1460 nm (6849 cm−1 ). The first one is a combination band while the second one is the first overtone of water absorption fundamental in the middle IR region. Figure 1.21 presents the NIR absorbance spectrum of poplar samples with different moisture content. The spectra were two-point baseline corrected at 1292 and 1860 nm (7740 and 5376 cm−1 ). Both water absorption band showed significant increase proportionally to moisture content increase. It is well visible that the abscissa of the maxima shifted during moisture content increase meaning that these bands are not single absorption bands. They are superpositions of minimum two bands. This phenomenon is discussed in Sect. 1.3.4. (Fig. 1.27). The maximum places of spectra are 1484, 1464, 1456 and 1456 nm (6739, 6831, 6868 and 6868 cm−1 ) at 9.43, 14.56, 23.69 and 30.94% wood moisture content, respectively. Similar peak positions for the other maximum are 1924, 1926, 1930 and 1930 nm (5198, 5192, 5181 and 5181 cm−1 ). The data represent that there is an intensive band on the right side of the water band at 1456 nm which does not change its intensity during wetting. This is the absorption band of semi-crystalline region in cellulose. The cellulose band was dominant at low moisture content. That is why the peak of the integrated band shifted considerably. The place of maximum for the other integrated band at 1930 nm hardly shifted representing that the sub-band on the left side is a small one. This sub band belongs to the second overtone of C=O stretching vibration of hemicellulose absorbing at 1910 nm (5236 cm−1 ). Figure 1.22 visualise the good correlation between wood moisture content and absorption intensities of the water bands.

Fig. 1.21 Relative absorbance spectra of poplar generated by different moisture contents

1.3 Reflectance Measurement

31

Fig. 1.22 Correlation between moisture content and NIR absorbance values of poplar at 1456 and 1930 nm

Straight trend lines prove that there is linear correlation between the absorbance values at the chosen wavelengths and the moisture content of the investigated sample. High values of the coefficient of determination confirms the strong relation. The change of wood moisture content during steaming can be monitored by NIR spectrum measurement. NIR spectrum analysis demonstrated that the equilibrium moisture content of wood decreased during steaming below 110 °C. This finding was first time published in scientific journal (Mahdiyanti et al. 2020). Measurement of the moisture content of wood under industrial conditions using NIR technique requires certain data manipulations. The measured reflectance values should directly be used (without any baseline correction and K-M transformation) to calculate the moisture content. Swelling or shrinkage modify the wood surface resulting in changes of reflectance properties. This phenomenon generates parallel and multiplicative shift of the spectrum. To eliminate these distortions, an internal reference point should be chosen where the absorption does surely not change during the applied treatment. In our case, this internal reference point can be at 1680 nm (Tsuchikawa et al. 1991). Calculating the absorption difference between the water peak and the internal reference point the effect of treatment appears. Mechanical parameters of wood are highly important for wooden constructions. The mechanical properties are partly determined by the ratios of the crystalline, semicrystalline and amorphous regions of cellulose (Fujimoto et al 2007). The crystalline region generates higher density and lower chemical reactivity compared the less arranged regions. The broad band between 1390 and 1640 nm (7200 and 6100 cm−1 ) belongs mainly to the first overtone of O–H stretching in cellulose. Tsuchikawa and Siesler determined the absorption band locations of different type of cellulose structures by deuterium exchange method and NIR polarisation spectroscopy (Tsuchikawa and Siesler 2003). The places of absorption maxima were found for amorphous, semi

32

1 Measurement Methods and Characterisation of the Optical Parameters …

-crystalline and two types of crystalline cellulose at 1428, 1488, 1548 and 1592 nm (7003, 6722, 6460 and 6281 cm−1 ), respectively. Thermal treatment of wood can modify and redistribute the structure of cellulose resulting in the change of the ratios of amorphous, semi -crystalline and crystalline structures (Mitsui et al. 2008; Inagaki et al. 2010). For demonstrating this phenomenon, black locust was steamed at 110 °C under saturated steam condition for two days. The NIR absorbance spectrum was measured before treatment and after treatment followed by slow drying. Then the spectra were two-point baseline corrected at 1390 and 1640 nm (7200 and 6100 cm−1 ). This broad O–H band is a sum of widely overlapping bands. The comparison of spectra generated before and after steaming does not show clearly the changes. The difference spectrum (treated-initial) shows only those bands that were modified by the applied treatment. This steam treatment generated difference spectrum of black locust can be studied in Fig. 1.23. The spectrum presents that the absorption intensity of amorphous cellulose structures decreased while that of crystalline cellulose increased due to steaming. Partly increased the absorption intensity of semi-crystalline cellulose as well but it is not clearly visible. The negative band of amorphous cellulose and the positive band of semi-crystalline cellulose overlap each other at around 1480 nm and the opposite values quench each other. The values of changes are small because of the relatively low steaming temperature and the pure absorption values in NIR region. None the less, the NIR measurement presents clearly the changes.

Fig. 1.23 Difference NIR spectrum of black locust generated by two-day steaming at 110 °C

1.3 Reflectance Measurement

33

1.3.4 Analytical (Middle) Infrared (IR) Spectrum of Wood The plot of absorption intensities versus wavelength or wavenumber in the 2500– 25,000 nm or 400–4000 cm−1 interval is referred to as the analytical infrared spectrum of the examined chemical compound. The absorption of IR radiation can modify the stretching and bending vibration amplitude of the chemical bonds in most covalent molecules. Only selected frequencies of IR radiation will be absorbed by a molecule. Each type of the bonds has a different natural frequency of vibration. The same bond type in different environment can have slightly different natural frequency. This is because of the perturbing effect of the neighbouring atoms. The importance of the IR spectrum lies in that it provides structural information about the investigated molecule. The bond strength and the mass of the atoms bonded together affects the IR absorption frequencies. Stronger bonds will vibrate at higher frequencies than weaker bonds and bonds between atoms of higher masses will vibrate at lower frequencies than bonds between lighter atoms. The IR absorbance spectrum of wood can be determined by DR method using K-M transformation or Log 1/R function. The ATR measurement method can be applied for chip type samples. In case of very thin samples, the transmittance measurement method can be used. In this study, the DR measurement method will mainly be used combined with K-M transformation. The wavenumber will be utilized as independent variable to present results that have already been published. In other cases, however, the combination of wavelength and wavenumber will be applied, to show the usefulness of wavelength as independent variable. The IR spectrum is commonly used to monitor the chemical changes in wood induced by different treatments. The K-M transformation of the reflectance spectrum calculates the quotient of the absorption and the scattering coefficients. The absorption coefficient is strongly while the scattering coefficient is only slightly wavelength dependent. The wavelength dependence of scattering coefficient is close to linear in such a short wavelength interval (2500–10,000 nm) used in wood science. Consequently, the K-M transformation provides the absorbance spectrum. Figure 1.24 shows the reflectance spectrum of black locust presented in the 2500– 10,000 nm wavelength interval (4000–1000 cm−1 wavenumber). The reflectance value at 2630 nm (3800 cm−1 ) is close to 100% and it should be 100% at 5260 nm (1900 cm−1 ) as well, because there is no absorption for wood at these wavelengths. The baseline shift causes that the reflectance values are not 100% at those wavelengths where there are no absorptions. This anomaly can be eliminated by baseline correction. There are two-point and multi-point baseline corrections. The two-point baseline correction is usually enough to amend the IR spectrum of wood. It is recommended to perform the baseline correction of the reflectance spectrum before K-M transformation. Computer supported IR spectrophotometers can complete the necessary baseline correction automatically if the proper points are chosen. It is important to know that selecting a wavelength for baseline correction where there is actual absorption, the spectrum may suffer dangerous alteration, creating unwanted artificial changes. The reflectance spectrum of black locust was two-point baseline corrected

34

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.24 Reflectance spectrum of black locust sample and its baseline corrected (BC) form

(solid line in Fig. 1.24) at 2630 and 5260 nm wavelengths (3800 and 1900 cm−1 wavenumbers). The baseline correction eliminates the wavelength dependent effect of scattering. It is well visible in Fig. 1.24 that linear baseline correction is enough for wood samples. Now, the baseline corrected reflectance spectrum is suitable to perform K-M transformation. Figure 1.25 shows the IR spectrum of beech and larch generated by K-M transformation. The absorbance spectra of this hardwood and softwood species are presented together for comparison. The IR absorbance spectrum of wood contains highly overlapped absorbance bands of cellulose, hemicelluloses and lignin. Left part of the spectrum consists of two broad bands. The higher peak belongs to the absorption of hydroxyl groups located at various places in cellulose, hemicelluloses and lignin. These diverse OH groups have absorption at different wavelengths causing a rather wide absorption band. Water in the wood also results in absorption in this region. The second band represents the absorption of methyl groups. The so called “fingerprint” region (right side) is an about 5500 nm (900 cm−1 ) wide interval and this interval slightly depends on the complexity of the material. In case of wood, the fingerprint region is located between 5500 and 11,000 nm (1800 and 900 cm−1 ) and is characteristic for every wood species. The place of maxima can be different for the individual species. Internal intensity differences also characterise the species. It is well visible if the spectra of beech and larch are compared. The absorbance band of unconjugated carbonyl groups at 5747 nm (1740 cm−1 ) shows great difference in intensity. Most carbonyl groups in wood are in the hemicellulose component (Owen and Thomas 1989). The reason of the absorbance difference lies partly in the diverse hemicellulose content of beech and spruce. The highly acetylated xylan is the main polyose component found in hardwoods, while glucose units in spruce are rarely acetylated. The absorption of carbonyl groups in the acetyl unit may generate the absorbance

1.3 Reflectance Measurement

35

Fig. 1.25 Absorbance spectrum of beech and larch samples

difference between beech and larch at 5747 nm (1740 cm−1 ). The next dissimilarity is induced by the different lignin content of hardwoods and softwoods. Hardwoods contain both syringyl and guaiacyl lignin absorbing at 6262 nm (1597 cm−1 ) and 6622 nm (1510 cm−1 ), respectively. The intensities of these bands are almost equal for beech, but they are highly different for larch. Lignin in larch is mainly guaiacyl type which is well represented by the intensity differences at 6262 and 6622 nm. The absorption intensity at 6622 nm is much greater than at 6262 nm. Larch sample has an additional peak at 6093 nm (1641 cm−1 ). This is the absorption region of conjugated carbonyl groups, generating the colour of wood. The colour of larch is redder compared to beech. The absorption bands overlap each other in the fingerprint region. The visible peak positions are some cases not the real ones because of the superpositions of two or more bands. This phenomenon can be visualised by artificially overlapped bands (Figs. 1.26 and 1.27). Two twin bands (with equal height) are visible in three different positions in Fig. 1.26. The solid line represents the sum of the two twin bands. The position of the left band was fixed and the band on the right side was gradually moved towards the left band. The distance between the maxima of the twin bands were 45, 28 and 20 units from left to right, respectively. It is especially important to know that the spectrum recorded by the spectrophotometer presents always the sum of individual bands. The integrated spectrum in the first (left) diagram shows clearly that the spectrum consists of two individual bands. The peak positions seem to be the same for the integrated band and for the individual twin bands. However, the intensity values show that the positions of maxima for the integrated spectrum moved towards each other slightly (one unit). The diagram in the middle presents the same twin bands, however, the distance between the bands is shorter (28 units) than for the left-side diagram. The integrated spectrum has a plateau in the middle with two

36

1 Measurement Methods and Characterisation of the Optical Parameters …

maxima closer to each other (15 units) than the initial maxima of the twin bands. The diagram on the right proves that the superposition can generate a single integrated band if the two bands are even closer to each other. This diagram demonstrates that the superposition of two (or more) bands may generate a new artificial apparent band. Looking at the measured absorbance spectrum of wood (Fig. 1.25, fingerprint region), it is not possible to find out whether the visible bands are the real ones or some of them are the sum of individual bands. The other combination of individual bands might be if the intensities are different whilst the distance remains the same demonstrated in Fig. 1.27. The left diagram presents that two bands close to each other may generate one new band. The maximum of this apparent band is in the middle between the twin bands. The diagram on the right side shows the superposition of bands with different height. The place of maximum for the integrated band shifted towards the dominant band. This phenomenon can often be observed when investigating changes in the absorbance spectrum due to different treatments. The shift of maximum place during a treatment indicates that the band is multiple, and one of the sub bands changed differently than the other(s). This happens during photodegradation for the unconjugated carbonyl band. The difference spectrum technique is an additional method to determine the exact places of changes in the absorption bands. After creating the difference spectrum

Fig. 1.26 Superpositions of two absorption bands in different positions

Fig. 1.27 Superpositions of two absorption bands with different height

1.3 Reflectance Measurement

37

(irradiated minus initial) only those absorption bands appear where changes occurred. The increase in absorption is represented by positive band while negative band represents the absorption decrease. The differences are usually small compared to the intensities of the original spectrum. That is why disturbing effects must be minimalized. These effects for wood are inhomogeneity and anisotropy of the sample and the roughness change during treatment. The same area of the sample must be measured before and after the treatment. Grain direction, angle between the fibre direction and the incident IR light must be identical for all measurement (Zavarin et al. 1990; Tolvaj and Mitsui 2004). The carbonyl band between 5460 and 5960 nm (1680 and 1830 cm−1 ) increased during 24-h UV irradiation (Fig. 1.28). Looking at the absorption change of the unconjugated carbonyl band, the shift of the band towards lower wavelengths is visible beside the increase of absorbance values. The maximum moves from 5747 nm (1740 cm−1 ) to 5720 nm (1748 cm−1 ). Moreover, the greatest difference between treated and initial spectra appears not at the maximum but at the left side of the band. The difference at the maximum is 0.14 while at 5672 nm (1763 cm−1 ) 0.37 units, which is more than double. These findings highlight the disadvantage and deceptiveness of the simple comparison method where the initial and the treated spectra are presented on top of each other. The difference spectrum method helps to make correct and more detailed interpretation. It is visible that two bands were grown up during photodegradation with maxima at 5666 nm (1765 cm−1 ) and 5843 nm (1711 cm−1 ). The simple comparison of two absorbance spectra (solid and dotted lines) does not provide as detailed information as the difference spectrum.

Fig. 1.28 Absorbance spectrum of beech before treatment (0 h) and after 24-h UV treatment (24 h) and the difference spectrum (Diff 24)

38

1 Measurement Methods and Characterisation of the Optical Parameters …

The maximum place of the band of unconjugated carbonyl groups around 5747 nm (1740 cm−1 ) is shifted towards shorter wavelengths demonstrating that the leftside sub band was growing faster than the one on the right side. It is a known phenomenon since 1995 that two bands are growing up in the unconjugated region during photodegradation (Tolvaj and Faix 1995). None the less, some papers evaluate the results incorrectly as it is a single band for unconjugated carbonyl groups. Figure 1.25 presents the absorbance spectra of a heartwood (beech) and a softwood (larch) species. Chemists created a list of absorption peaks of IR spectrum for identifying the absorbing chemical groups of the measured material. This list is presented in Table 1.2 with the assignment of the visible bands indicating the places of maxima for the peaks for both beech and larch (Fig. 1.25) in nm and cm−1 as well. Often happens that samples made of the same wood block gives visually different absorbance spectra. This phenomenon is visible in Fig. 1.29 introducing the absorbance spectra of 3 beech specimens. Although, band peaks are located at the same places, the intensities are different for the 3 individual samples. The intensity differences are proportional. The measured reflectance values were different for the investigated samples and the K-M transformation generated different absorbance values. Deviation among the reflectance values was not generated by real absorbance differences, but by the different diffuse reflectance conditions. Multiplicative type Table 1.2 IR absorbance bands of beech and larch wood species (place of maximum) and band assignments (Tolvaj and Faix 1995; Huang et al. 2012; Csanady et al. 2015) Larch

Beech

Assignment

(cm−1 )

(nm)

(cm−1 )

(nm)

3453

2896

3457

2893

O–H stretching

2939

3403

2933

3409

CH stretching

1740

5747

1738

5752

C=O stretching vibration of non-conjugated carbonyls

1652

6054

1641

6093

Conjugated C–O in quinines coupled with C=O stretching of various groups (flavones)

1597

6262

-

-

Aromatic skeletal breathing with CO stretching (syringyl lignin)

1504

6648

1511

6614

Aromatic skeletal vibrations (guaiacyl lignin)

1463

6836

1456

6864

C–H deformation

1428

7000

1428

7000

Aromatic skeletal vibration combined with C–H in-plane deformation

1379

7252

1376

7267

Aliphatic C-H stretching in CH3

1335

7493

1319

7581

Aromatic ring breathing (mainly syringyl)

1267

7892

1278

7821

Aromatic ring breathing (mainly guaiacyl)

1175

8514

1175

8508

C–O–C stretching (asymm.) in cellulose and hemicelluloses

1135

8811

1129

8856

C–O–C stretching (symm.), arom. C–H i.p. deformation, glucose ring vibration

1091

9170

1097

9113

C–O–C stretching

1.3 Reflectance Measurement

39

Fig. 1.29 Absorbance spectra of 3 beech specimens

of anomaly is a common phenomenon in IR spectroscopy, especially in diffuse reflectance spectroscopy. Absorbance values are usually calculated by the KubelkaMunk (K-M) equation where there is a hyperbolic (reciprocal) relationship between the reflectance and absorbance (K-M units) values. Diffuse reflectance units built in the IR spectrophotometers are unable to collect all reflected photons. Furthermore, reflected photons are collected, happens within a well-defined square angle determined by the size of the spherical mirror (Fig. 1.1). Any changes in the surface properties modify the number of the mirror-collected photons modulating the measured reflectance values. These changes appear in the absorbance spectrum, even though that the absorption itself did not change obviously. Roughness change is one of the usual surface property alterations during different treatments of wood. The increase in surface roughness causes more diffuse reflection, thus reducing the number of the mirror-collected photons. Ultimately, the measured reflectance intensity will therefore be lower due to the roughness increase. The K-M equation interprets this as an absorption increase. This fact increases the absorbance values generating distortion. This phenomenon was discussed in a previous paper (Tolvaj et al. 2011). The multiplicative type of anomaly can be recognised as diverse distortions within the spectrum. Small absorbance values suffer small distortions while high absorbance values undergo great distortions. The IR spectrometer measures approximately 15 mm2 (small) area of the sample. Roughness differences among the measured sample surfaces are responsible for the deviations among the reflectance values and among the absorbance values. These types of distortions can be eliminated by normalisation process. For a proper normalisation an absorption peak should be found that is not affected by the applied treatment. If there is more

40

1 Measurement Methods and Characterisation of the Optical Parameters …

than one choice it is better to choose a peak in central position. During the normalisation, the spectra are multiplied by different constants obtaining equal value for all spectra at the chosen peak. The other possibility is to multiply all of the spectra getting one unit value at the chosen peak. The second option is the better choice if the spectra of different samples or species are intended to be compared. The sample preparation should not generate detectable chemical changes. Figure 1.29 shows the absorbance spectra of 3 beech specimens. There are great intensity differences among the species. These differences originated from the diverse surface properties of the species. It is visible that there are multiplicative differences between the K-M values at all wavenumbers. The normalisation can minimalize the deviations. In case of Fig. 1.29 any of the peaks can be chosen for normalisation. Here, the C=O band of unconjugated carbonyls absorbing at 5723 nm (1747 cm−1 ) were chosen as the place of normalisation because its central position. The result of normalisation is shown in Fig. 1.30. Distortions were eliminated by the normalisation in most wavelength ranges only a little anomaly remained above 8500 nm. These results show that normalisation is able to eliminate the reflectance differences originated from the surface property differences among the specimens. According to these results, four to ten samples are enough to determine the absorption properties of a species. In some cases, the treatment generates surface roughness change. Photodegradation is a good example to such alteration. The phenomenon is discussed here according to a recent paper (Tolvaj 2022). Figure 1.31 represents the absorbance spectra of poplar wood measured before and after 5-day UV irradiation. Deviations between the plotted spectra are generated by both the photodegradation and the

Fig. 1.30 Normalised absorbance spectra of 3 beech specimens. Spectra were normalized at the peak of 5723 nm

1.3 Reflectance Measurement

41

Fig. 1.31 Absorbance spectra of poplar measured before and after 5-day UV irradiation

distortion caused by the roughness increase. The photodegradation of the aromatic ring in lignin reduced the absorption at 6644 and 6266 nm (1505 and 1596 cm−1 ) wavelength. Consequently, the intensity of unconjugated carbonyl groups increased at 5746 nm (1740 cm−1 ). The spectrum of treated sample is lifted in the 6897– 5263 nm (1450–900 cm−1 ) region (lifting happened in the carbonyl and aromatic regions as well, however, it is not evident because the photodegradation also caused absorption changes at the same places). This anomaly confuses the interpretation of the photodegradation generated changes in the 6900–10,000 nm (1450–1000 cm−1 ) region. Normalisation can help to reduce this anomaly, however. For correct normalisation it is necessary to find an absorption peak which is not affected by the applied treatment. The normalisation eliminates the anomaly at the place of the chosen peak and reduces it considerably in other places. The spectra of Fig. 1.31 were normalised to the band maximum at 7267 nm (1376 cm−1 ) thus the intensities were adjusted to 1.0. This C-H band of cellulose is often used as internal standard because of its high intensity, central position and strong stability. The normalised spectra are shown in Fig. 1.32. This data manipulation eliminated the lifting effect of the roughness increase. The effectiveness is only questionable around 9000 nm (1110 cm−1 ). The remained lifting around 9000 nm (1100 cm−1 ) demonstrates that the normalisation is not a perfect solution it can help only close to the chosen peak. That is why the central position of the chosen peak is important. The usefulness of normalisation is well demonstrated by the difference spectra (Fig. 1.33). Both difference spectra were calculated by subtracting of the spectrum of the initial sample form the treated one. In this case, an absorption increase is represented by a positive band while a negative band represents an absorption decrease. Figure 1.33 shows clearly the multiplication anomaly and how it was reduced by

42

1 Measurement Methods and Characterisation of the Optical Parameters …

Wavenumber (cm-1) 1.4

1667

1429

1250

1111

Relative unit

1.2 ←7267 (1376)

1 0.8 0.6 0.4

Norm 0UV Norm5UV

0.2 0 5000

6000

7000

8000

9000

10000

Wavelength (nm) Fig. 1.32 Normalised absorbance spectra of poplar recorded before and after 5-day UV irradiation

normalization. Photodegradation generates absorption decrease of the aromatic CH deformation at 6998 nm (1429 cm−1 ). This negative difference peak is a valley in the positive surrounding, if the spectrum is not normalised (dotted line). This phenomenon confirms the multiplicative effect. The calculation of the difference spectrum is a useful method if the spectrum contains highly overlapped bands such as that of the wood in the fingerprint region. The main advantage of the difference spectrum method is that only the altered bands Wavenumber (cm-1) 1

1667

1429

1250

1111 Norm5UV 5UV

0.8

Relative unit

0.6 0.4 0.2 0 -0.2 -0.4 5000

6000

7000

8000

9000

10000

Wavelength (nm) Fig. 1.33 Difference absorbance spectra of poplar calculated without (5UV) and with (Norm5UV) normalization. All changes in the spectra were generated by 5-day UV irradiation

1.3 Reflectance Measurement

43

are visible, reducing the number of overlapped bands. In order to get a correct difference spectrum, it is important that the measurements must be taken on the exact same surface area before and after the treatments. If this important requirement is ignored, the inhomogeneity of wood surface may destroy the difference spectrum and lead to drawing wrong conclusions. The usefulness of the difference spectrum method is demonstrated by some papers dealing with the temperature dependence of wood photodegradation (Tolvaj et al. 2013; Varga et al. 2017; Preklet et al. 2018) and the leaching effect of water during wood photodegradation (Bejo et al. 2019; Pasztory et al. 2020; Varga et al. 2020). Figure 1.33 contains two types of difference spectra but both generated by 5day UV irradiation of wood samples. One was calculated without normalizing the spectrum of both the untreated and the irradiated sample while the other was determined after normalisation. The interpretation of the difference spectrum is difficult without normalisation in the 6897–11,111 nm (1450–900 cm−1 ) wavelength interval. Negative peaks appear as valleys between two positive peaks. Normalisation reduces this anomaly, however, does not eliminate it completely. The reliability of the data remained questionable around 9091 nm (1100 cm−1 ) and further on. Although, normalisation seems to be a useful method to evaluate a spectrum, it can cause abnormal changes if it is applied incorrectly. Another important requirement beside the ones mentioned above is that the chosen peak must remain intact by the applied treatment. The highest peak is often used as a norming position. It can be correct or incorrect as well, depending on the stability of the chosen highest peak during the applied treatment. Unfortunately, some researchers apply normalisation knowing that the chosen peak alters during the applied treatment. In this regard, an example can be seen to present the consequence of using the wrong normalisation. Figure 1.34 presents the K-M spectrum of poplar before and after UV irradiation. The comparison of spectra shows the degradation of lignin absorbing at 1505 and 1596 cm−1 wavenumbers (6644 and 6266 nm). As a consequence, the intensity of unconjugated carbonyl groups increased at 1740 cm−1 . The right part of the spectra for irradiated samples presents a parallel shift as a result of scattering change. The absorption changes are located on the top of the parallel shifts. The alteration of scattering was generated by the changing roughness. Also, it is difficult to measure exactly the same area of the specimen before and after treatment and the inhomogeneity of the surface also causes a scattering change. Proper normalisation can minimize the effect of scattering (see Fig. 1.32). Figure 1.35 presents the effect of incorrect normalisation. It was mentioned above that the spectrum intensity at a proper normalisation location does not change during the treatment. Here, the result of normalisation at 5740 nm (1742 cm−1 ) is presented where the absorption intensity increased during photodegradation. The normalisation of the spectra made the values equal at 5740 nm but the other regions became confused. This process produced great distortion of the spectra. The normalisation spread out the spectra proportionally to the intensity differences at 5740 nm. This example clearly shows the possible results of a wrong spectra normalisation. Wood has no absorption at 1900 and 3800 cm−1 wavenumbers (5260 and 2630 nm). This means that the reflection should be 100% at these places, however,

44

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.34 The K-M spectrum of poplar wood before and after UV irradiation

Fig. 1.35 Normalised K-M spectra of poplar wood before and after UV irradiation

the reflection is less than 100% because of the light scattering; this is called the baseline shift in IR spectroscopy (Fig. 1.36). Before calculating the absorption spectrum, these reflection values have to be lifted up to 100% (lifting up the whole spectrum) to eliminate the scattering effect. This transformation is the so-called baseline correction, which is regularly used in diffuse reflectance IR spectroscopy (see Fig. 1.24). The scattering depends on the roughness of the surface and therefore the baseline shift gives information about the surface roughness. Here, the baseline shift is given as the quotient of two reflection intensities at 3800 and 1900 cm−1 .

1.3 Reflectance Measurement

45

Fig. 1.36 Diffuse reflectance IR spectrum of spruce wood indicating the baseline shift

Baseline shift = R(3800)/R(1900) where R(3800) is the reflection intensity at 3800 cm−1 and R(1900) is the same at 1900 cm−1 . The IR reflection spectra of several wood species were recorded and the surface roughness was determined by perthometer. The roughness alteration was generated by photodegradation. The roughness was represented by the Rz parameter (average peak-to-valley height). After calculating the baseline shift parameter, the correlation between roughness and baseline shift was evaluated. The results are presented in Fig. 1.37 for black locust and Scots pine wood species. The trend lines are linear presenting linear correlation between baseline shift and roughness. The linear trend lines are parallel and are located close to each other. The high values of the coefficients of determination (R2 ) prove good correlation between roughness and baseline shift in all cases. The results offer a rapid and easy roughness determination by measuring the IR spectra (Tolvaj et al. 2014). This close correlation also suggests that it is possible to monitor the roughness alteration effect of photodegradation by measuring the IR diffuse reflectance spectra and calculating the baseline shift. It is highly important to note that exactly the same area of the sample must be used for all reflectance measurements, otherwise, the surface inhomogeneity overlaps the roughness change. The diffuse reflectance measurement method is used consequently in this study. The IR reflectance spectra were measured by a JASCO FTIR spectrophotometer. The background spectrum was determined by using an aluminium plate having rough surface. The resolution was 4 cm−1 and the average of 64 scans gave the IR spectrum. Two-point baseline correction was applied at 1900 and 3800 cm−1 wavenumbers (5263 and 2632 nm). The K-M equation was used to determine the absorbance spectrum. The intensity of spectra was normalised to the band maximum at around 1375 cm−1 (7273 nm) in case of photodegradation. The intensity of spectra was

46

1 Measurement Methods and Characterisation of the Optical Parameters …

Fig. 1.37 Correlation between baseline shift and surface roughness of black locust and Scots pine samples. The roughness alteration was generated by photodegradation

adjusted to 1.0 by this normalisation at maximum around 1375 cm−1 . This C-H band of cellulose is often used as internal standard because of its high intensity, central position and strong stability.

References Anderson, E.L., Pawlak, Z., Owen, N.L., Feist, W.C.: Infrared studies of wood weathering. Part i: Softwoods. Appl. Spectrosc. 45(4), 641–647 (1991a). https://doi.org/10.1366/000370291433 6930 Anderson, E.L., Pawlak, Z., Owen, N.L., Feist, W.C.: Infrared studies of wood weathering. Part II: Hardwoods. Appl. Spectrosc. 45(4), 648–652 (1991b) Anderson, T.H., Weaver, F.W., Owen, N.L.: Anomalies in diffuse reflectance infrared spectra of wood and wood polymers. J. Mol. Struct. Struct. 249, 257–275 (1991c). https://doi.org/10.1366/ 0003702914336769 Bejo, L., Tolvaj, L., Kannar, A., Preklet, E.: Effect of water leaching on photodegraded spruce wood monitored by IR spectroscopy. J. Photochem. Photobiol. A: Chem. 382, Paper: 111948 (2019). https://doi.org/10.1016/j.jphotochem.2019.111948 Bisht, P., Pandey, K.K., Barshilia, H.C.: Photostable transparent wood composite functionalized with an UV-absorber. Polym. Degrad. Stab. 189, ID 109600 (2021). https://doi.org/10.1016/j. polymdegradstab.2021.109600 Csanady, E., Magoss, E., Tolvaj, L.: Quality of Machined Wood Surfaces. Springer, Heidlberg New York London (2015). https://doi.org/10.1007/978-3-319-22419-0 Defo, M., Taylor, A.M., Bond, B.: Determination of moisture content and density of fresh-sawn red oak lumber by near infrared spectroscopy. Forest Prod. J. 57(5), 68–72 (2007) Faix, O., Németh, K.: Monitoring of wood photodegradation by DRIFT-spectroscopy. Holz Roh Werkstoff 46, 112 (1988)

References

47

Faix, O., Böttcher, J.H.: The influence of particle size and concentration in transmission and diffuse reflectance spectroscopy of wood. Holz Roh Werkstoff 50, 221–226 (1992). https://doi.org/10. 1007/BF02650312 Fujimoto, T., Yamamoto, H., Tsuchikawa, S.: Estimation of wood stiffness and strength properties of hybrid larch by near-infrared spectroscopy. Appl. Spectrosc. 61(8), 882–888 (2007). https:// doi.org/10.1366/000370207781540150 Hon, D.N.S.: Photooxidative degradation of cellulose: reactions of the cellulosic free radicals with oxygen. J. Polym. Sci.: Polym. Chem. Ed. 17, 441–454 (1979) Hon, D.N.S.: Weathering and photochemistry of wood. In: Hon, D.N.S., Shiraishi, N. (eds.) Wood and Cellulose Chemistry. Marcel Dekker, New York (2001) Huang, A., Zhou, Q., Liu, J., Fei, B., Sun, S.: Distinction of three wood species by Fourier transform infrared spectroscopy and two-dimensional correlation IR spectroscopy. J. Mol. Sruct. 883–884, 160–166 (2008). https://doi.org/10.1016/j.molstruc.2007.11.061 Huang, X., Kocaefea, D., Kocaefea, Y., Boluk, Y., Pichette, A.: Study of the degradation behavior of heat-treated jack pine (Pinus banksiana) under artificial sunlight irradiation. Polym. Degrad. Stab. 97, 1197–1214 (2012). https://doi.org/10.1016/j.polymdegradstab.2012.03.022 Inagaki, T., Siesler, H.W., Mitsui, K., Tsuchikawa, S.: Difference of the crystal structure of cellulose in wood after hydrothermal and aging degradation: a NIR spectroscopy and XRD study. Biomacromol. 11(9), 2300–2305 (2010). https://doi.org/10.1021/bm100403y Kobori, H., Gorretta, N., Rabatel, G., Bellon-Maurel, V., Chaix, G., Roger, J.M., Tsuchikawa, S.: Applicability of VIS-NIR hyperspectral imaging for monitoring wood moisture content (MC). Holzforschung 67, 307–314 (2013). https://doi.org/10.1515/hf-2012-0054 Kubelka, P.J., Munk, F.: Ein Beitrag zur Optik der Farbanstriche. Zeitschrift für Technische Physik 11a, 593–601 (1931) Kubelka, P.J.: New contributions to the optics of intensely light-scattering materials. Part I. J. Opt. Soc. Am. 38, 448–457 (1948) Kuo, M., McClelland, J.F., Chien, P.L., Walker, R.D., Hse, C.Y.: Applications of infrared photoacoustic spectroscopy for wood samples. Wood Fiber Sci. 20(1), 132–145 (1988) Li, Y., Yang, X., Fu, Q., Rojas, R., Yan, M., Berglund, L.: Towards centimeter thick transparent wood through interface manipulation. J. Mater. Chem. A 6, 1094–1101 (2018). https://doi.org/ 10.1039/C7TA09973H Li, H., Gou, X., He, Y., Zheng, R.: A green steam-modified delignification method to prepare low-lignin delignified wood for thick, large highly transparent wood composites. J. Mater. Res. 34(6), 932–940 (2019). https://doi.org/10.1557/jmr.2018.466 Mahdiyanti, S.H., Tsuchikawa, S., Mitsui, K., Tolvaj, L.: Steaming-caused chemical changes of sugi (Cryptomeria japonica) wood monitored by NIR spectroscopy. Asian J. Forest. 4(1), 6–9 (2020). https://doi.org/10.13057/asianjfor/r040102 Mehrotra, R., Singh, P., Kandpal, H.: Near infrared spectroscopic investigation of wood. Thermochim. Acta. 507–508, 60–65 (2010). https://doi.org/10.1016/j.tca.2010.05.001 Mi, R., Li, T., Dalgo, D., Chen, C., Kuang, Y., He, S., Hu, L.: A clear, strong, and thermally insulated transparent wood for energy efficient windows. Adv. Funct. Mater. ID 1907511 (2019). https:// doi.org/10.1002/adfm.201907511 Michell, A.J.: Infra-red spectroscopy transformed—new applications in wood and pulping chemistry. Appita J. 41(5), 375–380 (1988) Michell, A.J.: An anomalous effect in the DRIFT spectra of woods and papers. J. Wood Chem. Technol. 11, 33–40 (1991). https://doi.org/10.1080/02773819108050260 Mitsui, K., Inagaki, T., Tsuchikawa, S.: Monitoring of hydroxyl groups in wood during heat treatment using NIR spectroscopy. Biomacromol. 9(1), 286–288 (2008). https://doi.org/10.1021/ bm7008069 Mora, C.R., Schimleck, L.R., Yoon, S.C., Thai, C.N.: Determination of basic density and moisture content of loblolly pine wood disks using a near infrared hyperspectral imaging system. J. Near Infrared Spectrosc. 19, 401–409 (2011)

48

1 Measurement Methods and Characterisation of the Optical Parameters …

Noupponen, M., Wikberg, H., Vuorinen, T., Jamsa, S., Viitaniemi, P.: Heat treated softwood exposed to weathering. J. Appl. Polym. Sci. 91(4), 2128–2134 (2004). https://doi.org/10.1002/app.13351 Ohkoshi, M.: FTIR-PAS study of light-induced changes in the surface of acetylated or polyethylene glycol-impregnated wood. J. Wood Sci. 48, 394–401 (2002). https://doi.org/10.1007/BF0077 0699 Owen, N.L., Thomas, D.W.: Infrared studies of “hard” and “soft” woods. Appl. Spectrosc. 43(3), 451–455 (1989) Pandey, K.K., Theagrajan, K.S.: Analysis of wood surfaces and ground wood by diffuse reflectance (DTIFT) and photoacoustic (PAS) Fourier transform infrared spectroscopic techniques. Holz Roh Werkstoff 55, 383–390 (1997). https://doi.org/10.1007/s001070050251 Pandey, K.K., Vuorinen, T.: Comparative study of photodegradation of wood by a UV laser and a xenon light source. Polym. Degrad. Stab. 93, 2138–2146 (2008). https://doi.org/10.1016/j.pol ymdegradstab.2008.08.013 Pasztory, Z., Tolvaj, L., Varga, D.: Effect of water leaching on photodegraded poplar wood monitored by IR spectroscopy. Wood Res. 65, 885–894 (2020). https://doi.org/10.37763/wr.1336-4561/65. 6.885894 Preklet, E., Tolvaj, L., Bejo, L., Varga, D.: Temperature dependence of wood photodegradation. Part 2: evaluation by Arrhenius law. J. Photochem. Photobiol. A: Chem. 356, 329–333 (2018). https://doi.org/10.1016/j.jphotochem.2018.01.008 Rao, A.N.S., Nagarajappa, G.B., Nair, S., Chathoth, A.M., Pandey, K.K.: Flexible transparent wood prepared from poplar veneer and polyvinyl alcohol. Composites Sci. Technol. 182, ID 107719 (2019). https://doi.org/10.1016/j.compscitech.2019.107719 Rosencwaig, A.: Photoacoustic spectroscopy of solids. Phys. Today 28(9), 23–29 (1975) Schramm, W.H.: The yellowing of woods. Jahresber. Verein. Angew. Bot. 3, 116–139 (1906a) Schramm, W.H.: The graying of woods. Jahresber. Verein. Angew. Bot. 3, 140–153 (1906b) Schwanninger, M., Rodrigues, C.J., Fackler, K.: A review of band assignments in near infrared spectra of wood and wood components. J. Near Infrared Spectrosc. 19, 287–308 (2011) Sitkei, G.: Zur Bestimmung der Farbtöne von Holzarten. Fortschrittbericht. No. 2. DWE, University of West-Hungary, Sopron (2013) Thygesen, L.G., Lundqvist, S.O.: NIR measurement of moisture content in wood under unstable temperature conditions. Part 1. Thermal effects in near infrared spectra of wood. J. Near Infrared Spectrosc. 8, 183–189 (2000). https://doi.org/10.1255/jnirs.277 Tolvaj, L., Faix, O.: Artificial ageing of wood monitored by DRIFT spectroscopy and CIE L*a*b* color measurements. I. Effect of UV Light. Holzforschung 49, 397–404 (1995). https://doi.org/ 10.1515/hfsg.1995.49.5.397 Tolvaj, L., Mitsui, K.: Surface preparation and direction dependence of DRIFT spectra of wood. Appl. Spectrosc. 58, 1137–1140 (2004) Tolvaj, L., Mitsui, K., Varga, D.: Validity limits of Kubelka-Munk theory for DRIFT spectra of photodegraded solid wood. Wood Sci. Technol. 4, 135–146 (2011). https://doi.org/10.1007/s00 226-010-0314-x Tolvaj, L., Molnar, Zs., Nemeth, R.: Photodegradation of wood at elevated temperature: infrared spectroscopic study. J. Photochem. Photobiol. B: Biol. 121:32–36 (2013). https://doi.org/10. 1016/j.jphotobiol.2013.02.007 Tolvaj, L., Molnar, Zs., Magoss, E.: Measurement of photodegradation-caused roughness of wood using a new optical method. J. Photochem. Photobiol. B: Biol. 134, 23–26 (2014). https://doi. org/10.1016/j.jphotobiol.2014.03.020 Tolvaj, L.: Traditions, anomalies, mistakes and recommendations in infrared spectrum measurement for wood. Wood Sci. Technol. 56, 1819–1834 (2022). https://doi.org/10.1007/s00226-022-014 25-7 Tsuchikawa, S., Hayashi, K., Tsutsumi, S.: Application of near-infrared spectrophotometry to the determination of fiber orientation and moisture content of wood. Mokuzai Gakkaishi 37(8), 758–760 (1991)

References

49

Tsuchikawa, S., Torii, M., Tsutsumi, S.: Application of near infrared spectrophotometry to wood. 4. Calibration equations for moisture content. Mokuzai Gakkaishi 42, 743–754 (1996) Tsuchikawa, S., Tsutsumi, S.: Adsorptive and capillary condensed water in biological material. J. Mat. Sci. Lett. 17, 661–663 (1998). https://doi.org/10.1023/A:1006672324163 Tsuchikawa, S., Siesler, H.W.: Near-infrared spectroscopic monitoring of the diffusion process of deuterium-labeled molecules in wood. Part I. Softwood. Appl. Spectrosc. 57, 667–674 (2003). https://doi.org/10.1366/000370203322005364 Tsuchikawa, S.: A review of recent near infrared research for wood and paper. Appl. Spectrosc. Rev. 42(1), 43–71 (2007). https://doi.org/10.1080/05704920601036707 Tsuchikawa, S., Schwanninger, M.: A review of recent near-infrared research for wood and paper (Part 2). Appl. Spectrosc. Rev. 48, 560–587 (2013). https://doi.org/10.1080/05704928.2011. 621079 Tsuchikawa, S., Kobori, H.: A review of recent application of near infrared spectroscopy to wood science and technology. J. Wood Sci. 61, 213–220 (2015). https://doi.org/10.1007/s10086-0151467-x Varga, D., Tolvaj, L., Tsuchikawa, S., Bejo, L., Preklet, E.: Temperature dependence of wood photodegradation monitored by infrared spectroscopy. J. Photochem. Photobiol. A: Chem. 348, 219–225 (2017). https://doi.org/10.1016/j.jphotochem.2017.08.040 Varga, D., Tolvaj, L., Molnar, Zs., Pasztory, Z.: Leaching effect of water on photodegraded hardwood species monitored by IR spectroscopy. Wood Sci. Technol. 54, 1407–1421 (2020). https://doi. org/10.1007/s00226-020-01204-2 Wislicenus, H.: Change in color of wood by the action of gases and vapours for technological improvement. Angewandte Chemie 23, 1441–1446 (1910) Yaddanapudi, H.S., Hickerson, N., Saini, S., Tiwari, A.: Fabrication and characterization of transparent wood for next generation smart building applications. Vacuum 146, 649–654 (2017). https://doi.org/10.1016/j.vacuum.2017.01.016 Yamauchi, S., Sudiyani, Y., Imamura, Y.: Depth profiling of weathered tropical wood using Fourier transform infrared photoacoustic spectroscopy. J. Wood Sci. 50, 433–438 (2004). https://doi. org/10.1007/s10086-003-0582-2 Zanuttini, M., Citroni, M., Martinez, M.J.: Application of diffuse reflectance infrared Fourier transform spectroscopy to the quantitative determination of acetyl groups. Holzforschung 52, 263–267 (1998). https://doi.org/10.1515/hfsg.1998.52.3.263 Zavarin, E., Jones, S.J., Cool, L.G.: Analysis of solid wood surfaces by diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy. J. Wood Chem. Technol. 10, 495–513 (1990). https:// doi.org/10.1080/02773819008050253 Hembree, DM., Smyrl, HR.: Anomalous Dispersion Effects in Diffuse Reflectance Infrared Fourier Transform Spectroscopy: A Study of Optical Geometries. Appl. Spectrosc. 43, 267-274 (1989)

Chapter 2

Measurement and Data Evaluation of Wood Colour and Gloss

Abstract The chapter covers the definition and measurement methods of colour and gloss parameters. The CIE chromaticity diagram is presented to understand the visualised meaning of colour hue, chroma and saturation. The CIE L*a*b* colour coordinate system is introduced as the most widely used system in research and in the wood industry. In addition to the advantages, the weaknesses of the CIE L*a*b* system are also presented. New and simple equations are introduced to determine the hue, lightness and saturation parameters of wood samples. These new equations give a better assessment of dark wood samples than the standardised CIE L*a*b* colour coordinate system. Although the L*, a*, b* colour parameters are independent variables, linear correlation was found between lightness and hue values. This linear correlation is presented in several situations. Examples are given to prove the weakness of the total colour change as one single parameter for monitoring colour alterations. The chapter provides extended experimental results in the field of gloss measurement. The basic regularities of the gloss of natural and treated wood surfaces are established. Keywords Wood · Colour parameters · Saturation · Lightness · Chromaticity diagram · Gloss parameters · Tropical wood

2.1 Introduction Colour and gloss are the most important aesthetic parameters of wood, which are mainly determined by the reflection properties. Colour parameters are defined by multiple equations. Since the calculation of colour parameter values is a difficult task without calculators or computers, regular colour measurement could only begin in the 1970s, after the advent of electronic calculators. The simple use of the computeraided colorimeter has accelerated the use of colour measurement in both scientific studies and industrial applications. This chapter presents the definition and measurement methods of colour and gloss parameters. New and simple equations are introduced to determine the hue, lightness

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 L. Tolvaj, Optical Properties of Wood, Smart Sensors, Measurement and Instrumentation 45, https://doi.org/10.1007/978-3-031-46906-0_2

51

52

2 Measurement and Data Evaluation of Wood Colour and Gloss

and saturation parameters of wood samples. Linear correlation is presented between lightness and hue values for natural and treated wood samples. Gloss measurement is discussed and supported by plenty of experimental results. The basic regularities of the gloss of natural and finished wood surfaces are also established in this chapter.

2.2 Colour and Its Measurement When sunlight falls on an object and it absorbs all the light, the surface of the object appears black. When all the light is reflected, the surface appears white. Most materials absorb part of the incident light and reflect and/or transmit the rest. The wavelength distribution of the reflected light determines the colour of the reflecting surface. Colour is a highly complex phenomenon. Individual wavelengths (between 380 and 750 nm) of sunlight determine the colours between violet and red, as presented by the rainbow. These individual colours of rainbow have maximum saturation. Illuminating a surface by helium–neon laser which emits light only at λ = 633 nm, the surface appears red independently on the surface properties. Illuminating a surface which has strong absorption around 495 nm by sunlight, the surface will also appear red. The human eye is unable to differentiate these two types of red. The second type of red is a compound colour where green is missing while all other “colours of sunlight” are present. These two colours (red and green) are referred to as complementary colours. (Some examples of complementary colour combinations are: red and green, yellow and purple, orange and blue, green and magenta.) The colour of wood is a compound type colour. The colour perceived by the human eye is not the colour corresponding to the wavelength of the light absorbed, but rather its complementary colour. The phenomenon described above makes the simulation of the human colour perception difficult. A light source illuminates an object by sending photons with different wavelengths. The object selectively absorbs these photons and reflects others. Reflected photons are absorbed by the cone shaped receptors in the eye and the brain generates the sensation of colour. This visual observing model shows that 3 items are necessary (illuminant, object and observer) to perceive colour. The colour measurement needs to reproduce the visual observing model. In order to build an instrument that will quantify human colour perception, each item in the visual observing situation must be represented by a table of numbers. The emission spectrum of a light source is the proper numerical representator of the illuminant. An object can be numerically represented by its reflection spectrum. The numerical representation of an observer is highly complicated procedure because of the complexity of the colour sensation. Neglecting the chemical, physiological and psychological aspects of colour sensing, only the relative sensitivity of the human eye to various wavelengths of light is discussed here. The cone shaped receptors in the eye are sensitive to different wavelength of light. Experiments were conducted to quantify the ability of the human eye to perceive colour. These experiments determined the average wavelength sensitivity of the tree different cones. CIE (Commission Internationale de l’Éclairage)

2.2 Colour and Its Measurement

53

Fig. 2.1 CIE 1964 colour matching functions of the 10° standard observer

determined the colour matching functions of the CIE 10° standard observer in 1964. These functions are introduced in Fig. 2.1. Thus, the observer can be quantified by the CIE colour matching functions. The three data collections: emission spectrum of the illuminant, reflectance spectrum of the object and the standard observer functions are used to measure colour. The CIE tristimulus colour values (X, Y, Z) as basic parameters of colour determination are calculated by multiplying and integrating the above listed three functions. 780

X = ∫ R(λ)S(λ)X(λ)dλ 380 780

Y = ∫ R(λ)S(λ)Y(λ)dλ 380 780

Z = ∫ R(λ)S(λ)Z(λ)dλ

(2.1)

380

where: R(λ) is the reflectance spectrum of the object, S(λ) is the emission spectrum of the illuminant and X(λ), Y(λ) and Z(λ) are the colour matching functions of standard observer. The computer aided spectrophotometer measures the reflectance spectrum and calculates the tristimulus values according to Eqs. (2.1). The spectrophotometer must be calibrated regularly to ensure correct colour measurement. The tristimulus colour values do not provide understandable colour presentation. However, they give basis for creating colour spaces that can easily be understood and represent colour differences correctly. Colour is expressed in terms of hue, lightness, and saturation (chroma, colourfulness). These three attributes of an individual colour can use as the dimensions of

54

2 Measurement and Data Evaluation of Wood Colour and Gloss

colour. A colour space specifies a colour numerically by the three colour coordinates. Colour spaces need to present clearly all three attributes of colour. There are different colour spaces. One of the simplest colour spaces is the CIE chromaticity coordinate system (1931). The coordinates x, y, z (in lower case) represent relative intensities of the tristimulus values as follows: x=

X X +Y + Z

y=

Y X +Y + Z

z=

Z X +Y + Z

(2.2)

As x + y + z = 1, only two coordinates are independent. In the CIE 1931 system, the x and y coordinates are used. Lightness is indicated by an additional value. Usually, the tristimulus value Y is used as a third coordinate perpendicular on the plane of the chromaticity diagram. Consequently, a colour dot is determined by the values Y, x and y. The CIE chromaticity diagram determines a horseshoe-shaped curve and the colour dots are located within this curve (Fig. 2.2). Each point of the curve boundary represents a pure hue of a single wavelength (in nanometres). There is a white (neutral) point (C) within the diagram with equal coordinates (x = y = 1/3 in case of “C” illuminant). The location of the neutral point depends on the illuminant. This neutral point represents the chromaticity values of black and white and grey between black and white. Colour dots of wood species are located within the dotted lines in Fig. 2.2. The chromaticity diagram is an excellent tool to demonstrate the meaning of colour hue and saturation. Equal colour hues are located on a straight line between the neutral point and the borderline of spectral colours. The relative distance of the measured point from the neutral point gives the saturation. If the measured point is close to C, the saturation is low and the hue is nearly grey. A point close to the boundary indicates high saturation. The ratio of the distance CP to CN gives a numerical index of saturation (Fig. 2.2). This definition of saturation has physical background and its maximum value is 100%. The diagram represents clearly the difference between chroma and saturation. Chroma is an absolute value represented by the CP distance, while saturation is a relative value determined by the CP/CN quotient. The main disadvantage of the chromaticity diagram is its visual non-uniformity. It means that equal distances in the x–y coordinate system are not perceived as being equal. The distances are compressed in the blue-red region. This weakness of the system is well visible on the horseshoe boundary line as well. The mostly used colour space is the CIE L*a*b* rectangular colour space: ISO 11664-4 (2008). (Not to be confused with the Hunter L, a, b, without stars!) The L*, a* and b* coordinates are determined by X, Y, Z tristimulus values. ( )1/3 Y − 16 L ∗ = 116 Y0

2.2 Colour and Its Measurement

55

Fig. 2.2 Chromaticity diagram and the possible region of wood colours

[(

)1/3 ( )1/3 ] X Y a ∗ = 500 − X0 Y0 [( ) ( )1/3 ] Y 1/3 Z ∗ − b = 200 Y0 Z0

(2.3)

where L* is the lightness axis: its value is between 0 and 100 (black-white). a* is the red-green axis: positive co-ordinates represent red hues, negative values represent green hues b* is the yellow-blue axis: positive co-ordinates represent yellow hues, negative values represent blue hues X, Y, Z are the tristimulus values X0 , Y0 , Z0 are the tristimulus values of a perfect reflecting diffuser (white etalon).

56

2 Measurement and Data Evaluation of Wood Colour and Gloss

Equations 2.3 are valid if X/X0 , Y/Y0 and Z/Z0 are higher than 0.008856. Below this limit modified equations are valid. The cube root parts of the equations must be replaced. The modified equations are as follows: Y Y0 ] [ X Y a ∗ = 3893.5 − X0 Y0 ] [ Y Z b∗ = 1557.4 − Y0 Z0

L ∗ = 903.3

(2.3a)

The CIE L*a*b* colour space is represented in Fig. 2.3. The proper terminology of colour difference (sample-target) is important. Positive ΔL* means that the sample is lighter than the target. If the sample is darker than the target, ΔL* is negative. The positive and negative values of a* and b* must be considered separately. If the positive a* is increasing, the sample will be more red. If the positive a* is decreasing, the sample will be less red or more neutral (not greener and doesn’t go towards green!). If the negative a* is increasing, the sample will be less green or more neutral (not redder and doesn’t go towards red!). If the negative a* is decreasing, the sample will be greener. Similar considerations can be applied to the b* co-ordinate as well. Unfortunately, lightness and brightness are confused in some scientific papers. The meaning and measurement methods of these two parameters are completely different. Determination of lightness (L*) uses almost the entire visible spectrum (Eqs. 2.1 and 2.3). In contrast, the determination of brightness uses a narrow wavelength interval Fig. 2.3 Schematic representation of the CIE L*a*b* colour space

2.2 Colour and Its Measurement

57

Fig. 2.4 Schematic representation of colour hue and chroma

around 457 nm. Brightness quantifies the percentage of blue light reflected from the surface of a material measured at a specific wavelength of 457 nm (with a half-peak bandwidth of 44 nm). Brightness does not indicate the colour or relative shade of the material, since a single number, reflectance value measured at 457 nm ignores all other wavelengths of light reflected across the visible spectrum. There are standards [TAPPI T452/GE (applying “C” illuminant), ISO 2470-1 (indoor daylight condition, applying “C” illuminant) and ISO 2470-2 (outdoor daylight condition, applying “D65” illuminant)] for defining the brightness and determining the measurement conditions. ISO brightness values are usually greater that TAPPI brightness values. Brightness is mainly used in the paper industry and in pulp and paper research. The CIE L*a*b* colour space can represent all three required colour parameters (lightness, hue, chroma). Lightness is represented by the vertical L* axis. Hue and chroma can be visualized on the a*–b* plane as introduced in Fig. 2.4. The hue (h*) and the chroma (C*) of a colour dot P(a*,b*) are presented in the a*– b* polar coordinate system. The L* axis is perpendicular to the plane (not presented here). The colour hue is indicated by the polar angle of the colour dot while the chroma is the distance between the colour dot and the L* axis. The h* is determined by the following equation. h ∗ = arc tan

b∗ a∗

(2.4)

Zero h* represents red colour, while 90°, 180° and 270° represent yellow, green and blue colours, respectively. The h* values are often given in radian. (180° = 3.14 rad). Chroma (C*) has exact physical meaning, as chroma value is represented by the distance between the colour dot and the L* axis. Chroma is the attribute that expresses the purity of a colour. If a colour dot is moving towards the L* axis, the chroma decreases meaning that the colour becomes more and more gray. C* is determined by the following equation.

58

2 Measurement and Data Evaluation of Wood Colour and Gloss

C∗ =

/ a ∗2 + b∗2

(2.5)

The definition of chroma does not determine the upper limit of chroma values. Chroma parameter does not represent correctly the visual saturation (S) for dark wood species. Chroma and saturation have similar meaning. While chroma defines the brilliance of a colour in absolute terms as a distance in the CIE colour space, (S) is more in line with the human perception of saturation. Saturation of a colour is defined by the combination of lightness and croma. Saturation is the proportion of pure chromatic colour in the total colour sensation (Lübbe 2010). It indicates the balance of the pure colour and white between 100 and 0%. The most saturated colour can be achieved by laser (emitting only one wavelength) illumination. 100% saturation means that there is no addition of gray to the hue. Saturation is defined by the following equation. (CIE L*a*b* system does not contain this definition.) C∗ S=√ 100(%) C ∗2 + L ∗2

(2.6)

The distance between two colour dots [P1 (a1 *,b1 *,L1 *) and P2 (a2 *, b2 *, L2 *)] is defined by the total colour difference (ΔE*). ΔE ∗ =

/ Δa ∗2 + Δb∗2 + ΔL ∗2

(2.7)

where: Δa* = a2 * − a1 *; Δb* = b2 *-b1 *; ΔL* = L2 * − L1 *. All colour measurements presented in this book were performed by a KonicaMinolta 2600d spectro-colorimeter. This spectrometer supplied the reflectance spectra in the visible region. The L*, a*, b* colour co-ordinates were calculated based on the D65 illuminant and 10° standard observer with a test-window diameter of 8 mm. If the colour alters (due to a treatment) the ΔE* is called: total colour change. The colour change is a highly complicated chemical process since a small amount of chemical alteration can generate intense colour change. Usually, all three above discussed colour parameters are affected during a colour change. The total colour difference seems to be an obvious and useful single parameter to represent the changes. This could be true in pass/fall decision, but it is not enough in scientific work. Nowadays, many papers demonstrate the colour alteration only by the total colour change ΔE*(Méndez-Mejías and Moya 2018; Straze et al. 2018; Zhu et al. 2021; Budakçi et al. 2023; Kilinç et al. 2023). It must be stated, however, that ΔE* is not always reliable by itself. In some cases, ΔE* values do not match the visual observations. Human eye is highly sensitive to the changes of hue determined by the alterations of a* and b*. The total colour change is mainly determined by the change of the dominant component. During the photodegradation of wood, the yellowing is the dominant change since the red hue shift is usually much smaller. These findings are well demonstrated in Fig. 2.5 representing the colour changes of beech caused by UV irradiation at 53 °C under wet condition (100% air humidity).

2.2 Colour and Its Measurement

59

Fig. 2.5 The change of colour parameters for beech caused by UV light irradiation under wet condition at 53 °C. (ΔE. is the total colour change calculated without the redness change.)

It is clearly visible that the total colour change ΔE* is determined mainly by the yellowness change. For comparison the total colour change value was calculated without the redness change (depicted as ΔE. ) as well. The difference between the two total colour change curves (ΔE* and ΔE’) is negligible suggesting that the effect of redness change is also negligible from point of view of the total colour change. This redness change is, however, high enough to be visible to the naked eye. This example shows the disadvantage of the total colour change as one single indicator. Another example is given in Fig. 2.6. The colour of black locust can be modified by steaming considerably. Its dominant unattractive yellow colour decreases and the red colour increases similarly during the first day of the process. The redness decreases slowly but gradually for the remaining period of the treatment. The total colour change (ΔE. ), calculated without the lightness change does not follow the tendency of the redness change, however. Obviously, it is determined by the more intensive yellowness change after one day steaming. Banadics et al. (2016) investigated the colour similarity between dark exotic species and steamed black locust. It was observed that the total colour difference (ΔE*) is inefficient in monitoring the visually observed colour difference, if the difference was created by the alteration of red hue component. These examples prove the weakness of the total colour change as one single parameter for monitoring colour alterations. The total colour change is determined mainly by the dominant changes and it disregards the small but visible alterations. Therefore, the three standalone colour coordinates supply much more and detailed information than the total colour change. Colour is a highly complex phenomenon. Chromophore molecules with double bound chemical systems are responsible for the colour of wood. Chromophores have

60

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.6 Change of black locust colour parameters caused by steaming at 115 °C. (ΔE. is the total colour change calculated without the lightness change.)

special absorption bands in the visible region. Wood polyoses do not absorb visible light. Lignin, absorbs and reflects light below 400 nm wavelength and has a paleyellow colour. In general, this yellow colour is typical for the sapwood of almost all wood species. As sapwood turns into heartwood, its oldest cavities are filled with extractives including terpenes, polyphenols, wood resins, tannins, sugars and fatty acids. These constituents make significant contributions to such properties of wood as colour, odor and decay resistance. Some of these materials are reflecting more light at higher wavelengths and have deeper colours. Colouring materials in wood are mainly the aromatic systems. Basic colours of these compounds are red, yellow and brown. Other colours result from mixtures of these basic colours. Red colouring agents are mainly the substituted derivatives of naphthoquinone and anthraquinone with varying numbers of different groups in different positions (Mills and White 2011). Characteristic groups are the methyl, hydroxyl, carboxyl, carbonyl and hydroxymethylene. The colouring material is a mixture of several components such as alizarin, purpurin, rubiadin, brasilein, haematin etc. Sappan wood (Caesalpinia sappan), also called Brasilwood in trade, has the main constituent of brazilin which is oxidized by air and light to produce brasilein. Haematin is the main colouring agent of logwood from Haematoxylon campechianum in Central America. Yellow-coloured constituents are mostly flavonoid components such as quercetin in many wood species, rhamnetin from Rhamnus species, fisetin from Cotinus coggygria and Rhus verniciflua (Urushi), berberin from Berberis and Phellodendron species. In many cases several of compounds may exist together giving different shades. For instance, the Brasil wood contains not only brasilein but also quercetin and emodin, producing its unique orange- red colour (Badami et al.2004). Brown colours are mostly produced by tannins and their oxidized and polymerized derivatives (Mills and White 2011). It is believed, for example, that the characteristic yellow colour of Urushi (Rhus verniciflua) is due to flavonoids such as fisetin and sulfetin mainly. The age of a tree in many cases influences the quantity and colour of extractive materials in the heartwood considerably. The Karri wood (Eucalyptus diversicolor)

2.2 Colour and Its Measurement

61

has darker red colour over time. It is thought that the richness of colour may also be imparted by the soil where the trees were grown. For example, the deepest red Myrtle (Nothofagus cunninghamii) comes from highly fertile soils on basalt. The surface texture of wood can modify the colour considerably. Slender wood cells are arranged in layers in one direction. During woodworking operations, these cells are cut in a given plane in relation to the grain. Depending on the direction of cut, a different microphysics-relief, topography can be achieved. Furthermore, the cut of wood surface is always associated with brittle fracturing of cell walls causing diffuse reflectance. Because the microrelief of the surface has generally a definite direction, the irradiating direction of light may slightly modify the wood colour. If the vessels and lumens are cut, bottom surface of lumens reflect the incident light similarly to bundles of small mirrors. The bunch of mirrors are aligned and divided by cells walls. The incident light is much more reflected parallel to the grain (directional reflection) giving a high gloss. In contrast, the light will be more scattered (diffusely reflected) perpendicular to the grain. For example, the sanding of a wood surface always endows with a greyish blear (veil) which can be removed using wax, or oil treatment (Csanady et al. 2015). The colour coordinates (L*, h*, C*) are independent variables. There are no direct correlations among the coordinates that would be valid in all conditions. Non the less, it is possible to find corelations which are valid only in special conditions. One example is the correlation between lightness and hue in case of wood. Species with high lightness value have hue close to yellow, while dark species have brown colour. The hue angle for most of the natural, untreated wood species is between 45° and 90°. There are only a few extremely dark species (mainly tropical wood species) with hue angle below 45°. The colour of European wood species is between yellow and brown. The hue angle limits for European wood species are 60° and 85°. Natural colour of European species was determined by spectrophotometric colour measurement. The involved species were: alder (1), ash (2), beech (3), beech red heart (4), birch (5), black locust, cherry (Prunus avium 7), cherry (Prunus serotina 8), Douglas fir sapwood (9), Douglas fir heartwood (10), hornbeam (11), larch sapwood (12), larch heartwood (13), lime (14), maple (15), oak sapwood, oak heartwood, poplar (18), Scots pine sapwood (19), Scots pine heartwood (20), spruce (21), Turkey oak sapwood (22), Turkey oak heartwood. Results are presented here according to a previously published paper (Tolvaj et al. 2013; with written permission of Wood Research). The colour dots of species are presented in Fig. 2.7. The numbers show the places of the individual colour dots of the species on the L*–h* plane. Separate dots (black locust heartwood, oak (Quercus petrea) sap- and heartwood, Turkey oak heartwood) were excluded from the regression analysis. For the rest of the investigated European wood species, the analysis revealed strong correlation between their lightness and hue values with coefficient of determination R2 = 0.90. The high extractive content of outlier species certainly justifies their removal from the analysis. For instance, black locust has the highest extractive content among the trees grown in Europe. The yellowish colour of this hardwood is due to its high robinetin content. Furthermore,

62

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.7 Colour dots of European wood species in the L*–h* plane

all of the eliminated species are ring-porous angiosperms so that the large vertically cut vessels may have influenced the lightness under incident illumination. It appears that except for a few species, the lightness of wood is a good predictor of its hue. The developed prediction equation numerically supports the general observations that light wooden colours are more yellowish while dark species are rather brownish. This finding gives the possibility to monitor the colour alteration effects of a process, measuring only the lightness. Fortunately, the lightness depends only on the Y tristimulus colour value and this component can be measured by a proper colour filter and a detector. As this kind of measurement does not need any expensive colorimeter, it allows a fast and easy colour monitoring also during steaming. A wide range of colours can be imparted to black locust wood by steaming, from greenish yellow up to chocolate brown applying different treatment times and temperatures. This wide range of colours is represented in Fig. 2.8 where the hue ranges from 0.95 to 1.5 units (from 55° to 86°). In spite of this wide colour range, linear correlation was found between the lightness and the colour hue as presented in Fig. 2.8. The coefficients of determination show high correlation between the hue and the lightness, R2 was 0.99 in all cases (Tolvaj and Nemeth 2008). The dots representing the unsteamed samples are located in the top-right corner followed by the colour dots of the steamed samples to the left generated by 1; 2; 4; 6; 9; 12; 15; 18- and 22-day steaming. Hue data show that increasing steaming temperatures induce decreasing hue values presenting the colour shift towards brown. Steaming generated colours of species other than black locust (poplar, Scots pine, spruce) are also featured by linear correlation between lightness and hue values (Tolvaj et al. 2012; Banadics 2023). Regression equations (see in Fig. 2.8) give the possibility to

2.2 Colour and Its Measurement

63

Fig. 2.8 Correlation between lightness and hue of steamed black locust. (Colour dots of unsteamed wood are in the top-right corner followed by the dots of steamed wood with increasing steaming time.)

calculate the hue values after measuring the lightness. Measurement of lightness is much easier than that of hue because lightness is determined by the Y tristimulus value alone while the calculation of hue requires all tree (X, Y, Z) tristimulus values. The linear correlation between lightness and hue is a general relationship for wood. The colour change during photodegradation also presents this correlation (Tolvaj and Mitsui 2010). To demonstrate this relationship, black locust, spruce and Japanese cypress (Chamaecyparis obtusa) samples were put outdoor (only in sunny periods) and the colour values were measured after 5, 10; 20; 30; 60 and 120 h of irradiation. The measured lightness and the calculated hue values are presented in Fig. 2.9. The coefficients of determination (R2 ) are presented in Fig. 2.9 as well. The initial colour dots (representing the colour of untreated samples) are located in the top-right corner, followed by the colour dots of irradiated samples with increasing treatment time. The trendlines are straight lines and the values of coefficients of determination are above 0.9 representing the strong linear correlation between lightness and hue values. The slope of the trendlines is different. Trendline of black locust is the steepest while that of spruce has the lowest gradient. This trend reflects the diverse extractive content of the investigated species. Chemical changes during photodegradation are highly temperature dependent. These changes generate intensive colour changes. A previous study demonstrated that the elevated temperature multiplied the effect of photodegradation (Tolvaj et al. 2015). It is interesting to test whether the linear correlation between lightness and hue applies also to such intensive colour change induced under high temperature conditions. The UV irradiation was produced by mercury vapour lamp and the experiments were carried out at 30, 80, 120 and 160 °C. The irradiation times were different

64

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.9 Correlation between lightness and hue during photodegradation of different wood species caused by sun radiation

depending on the applied temperatures. Increasing temperature required the shortening of the irradiation time in order to obtain comparable colour changes. Irradiation intervals were 8, 20, 40, 90 and 200 h at 30 and 80 °C. The irradiation at 120 °C required shorter periods (6, 11, 16 and 36 h). The colour change at 160 °C was extremely intensive. That is why the treatment was interrupted after 16 h, and the colour was measured after 6, 11 and 16 h of irradiation. Figure 2.10 presents that the 16-h irradiation at 160 °C generated much more intensive colour alteration than the 200-h treatment at 30 °C. These data show that the photodegradation induced colour changes are highly temperature dependent (Fig. 2.10). The initial colour dots (representing the colour of untreated samples) are located in the top-right corner, followed by the colour dots of irradiated samples with increasing treatment time. The trendlines are straight lines and the values of coefficients of determination are above 0.9 representing the strong linear correlation between lightness and hue values. Trendline gradients are almost equal representing that the chemical changes are the same, however, the intensities are different depending on the temperature. Correlation between the lightness and hue parameters of wood during photodegradation and water leaching also follows the linear scheeme (Kannar et al. 2018) The above presented examples strengthen that the linear correlation between lightness and hue is a strong relationship for wood material. There are some generally valid observations regarding the colour of wood species. Reflection spectrum of wood always has a continuously increasing character in the visible light range between 400 and 700 nm. It never has maximum that is characteristic for pure colour hues meaning that colours of wood are always mixed colours.

2.2 Colour and Its Measurement

65

Fig. 2.10 Correlation between lightness and hue during photodegradation of spruce specimens caused by mercury lamp irradiation at different temperatures

There is agreement among scientist that the reflection spectrum uniquely characterises and determines the colour that can be seen with naked eye. Therefore, the best way towards a better evaluation method is using the measured reflection spectrum directly as much as possible. The calculation of L* in CIE L*a*b* system leaves the X and Z tristimulus values out of account and uses only the Y value (Eq. 2.3). It is hard to imagine that other parts of the reflection spectrum do not contribute to the perceptible lightness. Figure 2.1 shows that Y tristimulus value ignores the reflected light in the 400–450 nm and 650–700 wavelength intervals. Non the less, human eye perceives these reflected photons but with low intensity. The first wavelength interval covers half of the wavelength interval belonging to the Z( colour)matching function. √ The geometric mean of X and Y tristimulus values X Y might be a proper value for lightness determination. This geometric mean covers the 400–450 and 650– 700 nm wavelength intervals as well, which is ignored by Y. The geometric mean is often used as an “average” of two independent variable. The correlation between √ X Y and L* was determined experimentally. The results are plotted in Fig. 2.11. The trendline is curved but √ the deviations are almost zero. This fact confirms strong correlation between L* and X Y . Figure √ that L* is more sensitive √ 2.11 demonstrates below L* = 50 to the changes than X Y . However, X Y is more sensitive to the changes than L* above L* = 60. It is important to mention √ that most of the wood species are located in this second L* ≥ 60 interval. (L* and X Y have equal sensitivity between 50 and 60 L* values.) The only problem is that the correlation is not linear. This fact makes the calculation more complex. Sitkei (2020) recommended a simpler solution (described in Sect. 2.2.1). The physical lightness Rm is defined by the integral of the reflectance spectrum between 400 and 700 nm (Eq. 1.9). Its meaning is the area below the reflectance

66

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.11 Relationship between CIE lightness (L*) and



XY

curve which is proportionate to the √ number of reflected photons. The correlation between physical lightness (Rm ) and X Y is presented in Fig. 2.12. The measured data √ of many wood species show linear correlation between physical lightness (Rm ) and X Y . The strong correlation is confirmed by the high value of coefficient of determination (R2 = 0.98). √ As the correlation between physical lightness (Rm ) and X Y is linear, the correlation between physical lightness (Rm ) and L* can be determined. Comparing Figs. 2.11 and √ 2.12, it can be stated that the correlation between Rm and L* is the same as between X Y and L*. As wood colours are located between brown and yellow occupying a relatively small area, some simplifying can be done. One is that the equation of hue (arc tan (b*/ a) can be simplified by b*/a* quotient. It is possible because the (arc tan) function is close to linear in the range of wood colours (Fig. 2.13). The (arc tan) function is basically curved. None the less, the linear substitution has high coefficient of determination (R2 = 0.92) in the examined angle interval. Consequently, the hue (h* = b*/a*) will be used in this study in some cases. For precise comparison the usage of (arc tan) function is needed. It is mentioned in Sect. 1.3.2. that the intersection wavelength λ0 characterises the colour hue of wood. The definition of this characteristic wavelength is presented in Fig. 1.15. The representation of different colours by intersection wavelength λ0 is presented in Fig. 2.14. The intersection wavelength of many wood species was determined and the results were plotted in Fig. 2.15. The correlation is close to linear between intersection

2.2 Colour and Its Measurement Fig. 2.12 Relationship between physical lightness √ (Rm ) and X Y

Fig. 2.13 The (arc tan) function in the 57–86-degree range

Fig. 2.14 Representation of different colours by intersection wavelength λ0

67

68

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.15 Plot of experimental results representing the correlation between λ0 and h*

wavelength and colour hue, but the real trendline is slightly curved. The correlation is rather week at h* values below 67°, where the dark tropical species are located. The correlation is much better above the h* = 67° limit, where the European wood species are located. Below 67°, the h* scale is more expanded occupying 63% of the whole horizontal scale. The λ0 values cover here 43% of the whole vertical scale. Above 67°, the λ0 scale is more expanded occupying 57% of the whole vertical scale. The h* values cover here 37% of the whole horizontal scale. These results show that the h* parameter is highly sensitive below 67° and less sensitive above 67° to monitor the colour alterations. In contrast, the sensitivity of the λ0 parameter is more balanced than the similar property of the h* parameter in the whole wavelength range. It is important to note that the new colour evaluation methods proposed here were derived from plenty of measured colour parameters of a wide range of wood species. Since the colour data of the wood samples occupy a small volume in the L*, a*, b* colour space, the proposed equations may only be valid for the wood samples.

2.2.1 Anomalies in the CIE Lab Colour System Regarding to the Colour Determination of Wood The adequate designation of a colour is an old problem and the first international standard for colour assessment, the CIE Lab evaluation system, which was modified several times, is dated back to 1931. The development of a physically correct evaluation system is encumbered by the fact that colour does not physically exist, only our eyes see different colours depending on the wavelength distribution of the reflected

2.2 Colour and Its Measurement

69

light from a given surface. It means that also the colour vision provided by the human eye and brain play an important role. Due to the fact that the colour vision of human eye cannot be exactly measured in all its details, an inherent feature of all colour estimation methods is the use of some hypothetical relations with consequences. A colour space is uniform if equal distances in the coordinate diagram correspond to equal perceived colour differences. None of the colour coordinate systems created is perfect. CIE L*a*b* system is over expanded in the yellow zone. In contrast, the second most frequently used Hunter L, a, b system is over expanded in the blue region. Both systems show anomalies in the very white and very black regions (where L* value is close to 100 or to zero). Regarding the colour of wood, the dark region is highlighted. The observed anomalies in the CIE L*a*b* system are the followings: lightness values of dark wood samples, evaluation of colour hue evaluation of saturation. The weakness of the L* determination observed for very dark wood species. The measured L*value is usually lower than the perceived lightness. Reflectance spectrum of dark species hardly have measurable intensities below 600 nm (Fig. 2.16). The studied species were sassafras (Atherosperma moschatum), Bois de Rose (Dalbergia maritima) and Saint Johansson’s bread (Ceratonia siliqua); their measured L* values were 1.58, 6.63 and 11.1, respectively. Figure 2.16 shows the shape of the Y colour matching function of standard observer and the sum of X, Y and Z colour matching functions to present how they take account of the reflectance intensities of dark wood species. The values of colour matching functions were multiplied by 10 for the proper comparison to the reflectance values. The X + Y + Z function shows the wavelength sensitivity of the human eye. It is vell visible that the Y colour matching function ignores most part of the blue region where the human eye is sensitive. Less but noticeable intensity difference can be seen in the red region between Y and X + Y + Z as well. Figure 2.16 represents that only some percentage of area under the reflectance curve is covered by the Y function. The much bigger part of the reflected light is not integrated into the L* value by the Y colour matching function. This might be the reason why the perceived lightness is higher than the measured one. The X + Y + Z function covers a much larger part of the reflectance spectra than the Y function. This finding suggests that the X + Y + Z parameter might be a proper function to determine the lightness value of dark wood species. Sitkei (2020) introduced a new equation for determining the lightness by using all three tristimulus values (X, Y, Z). ) ( X + Y + Z 0.4 L = 16.7 3

(2.8)

70

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.16 Reflectance spectra of very dark Bois de Rose, sassafras and Saint Johansson’s bread wood species and the relative intensity distribution of Y colour matching function of standard observer together with the sum of X, Y and Z values (multiplied by 10)

Sitkei’s formula accounts the reflected red (and blue) photons as well. For determining the usefulness of Eq. 2.8, 48 wood samples having lightness (L*) between 83 and 1 units were involved to the test. The CIE lightness (L*) and lightness (L) calculated by Eq. 2.8 were determined for all chosen samples. The generated datacouples are plotted in Fig. 2.17. (The aim was to keep 2 L* units of distances among the samples, but in some cases the proper sample could not be found. Consequently, the distance was 1 or 3 units in some cases.) The values of the two types of lightness (L* and L) were almost identical from the top to 40 units. Below this threshold, L* values decreased faster than L values. The last difference was 6.5 units. Decrease of L values was more balanced that the decrease of L*. The results suggest that Eq. 2.8 provides more equalised lightness values for dark wood species than the CIE lightness (L*). Colour hue is determined by the b*/a* ratio according to the following equation b∗ (Y/Y0 )1/3 − (Z /Z 0 )1/3 = 0.4 a∗ (X/ X 0 )1/3 − (Y /Y0 )1/3 where X0 , Y0 and Z0 are constant values. The denominator is determined by the difference between X and Y tristimulus values. If this difference is small, a small alteration in X and Y can cause great deviation in the calculated value of a*. It means that the determination of colour hue may be prone to calculation error because of the uncertainty of a* value. In our wide-ranging investigations, hue anomalies were found generated by the CIE L*a*b* evaluation method in several cases. The following examples demonstrate

2.2 Colour and Its Measurement

71

Fig. 2.17 Lightness couples of different wood species between 83 and 1 units. (L* represents the CIE lightness and L characterises the lightness determined by Eq. 2.8. The smallest 5 L* values were determined by Eq. 2.3a.)

the nature of these type of anomalies. Studying the colour data of Kashmir walnut (Juglans regia) and Tasmanian blackwood (Acacia melanoxylon) the standard CIEL*a*b* system gives the same colour hue for both species although visually they show quite different tint (Table 2.1). Additionally, their reflection curves are not similar either (Fig. 2.18). Blackwood has much higher reflectance values in the (650– 700 nm) red interval than Kashmir walnut. This fact supports the visual observation that the tint of blackwood is much redder than that of Kashmir walnut. It is important to mention that the (λ0 ) intersection wavelength (see Figs. 1.15 and 2.14) shows correctly the perceived hues of both species. Another interesting example is presented by the golden-brown Kadam wood (Anthocephalus chinensis) and the middle brown African mahogany (Khaya ivorensis). Despite the widely different perceived hues, the CIE L*a*b* assessment method gives almost the same hue (h*) values (Table 2.1). The reflection spectra of Table 2.1 Colour parameters and perceived hues of selected wood species Species Kashmir walnut

X 7.98

Y 5.72

L*

a*

b*

b*/a*

h*

λ0

Perceived hue

38.4

15.07

27.94

1.85

61.6

565

Yellowish brown

Blackwood

11.8

10.33

26.69

26.48

48.14

1.82

61.2

589

Brownish red

Kadam wood

27.32

24.29

56.99

15.65

43.5

2.78

70.2

562

Golden-brown

7.73

6.42

30.46

16.57

43.73

2.64

69.3

576

Brown

African mahogany

72

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.18 Reflectance spectra of selected wood species

these species are also quite different (Fig. 2.18). The (λ0 ) intersection wavelength shows correctly the perceived hues of both species. There are two yellowish brown species in the list of Table 2.1 (Kashmir walnut and kadam wood). Their perceived hue is close to each other. Although the intersection wavelengths show small difference (only 3 nm), the hue difference is 8.6°, which is rather high deviation. The observed anomalies were caused by the unpunctuality of a* values generated by low difference between the values of X and Y tristimulus values (Table 2.1). Only kadam wood has a proper (X–Y) difference (3 units) which is acceptably high. It can be concluded that the CIE hue (h*) gives properly the perceived hue if the (X–Y) tristimulus value difference is considerably higher than one unit. The (λ0 ) intersection wavelength determines the hue properly in these special cases as well. To find the similarities and the differences between chroma and saturation values of wood samples, untreated and steamed specimens were chosen to create a list with decreasing lightness value (Table 2.2.). This summary chart contains the colour coordinates (L*, a*, b*), chroma (C*) and saturation (S) values as well as their difference (S-C*) and quotient (S/C*) values. Mostly European species as well as steamed poplar and black locus samples were chosen. These samples covered the 85–37 lightness value range. Tropical samples were also chosen (below the doubled line in Table 2.1) to present even lower lightness values. Colour parameters of European species showed some regularities but tropical species gave random S and C* data. Saturation values were higher than the chroma values in all cases. Similar results were found by Klement et al. (2019, 2021). Both saturation and chroma values increased by increasing lightness values up to L* = 65 and decreased afterward (within the 37–85 interval). The difference and the quotient

2.2 Colour and Its Measurement

73

Table 2.2 Colour parameters of selected wood samples generating a decreasing lightness value list b*

C*

S

S-C*

S/C*

85.18

3.55

20.35

20.66

23.57

2.91

1.14

84.02

4.12

21.14

21.53

24.83

3.29

1.15

83.07

4.35

21.46

21.90

25.49

3.59

1.16

82.15

4.45

22.66

23.09

27.06

3.97

1.17

81.46

4.70

23.37

23.83

28.08

4.25

1.18

80.58

5.10

23.73

24.27

28.84

4.57

1.19

79.37

4.63

23.94

24.39

29.37

4.98

1.20

74.58

7.69

27.72

28.77

35.99

7.22

1.25

71.39

13.22

29.57

32.39

41.32

8.93

1.28

70.59

10.56

30.51

32.29

41.60

9.31

1.29

64.96

4.34

32.49

32.78

45.05

12.27

1.37

63.48

8.75

28.54

29.85

42.56

12.70

1.43

55.62

8.72

23.07

24.66

40.54

15.87

1.64

49.95

8.64

19.34

21.18

39.04

17.86

1.84

45.92

8.53

17.13

19.14

38.47

19.33

2.01

43.02

8.40

15.51

17.63

37.93

20.29

2.15

40.96

8.26

14.47

16.66

37.67

21.02

2.26

38.78

8.01

13.24

15.47

37.06

21.59

2.40

37.30

7.70

12.00

14.26

35.71

21.45

2.50

36.27

20.27

20.96

29.16

62.66

33.50

2.15

35.86

19.91

31.27

37.07

71.87

34.80

1.94

34.72

0.64

5.89

5.91

16.78

10.87

2.84

33.88

14.71

14.88

20.92

52.55

31.62

2.51

33.34

20.3

38.01

43.09

79.09

36.00

1.84

31.78

15.32

12.67

19.88

53.03

33.15

2.67

29.00

33.01

49.58

59.56

89.91

30.35

1.51

28.11

19.84

37.60

42.51

83.41

40.90

1.96

27.46

14.29

19.87

24.47

66.54

42.06

2.72

26.10

9.76

25.84

27.62

72.68

45.06

2.63

L*

a*

of S and C* decreased continuously with increasing lightness values for European species. Saturation values of European species are below 50% because the colour of wood is a mixed colour and the individual colour dots are fare to the boundary of the colour space where the pure colours are located. As the lightness value decreases, the boundary of the colour space gets closer and closer to the lightness axis. Within a short distance, a small change in chroma can result in a large change in saturation percentage (see Fig. 2.2 for the meaning of chroma (CP distance) and saturation (CP/

74

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.19 Correlation between lightness (L*) values and the difference between saturation and chroma values (S–C*)

CN)). This phenomenon is the reason why dark wood species have high saturation values. Figure 2.19 shows the difference between saturation and chroma as a function of lightness for 82 different samples. The S–C* values show an exactly linear increase with decreasing lightness values up to L* = 57. This linear increase represents that the change in chroma is proportional to the change in saturation. The linear increase persists at lower lightness values as well, but the S-C* values show an increasing dispersion, indicating that saturation increases faster than chroma at low lightness values. Plenty of colour evaluation results revealed that the best way toward a better colour evaluation method is to use the measured reflection spectrum directly by the tristimulus values as simple as possible. Definition equations of chroma (C*) and saturation (S) do not fulfil this requirement. Their equations (Eqs. 2.5 and 2.6) are highly complicated functions of the tristimulus values as it is visible below. That is why we tried to find simpler equation. [ [( ) [( |{ ) ) ]}2 ( )1/3 ]}2 { ( | Y 1/3 X 1/3 Y Z 1/3 − + 200 − C ∗ = | 500 X0 Y0 Y0 Z0 /{ ]} [( [( ) )1/3 ( )1/3 ]}2 { 1/3 ( Z )1/3 2 500 XX − YY + 200 YY − Z 0 0 0 0 S = /{ [( [( ) [ ( ) ]2 100(%) ]} )1/3 ( )1/3 ]}2 { 1/3 ( Z )1/3 2 1/3 500 XX − YY + 200 YY − Z + 116 YY − 16 0

0

0

0

0

A detailed analysis of the saturation data of a wide range of wood samples revealed the surprising result that there is a very √ close correlation between the√saturation values and the dimensionless quantity Z / X · Y (Sitkei 2020), where X Y is the

2.2 Colour and Its Measurement

75

geometric mean of X and Y tristimulus values. The correlation is described by the following simple equation: [

(

Ss = 100 1 − Valid if Z
15 GU

2.3.1 Regularities Related to the Gloss Parameters of Natural and Treated Wood Surfaces The following section is presented here according to Csanady et al. (2015). The surface of a planned wood sample looks like a layer of cylindrical halved tubes. In the direction along the fibres a combination of specular and diffuse reflection occurs while across the fibres mainly diffuse reflection can be expected. Polishing the wood surface, the reflection ability will be enhanced in all directions and, therefore, across the fibre a combined specular and diffuse reflection will take place. It was observed that the reflection of a surface is more directed for incident light parallel to the grain compared to perpendicularly incident light (Schmidt and Eckert 1935). Boruvka et al. (2021) investigated the gloss change of birch wood during thermal treatment. Five stages of heat treatment were used, ranging from 160 to 200 °C, with 10 °C increments applying a peak treatment duration of 3 h for each level. All standardized incident angles (20°, 60° and 85°) were employed. The measured gloss values were considerably different depending on the standardized incident angles. It means that the measured gloss values are highly influenced by the illuminating angle. This circumstance alone may cause difficulties in the correct interpretation of readings generated by different measuring angles. Gloss measurements were undertaken on samples of wood species from all over the world having wide gloss range. Gloss values were measured using 20°, 60° and 85° measuring angles to discover the differences. Figure 2.22 presents the gloss values of different natural and finished wood specimens measured with angle of 20° and 60°. The measured values are located on a curved line with low deviations, representing strong correlation. Gloss values show that the 60° is a more sensitive measurement position below 60-unit presented by GU60 . This interval counts only 22 units using 20° incident angle. In contrast, the GU60 values occupy 60-unit interval. The slope of the trendline between 60 and 80 GU60 units is close to 45° proving the equal sensitivity of GU20 and GU60 measurement. The data of Fig. 2.22 show that the recommended measuring angle is 60° for wood. (Gloss values of natural wood are below 80 units.) The gloss dot in the upper right corner does not belong to real wood. It was measured on the surface of a laminated fibreboard. Natural wood does not produce such a high gloss. This dot was inserted to demonstrate that 20° measurement angle is more sensitive at high gloss than 60° measurement angle. (If the GU60 value is above 80 units.)

2.3 Gloss and Its Measurement

79

Fig. 2.22 Correlation between GU20 and GU60 values for different wood surfaces

The 85° measurement angle is located closer to 60° than 20°. Consequently, most of the measured gloss values at 85° have stronger correlation to the GU60 values than to the GU20 values. For determining the correlation between GU60 and GU85 values, measurements were undertaken on wood samples from all over the world and on different furniture’s including also an old Biedermeier cabinet. Some wooden artefacts, carvings and turneries, were also measured with their original surface finishing. Machined, sanded and polished surfaces were also involved in the test. The measured glossiness data are presented in a GU60 –GU85 coordinate system. The measured points indicate the range of gloss on the given surface as well. The results showed that gloss ratios measured with 85º and 60º incident angles are distinct for a given surface and these ratios are always constant. The trendlines were always straight lines crossing the origin of the coordinate system. Consequently, the GU85 / GU60 ratios can be taken as constant for a given wood species and surface condition. The surface roughness was characterized by the following parameters: Rpk (average height of protruding peaks), Rk (vertical difference in the core section). Figure 2.23 shows the correlation between GU85 and GU60 values for planed sugi heartwood and sapwood in both parallel and perpendicular directions to the fibres. The values of surface roughness (Rpk + Rk ) were between 6.8 and 7.7 μm. The value of the GU85 /GU60 ratio is close to 1.0 in the latewood if the incident light is parallel to the grain, both in heartwood and sapwood. The same ratio is less than 1.0 in the other cases. Using high incident and viewing angle (85°), the protruding parts of the surface (bumpiness) disturb the local reflection and, as a consequence, the GU85 data show lower gloss. The earlywood is bumpier than latewood therefore readings with 85º angle have lower values. Furthermore, the GU60 values perpendicular to the grains are always lower than those parallel to the grains. Latewood readings for 60° angle are almost the same independently of grain direction in case of sapwood, however, their GU85 /GU60 ratios are different.

80

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.23 Gloss ratios for sugi sapwood and heartwood. (Rpk + Rk = 6.8–7.7 μm) (Reproduced with the written permission from Springer Nature)

Sugi has relatively low density (430 kg/m3 ). Its glossiness data are below 10 units measured with both 60° and 85° incident angles. Glossiness values measured with 85° incident angle are more sensitive to the density of wood than those measured with 60°. It is well presented in Fig. 2.24 by karri wood with a density of 920 kg/m3 . Its glossiness is less than 10 units if measured under 60° similarly to that of sugi, however, the data measured under 85° are between 10 and 34 units. The GU85 /GU60 ratios are 3.5 and 2.0 parallel and perpendicular to the grain, respectively. Density of rosewood is even higher (1120 kg/m3 ) than that of karri, and its machined surface is usually smoother (Rpk + Rk = 3.6–3.9 μm) than the surface Fig. 2.24 Gloss ratios for karri wood (Eucalyptus diversicolor). (Rpk + Rk = 7 μm) (Reproduced with the written permission from Springer Nature)

2.3 Gloss and Its Measurement

81

Fig. 2.25 Gloss properties of the extreme dense (1120 kg/m3 ) rosewood (Dalbergia cochinchinensis). (Rpk + Rk = 3.6–3.9 μm) (Reproduced with the written permission from Springer Nature)

of karri processed with the same machining technology. The Rpk + Rk roughness parameter for karri is twice as high than the same parameter of rosewood. Gloss ratios for rosewood are extremely high, also across the grain. The GU85 /GU60 ratios are 5.7 and 4.5 parallel and perpendicular to the grain, respectively (Fig. 2.25). Irregular growth structures in wood may diminish or eliminate the difference between the gloss ratios parallel and perpendicular to the grains. Figure 2.26 shows the gloss ratios for two eucalyptus wood species with strongly irregular growth. The grains had more or less a curly character and random direction in relation to the main axis. The gloss values of bird’s eye did not present fibre direction dependence at all, however, the linear relationship between glossiness values measured with 60° and 85° incident angles could be observed. Some of the measured dots in GU60 –GU85 coordinate system for red tingle burl show some kind of scattering, but they did not create definite directional structure. Figure 2.27 presents the gloss ratios for finished walnut veneer with irregular growth used on a bed-clothes cabinet. The measured gloss values are high (both measured with 60° and 85°), but the gloss ratio is rather low, it is only 1.6. It was impossible to determine the real fibre direction at the place of measurement. The directions indicated on the figure are the fibre directions of the regular portion. All measured dots are located on the same straight line. This fact confirms that the irregular area has random fibre directions. All of figures above present that the straight trendline crosses the origin of the GU60 –GU85 coordinate system representing the proportional correlation. Wooden surfaces with irregular surface pattern also fulfil this directly proportional correlation.

82

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.26 Glossiness properties of red tingle (Eucalyptus jacksonii) and karri (Eucalyptus diversicolor) wood species with irregular growth. (Reproduced with the written permission from Springer Nature)

Fig. 2.27 Gloss properties of a walnut veneer with irregular growth. (Reproduced with the written permission from Springer Nature)

Figure 2.28 shows the gloss ratios of planed, sanded and polished ash surfaces. The volume density was 730 kg/m3 and the (Rpk + Rk ) roughness parameters were 22.7 and 8.2 μm for planned and polished samples, respectively. The gloss ratio for planed ash perpendicular to the grain is comparatively low due to the higher surface roughness. The same species in polished condition (sanded with P 400 paper) shows much higher gloss ratios. The high gloss ratio of polished ash perpendicular to the grain indicates a low surface roughness allowing light reflection under large

2.3 Gloss and Its Measurement

83

Fig. 2.28 Gloss ratios for planed, sanded and polished ash specimens. (Reproduced with the written permission from Springer Nature)

viewing angle (85°). Polished ash generates relatively high gloss values also with 60º measuring angle because of the low roughness value. The gloss values of sanded samples are similar in parallel and perpendicular directions showing that sanding makes “smeary” surface. The gloss properties of a surface made using Japanese planer provides remarkable results. This cutting method generates extremely smooth surface with excellent gloss properties providing a gentle lustre and vivid colour. Reflectance measurements showed that the reflection values decreased considerably in the 400–500 nm interval due to the clean cutting. This planing method produces less pressed cellular structures than the rotating planer. Figure 2.29 presents the gloss data of cherry samples planed with 15 μm chip thickness. The GU85 values are at least 50% higher compared to conventional planing. Results described above suggest that correlation may exist between the gloss ratios and the (Rpk + Rk ) roughness parameter both parallel and perpendicular to the grain. Using our measurements results, Fig. 2.30 shows the discovered correlation functions between the roughness parameter (Rpk + Rk ) and the gloss ratios parallel and perpendicular to the grain. The curved trendlines run almost parallel. The measured dots sit on the trendlines representing good correlation. This correlation is valid if the dominant effect influencing the reflection is originated from the surface structure of wood. Figure 2.31 presents the gloss ratios of an oak veneered cabinet in natural colour (having thin clear coating) manufactured in the 1970s. The gloss parallel to the grain varies in wide range corresponding to the silky mat grade. The GU85 /GU60 ratios are 2.5 and 1.2 parallel and perpendicular to the grain, respectively. These values are slightly higher that the same data of natural oak. The difference is due to the clear coating highly amplifying the gloss values measured with both 60° and 85° incident angles.

84

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.29 Gloss properties of cherry wood planed with Japanese planer

Fig. 2.30 Correlation functions between roughness parameter (Rpk + Rk ) and gloss ratios parallel and perpendicular to the grain. (Reproduced with the written permission from Springer Nature)

The gloss ratios of a walnut-veneered table stained to a middle brown colour are presented in Fig. 2.32. The gloss parallel to the grain varies in a broad range corresponding to the silky mat grade. The gloss perpendicular to the grain is changing in a narrower range and its ratio is somewhat higher than that would correspond to the correlation function. The GU85 /GU60 ratios are 2.2 and 1.5 parallel and perpendicular to the grain, respectively.

2.3 Gloss and Its Measurement

85

Fig. 2.31 Gloss ratios of an oak veneered cabinet. (Reproduced with the written permission from Springer Nature)

Fig. 2.32 Gloss ratios of a walnut-veneered table slightly stained to middle brown colour. (Reproduced with the written permission from Springer Nature)

The glossiness of wood surfaces depends much more on the surface roughness than on their colour hue. The porosity of wood influences the volume density and the roughness as well. The roughness of a surface determines the reflection properties together with glossiness. Therefore, it is interesting to examine a possible correlation between gloss ratio and volume density. Selected measurement results (glossiness

86

2 Measurement and Data Evaluation of Wood Colour and Gloss

Fig. 2.33 Gloss ratios for polished samples as a function of volume density. (Reproduced with the written permission from Springer Nature)

ratios and densities) of polished wood samples are presented in Fig. 2.33. The curved trendlines and the locations of the dots are representing that the measured values show correlation between the gloss ratio and volume density. Based on the theoretical considerations and experiments performed with widerange of wood species and types of furniture’s, the following main conclusions can be drawn: • Due to the specific structure of wooden surfaces, light reflection properties depend on many factors, and on a given surface area different reflection values can be measured. • Gloss properties of a wooden surface depend on the incident angle of the applied light. Therefore, unified approach and standardised measurement method is essential. • An incident and viewing angle of 60º is recommended for gloss measurement on wooden surfaces. • Determination of GU85 /GU60 ratios is recommended to get more detailed information. • The GU85 /GU60 ratios are generally constant, the scattering of gloss ratios depends on the homogeneity of surface structure. • There is a distinct difference in the gloss ratios parallel and perpendicular to the grain. • Local gloss properties on a given sample, i.e. early- and latewood, may differ considerably. Different treatments applied in the wood industry modify the gloss values of wooden surfaces. Dry thermal treatment above 160 °C is often used to improve some properties of wood. Heat treated wood demonstrates new characteristics such

2.3 Gloss and Its Measurement

87

as reduced water absorption, improved dimensional stability, better resistance to destruction by insects and microorganisms. Heat treatment may improve the aesthetic appearance of unattractive species. Thermal modification of wood also causes loss of glossiness in some extent. It was reported that gloss values of heat-treated wood decreased with increasing treatment temperature and treatment time (Aksoy et al. 2011; Korkut et al. 2013; Gurleyen et al. 2018; Lu et al. 2022). Korkut demonstrated that surface roughness values of the specimens decreased under all treatment conditions. Esteves et al. (2019) investigated the gloss change of black locust, wild pear, linden, alder and willow wood species applying heat treatment at 212 °C for 1 and 2 h. It was found that gloss values obtained using a measuring angle of 20° decreased due to heat treatment, in both perpendicular and parallel directions. Glossiness at 60° angle was similar to that at 20°. In case of 85° angle an equilibrium was found between the decrease in gloss due to the heat treatment and the increase due to the lower roughness of the heat-treated surface. Glossiness measured perpendicular and parallel to the grain at the angles 20°, 60° and 85° were found significantly different for the investigated wood species and for the applied heat treatments as well. Exotic wood samples were heat treated at 212 °C for 1- and 2-h and the gloss and colour change were monitored (Ayata et al. 2017). The results were similar to Esteves’s results. The densifying process reduces the surface roughness and generates glossiness increase. Thermo-mechanical densification improves the attractiveness of wood surfaces. Bekhta et al. (2014) densified alder, beech, birch and Scots pine veneers at 100, 150, and 200 °C temperature, 4, 8, and 12 MPa pressure for 4 min. It was found that short-term thermo-mechanical densification causes considerable glossiness increase, in comparison with non-densified veneer. The gloss of all investigated wood species increased significantly with increasing densification temperature and pressure. It is interesting to mention that gloss values of non-densified veneers obtained at angles of 20° and 85° are practically identical. In contrast, great differences can be found between the gloss values obtained at angles of 20° and 85° after densification. The maximum value of glossiness for Scots pine measured at 85° was 15 times higher than that measured at 20°. These results show clearly that the comparison of glossiness data measured at different incident angles is meaningless if the aim is to monitor the changes. This kind of comparison has only theoretical meaning. Weathering modifies the surface roughness of wood and as a consequence gloss values change as well. Arpaci et al. (2021) investigated outdoor weathering properties of sixteen wood species including European and exotic samples. Duration of the weathering was 393-day. It was found that anigre, ash, beech, cherry, eucalyptus, maple, mazel, oak, pine and tulip samples had a slight gloss loss while the surface of beli, limba, teak, sapele, walnut and American walnut samples showed slightly higher glossiness after weathering. It is interesting to mention that samples producing gloss increase have high extractive content. This high extractive content is partly able to protect the wood surface against photodegradation. One year long outdoor weathering test was done in Prague involving 9 European wood species (Oberhofnerova et al. 2017). Change of gloss mean values

88

2 Measurement and Data Evaluation of Wood Colour and Gloss

did not show any systematic trend during exposure. Generally, slight increase was observed in most cases after 6 months followed by a large decrease by the end of the 12-month natural weathering test. The cited two outdoor weathering tests present different results, representing that outdoor weathering is unable to produce repeatable outcomes because plenty of randomly influencing parameters determine the process. Basralocus, mabberley and teak specimens were exposed to sunlight to determine the effects of long-term (733 days) treatment (Liu et al. 2017). The sunlight irradiation generated only a little gloss decrease (only 1–2 GU). Mabberley showed the highest gloss change while teak was the most stable species. Esteves et al. (2020) studied the gloss properties of two heat-treated pine species (Pinus pinaster and Pinus radiata) before and after artificial weathering. Thermal treatment was done in accordance with the Thermowood process at 212 °C. The artificial exposure was generated by UV A lamp and the duration of exposure was 750 h. The untreated (natural) wood showed gloss decrease due to artificial weathering, while the glossiness of heattreated wood remained almost constant. Important result was that the measurement at 85° incident angle produced high data dispersion. Similar results were found by Baysal et al. (2014) after artificial weathering of thermally modified Scots pine. The artificial weathering induced gloss change of unmodified Scots pine was higher than that of thermally modified Scots pine. Based on the results described above, it can be stated that gloss is not a real material property (such as density) that has exact meaning alone. Gloss values are highly dependent on the measurement conditions. The quotient of two types of gloss values (e.g. GU85 /GU60 ) may be used as a material constant. (The quotient eliminates most of the influencing factors.) The measured data can be accepted as valid only if the measurement conditions are presented together with the measured data. The citation of a standard (e.g. ISO 2813) is not enough. It is meaningless to present the data measured at all 3 standardised angles if only the changes are presented. Gloss data measured at different angles are hardly comparable. This kind of comparison has only theoretical meaning and it is appropriate only for choosing the proper one to demonstrate the changes. Experiences prove that 60° is usually useful to monitor the glossiness of natural wood. Nevertheless, for small gloss values, the measurement using 85° angle is recommended, keeping in mind that this high angle may have some disadvantages for such inhomogeneous surfaces as wood, because the measured data have high standard deviations. The 20° incident angle is recommended only for finished wood surfaces.

References Aksoy, A., Deveci, M., Baysal, E., Toker, H.: Colour and gloss changes of Scots pine after heat modification. Wood Res. 56(3), 329–336 (2011) Arpaci, S.S., Tomak, E.D., Ermeydan, M.A., Yldirim, I.: Natural weathering of sixteen wood species: changes on surface properties. Polym. Degrad. Stab. 183, ID 109415 (2021). https:// doi.org/10.1016/j.polymdegradstab.2020.109415

References

89

Ayata, U., Gurleyen, L., Esteves, B.: Effect of heat treatment on the surface of selected exotic wood species. Drewno 60, 105–116 (2017) Badami, S., Moorkoth, S., Suresh, B.: Caesalpinia sappan, a medicinal and dye yielding plant. Nat. Prod. Radiance 3(2), 75–82 (2004). http://nopr.niscpr.res.in/handle/123456789/9400 Banadics, E.A., Gálos, B., Tolvaj, L.: A sötét egzóta faanyagok helyettesítése g˝ozölt akác faanyaggal. [Substitution of dark exotic wood species by steamed black locust.] Faipar 64(1), 22–28 (2016) Banadics, E.A.: A nyár faanyag g˝ozölési tulajdonságainak vizsgálata. [Investigation of the steaming properties for poplar wood.] Ph.D. thesis, University of Sopron, Hungary (2023) Baysal, E., Degirmentepe, S., Simsek, H.: Some surface properties of thermally modified Scots pine after artificial weathering. Maderas, Ciencia Tecnol. 16(3), 355–364 (2014) Bekhta, P., Porszyk, S., Lis, B., Krystofiak, T.: Gloss of thermally densified alder (Alnus glutinosa Goertn.), beech (Fagus sylvatica L.), birch (Betula verrucosa Ehrh.), and pine (Pinus sylvestris L.) wood veneers. Eur. J. Wood Prod. 72, 799–808 (2014). https://doi.org/10.1007/s00107-0140843-3 Boruvka, V., Sedivka, P., Novak, D., Holecek, T., Turek, J.: Haptic and aesthetic properties of heat-treated modified birch wood. Forest 12, ID 1081 (2021). https://doi.org/10.3390/f12081081 Budakçi, M., Korkmaz, M., Karal, I.: Antifungal effects of staining process on wood: hardness, gloss, and color change. BioResources 18(1), 302–316 (2023). https://doi.org/10.15376/biores. 18.1.302-316 Csanady, E., Magoss, E., Tolvaj, L.: Quality of Machined Wood Surfaces. Springer, Heidlberg New York London (2015). https://doi.org/10.1007/978-3-319-22419-0 Eckert, E.: Messung der Reflexion von Wärmestrahlen an technische Oberflächen. Forschung 7(6), 265–270 (1936) Esteves, B., Ayata, U., Gurleyen, L.: Effect of heat treatment on the colour and glossiness of black locust, wild pear, linden, alder and willow wood. Drewno-Wood 62, 39–52(2019). https://doi. org/10.12841/wood.1644-3985.267.10 Esteves, B., Herrera, R., Santos, J., Carvalho, L., Nunes, L., Ferreira, J., Domingos, I.J., Cruz-Lopes, L.: Artificial weathering of heat-treated pines fron the Iberian Penesula. BioResources 15(4), 9642–9655 (2020) Gurleyen, L., Esteves, B., Ayata, U., Gurleyen, T., Cinar, H.: The effects of heat treatment on colour and glossiness of some commercial woods in turkey. Drewno-Wood 61, ID 201 (2018). https:// doi.org/10.12841/wood.1644-3985.227.03 ISO 11664–4: Colorimetry-Part 4: CIE 1976 L*a*b* colour space (2008) Kannar, A., Tolvaj, L., Magoss, E.: Colour change of photodegraded spruce wood by water leaching. Wood Res. 63(6), 935–946 (2018) Kilinç, ˙I., Budakçi, M., Korkmaz, M.: The use of environmentally friendly abrasive blasting media for paint removal from wood surfaces. BioResources 18(1), 1185–1205 (2023). https://doi.org/ 10.15376/biores.18.1.1185-1205 Klement, I., Vilkovská, T.: Color characteristics of red false heartwood and mature wood of beech (Fagus sylvatica L.) determining by different chromacity coordinates. Sustainability 11(3), ID 690 (2019). https://doi.org/10.3390/su11030690 Klement, I., Vilkovsky, P., Vilkovská, T., Orlowski, K.A., Baranski, J., Chuchala, D., Suchta, A.: The influence of drying temperature on color change of hornbeam and maple wood used as surface and inner layers of wood composites. Appl. Sci. 11(22), ID 10673 (2021). https://doi. org/10.3390/app112210673 Korkut, D.S., Hiziroglu, S., Aytin, A.: Effect of heat treatment on surface characteristics of wild cherry wood. BioResources 8(2), 1582–1590 (2013) Liu, R., Pang, X., Yang, Z.: Measurement of three wood materials against weathering during long natural sunlight exposure. Measurement 102, 179–185 (2017). https://doi.org/10.1016/j.measur ement.2017.01.034 Lübbe, E.: Colours in the Mind—Colour Systems in Reality—A Formula for Colour Saturation. Muster-Schmidt Verlag, Zurich (2010). ISBN 978-3-7881-4057-1

90

2 Measurement and Data Evaluation of Wood Colour and Gloss

Lu, C., Liu, Y., Jiang, H., LU, Q.: Impact of heat treatment on the surface color and glossiness of young eucalyptus wood. Wood Res. 67(3), 348–360 (2022). https://doi.org/10.37763/wr.13364561/67.3.348360 Méndez-Mejías, L.D., Moya, R.: Effect of thermo-treatment on the physical and mechanical, color, fungal durability of wood of Tectona grandis and Gmelina arborea from Forest Plantations. Mat. Sci. (medziagotyra) 24(1), 59–68 (2018). https://doi.org/10.5755/j01.ms.24.1.17545 Mills, J., White, R.: The Organic Chemistry of Museum Objects. Routledge, London (2011) Oberhofnerova, E., Panek, M., Garcia-Cimarras, A.: The effect of natural weathering on untreated wood surface. Maderas, Ciencia Tecnol. 19(2), 173–184 (2017). https://doi.org/10.4067/S0718221X2017005000015 Schmidt, E., Eckert, E.: Über die Richtungsverteilung der Wärmestrahlung von Oberflächen. Forsch. Gebiet Ing. – Wes. 6, 175–183 (1935) Sitkei, G.: Further studies on the characterization of wood colours. Progress report No. 7. Department of Wood Engineering, University of Sopron, Hungary (2020). ISBN 978-963-359-100-0 Straze, A., Fajdiga, G., Gospodaric, B.: Nondestructive characterization of dry heat-treated fir (Abies alba Mill.) timber in view of possible structural use. Forests 9(129), ID 776 (2018). https://doi. org/10.3390/f9120776 Tolvaj, L., Nemeth, K.: Correlation between hue-angle and colour lightness of steamed black locust wood. Acta Silv. Lign. Hung. 4, 55–59 (2008) Tolvaj, L., Mitsui, K.: Correlation between hue angle and lightness of light irradiated wood. Polym. Degrad. Stab. 95, 638–642 (2010). https://doi.org/10.1016/j.polymdegradstab.2009.12.00 Tolvaj, L., Papp, G., Varga, D., Lang, E.: Effect of steaming on the colour change of softwoods. BioResources 7(3), 2799–2808 (2012) Tolvaj, L., Persze, L., Lang, E.: Correlation between hue angle and lightness of wood species grown in Hungary. Wood Res. 58, 141–145 (2013) Tolvaj, L., Tsuchikawa, S., Inagaki, T., Varga, D.: Combined effects of UV light and elevated temperatures on wood discoloration. Wood Sci. Technol. 49, 1225–1237 (2015). https://doi.org/ 10.1007/s00226-015-0749-1 Zhu, T., Sheng, J., Chen, J., Ren, K., Wu, Z., Wu, H., Li, J., Lin, J.: Staining of wood veneers with anti-UV property using the natural dye extracted from Dalbergia cohinchinensis. J. Cleaner Prod. 284, ID 124770 (2021). https://doi.org/10.1016/j.jclepro.2020.124770

Chapter 3

Applications of Colour Measurement in Wood Research

Abstract The chapter discusses the applications of colour measurement in wood research dealing with the colour modification effects of dry thermal treatment, steaming and surface wetting. The experimental results demonstrate that the temperature dependence of a* and b* colour coordinate values is exponential verified by the Arrhenius law. The chapter contains detailed information regarding the steaming properties of black locust, beech, poplar, Turkey oak, Scots pine, larch, spruce and sugi wood species. The presented figures give the possibility to determine the proper steaming parameters (steaming time and temperature) to create the desired wood colour using this technology. Colour homogenisation effect of steaming is presented for those species that have inhomogeneous natural colour. The presented experimental results show that wetting and lacquering result in intensive darkening and saturation increase on wooden surfaces. Because of this phenomenon, colour measurement of wood material requires attention to guarantee equal moisture content of the samples during the applied treatment period. Keywords Wood · Thermal treatments · Colour modification · Steaming · Wetting · Colour homogenisation · Extractives

3.1 Introduction Colour is the most important aesthetic feature of wood. Extremely small differences in colour are visible if two objects are side by side. The differences are less obvious if the surfaces are textured. Colorimeters always measure average colour parameters of the illuminated surface. Decreasing the diameter of the aperture allows to determine the colour texture of a specific wooden surface. Colour measurement is an extremely sensitive method for monitoring the changes during treatment. As colour parameters are highly complex, the chemical changes that generate the alterations cannot be precisely identified. The colour of different types of wood strongly depends on the species. Some of them have a uniform greyish-white appearance, while others have a

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 L. Tolvaj, Optical Properties of Wood, Smart Sensors, Measurement and Instrumentation 45, https://doi.org/10.1007/978-3-031-46906-0_3

91

92

3 Applications of Colour Measurement in Wood Research

highly inhomogeneous texture. According to our experiences, ten randomly chosen measuring spots on wooden surface are enough to get repeatable colour data. Thermal effects modify the colour of wood considerably. The rate of colour change is strongly temperature dependent and is accelerated by the presence of water. Thermal treatment of wood is the most advanced commercial modification technology not only for colour alteration, but also for improving dimensional stability and increasing decay resistance. Tieman (1915) reported first that the elevated temperature modifies the physical properties of wood mainly by reducing the equilibrium moisture content. Recently, several commercial processes have been developed to the large-scale production of thermally treated timber. In terms of colour modification, steaming is a more preferable process than dry thermal treatment at high temperature because the presence of water enhances the effect of thermal treatment even at much lower temperatures.

3.2 Colour Modification by Dry Thermal Treatment Extractives are the thermally most sensitive molecules in wood. Among the main chemical components, hemicelluloses are the most susceptible and can undergo thermal degradation above 100 °C. The intensity of a thermally activated process increases usually exponentially by rising temperature according to the Arrhenius law. The Arrhenius Eq. (3.1) relates the rate of a chemical reaction k to temperature T and it includes the activation energy (E a ). The equation is simple if the activation energy is constant. The Arrhenius equation can be given in the form: k = Ae−Ea /(RT )

(3.1)

where R is the universal gas constant. The former form can be written equivalently taking natural logarithm as follows: ln k =

−E a 1 + ln A R T

An Arrhenius plot displays the logarithm of a kinetic constant (k, ordinate) plotted against inverse temperature (1/T, abscissa). Arrhenius plots are often used to analyse the effect of temperature on the rates of chemical reactions. For a single rate-limited thermally activated process, an Arrhenius plot gives a straight line, from which the activation energy (E a ) and the pre-exponential factor (A) can both be determined. The colour change of wooden surfaces at room temperature is a slow process. Nevertheless, an indoor wooden structure will darken in a few years, even in total darkness. Black locust heartwood (Robinia pseudoacacia L), poplar (Populus x euramericana cv. Pannonia), larch heartwood (Larix decidua L.), Scots pine heartwood (Pinus sylvestris L.) and spruce (Picea abies Mill.) samples were prepared to investigate the

3.2 Colour Modification by Dry Thermal Treatment

93

colour changes generated by dry thermal treatment at 90, 110 and 130 °C. Temperatures less than 90 °C generates extremely slow colour change. Therefore, this temperature value was chosen as the lower limit. The effects of temperatures above 130 °C will be discussed later on separately, because of the rapid colour change. Long treatment time (18 days) was chosen because of the slow process at 90 °C. Colour parameters were measured on the radial surface of the samples to get the average colour of earlywood and latewood. The results are presented here according to a recently published paper (Preklet et al. 2023) with written permission of Acta Silvatica et Lignaria Hungarica. Colour parameters are derived from the reflectance spectrum thus the change of reflectance spectrum represents the colour change. Changes of the reflectance spectra are represented here by spruce wood thermally treated for 11 days at different temperatures (Fig. 3.1). The reflectance spectrum of natural spruce samples contains close to linearly decreasing values in the 700–440 nm wavelength interval. This tendency is followed by a rapid decrease in the 440–400 nm wavelength interval. The reflectance value at 700 nm is rather high (87%) and it is close to 50% at 450 nm as well. The high reflectance values almost in the whole visible region generates white colour. High reflectance values in the red colour region and relatively low reflectance values in the blue range result in warm white colour. The 11-day treatment at 90 °C generated only a small reflectance intensity decrease. The result of this decrease was a slight darkening. The decrease was more pronounced in the blue-green region than in the yellow–red region, producing slight colour shift towards brown tint. Maximum reflectance decrease was at a wavelength of around 450 nm. Dry thermal treatment at 110 °C induced more intensive reflectance decrease than that at 90 °C. The result was an intensive darkening. The greatest reflectance difference evolved at a wavelength of around 500 nm. This change is

Fig. 3.1 Reflectance spectra of spruce generated by 11-day dry thermal treatment at different temperatures

94

3 Applications of Colour Measurement in Wood Research

twice as much as the change at 700 nm. All these alteration differences pushed the hue towards brown. Thermal treatment at 130 °C created similar but more intensive reflectance decreases than the treatment at 110 °C. Plotting the L*, a* and b* colour coordinates as a function of the treatment time could provide more detailed information regarding the colour alteration than the analysis of reflectance value changes. Figures 3.2 and 3.3 show the lightness alterations of certain investigated wood species. Dry thermal treatment at 90 °C generated almost no lightness decrease except for black locust where the total lightness decrease was 10 units at the 18th day of the treatment. Elevated temperatures amplified the lightness decreases. The investigated conifer species presented similar lightness decreases. Trendlines belonging to the same treatment temperature run close to parallel. Only spruce presented a little deviation as its lightness decrease was less intensive during the first 3 days of the treatment compared to the other two conifers. Poplar samples showed somewhat different lightness decrease character that the other species. All tree trendlines of poplar are close to straight lines and the measured lightness values generated by 90 and 110 °C are close to each other showing very little lightness decrease. In contrast, the treatment at 130 °C produced intensive lightness decrease. Poplar has low extractive content. Consequently, the darkening of the samples was generated mainly by the degradation products of hemicelluloses. 130 °C is high enough to degrade wood hemicelluloses. The missing intensive change at the beginning of the treatment procedure is due to the low extractive content of poplar and the same is valid partly for spruce as well. Black locust showed the most rapid lightness decrease during the first 3 days of treatment at all temperatures compared to the other investigated species. It is not surprising because black locust has the highest extractive content among the investigated species. Results show that the extractives are the most sensitive chemical components in wood during thermal treatment, followed by the hemicelluloses.

Fig. 3.2 Lightness change of black locust and poplar samples during dry thermal treatment

3.2 Colour Modification by Dry Thermal Treatment

95

Fig. 3.3 Lightness change of conifers during dry thermal treatment

When investigating colour change procedures, a* and b* colour coordinates provide more detailed information than L* alone thus allow more subtle conclusions because the calculation of a* and b* use both X and Z tristimulus values beside Y. (Calculation of L* incorporates Y only.) Furthermore, it is important to know that the calculation of Y encompasses the reflection values of almost the whole visible wavelength interval. In contrast, determination of X utilizes two wavelength intervals around 600 and 450 nm, and the calculation of Z utilizes only one narrower wavelength interval around 450 nm. These properties of a* and b* show that they are more sensitive to the wavelength dependence of the reflectance changes than L*. Figures 3.4 and 3.5 present the redness change of the investigated species generated by dry thermal treatment. The initial redness values of the investigated species were for poplar, spruce, black locust, Scots pine and larch 2.7, 4.5, 5.9, 6.4 and 11.6, respectively. This order represents the extractive content values responsible for the redness of the species. Initial redness value of larch was more than twice as high as the initial a* values of the other species. The unequalled high durability of larch is lying in its special extractives responsible also for its high redness value. Gierlinger et al (2004) investigated different larch species having large distribution of a* values (between 5.6 and 9) and found that the redness values were strongly correlated with phenolics content. Heating at 90 °C caused only a little redness increase during the 18-day treatment. There were two exceptions. The redness value of larch did not change at all while black locust produced considerable redness value increase (59%) even at this temperature. Extractives in black locust responsible for redness increase are highly sensitive to thermal treatment. Treatment at 130 °C showed that the heat generated degradation products of extractives are chemically not stable enough. These degradation products underwent secondary degradation and reduced the redness values after 6-day treatment of black locust and after 11-day treatment of larch. Larch seems to

96

3 Applications of Colour Measurement in Wood Research

Fig. 3.4 Redness change of black locust and poplar samples during dry thermal treatment

Fig. 3.5 Redness change of conifers during dry thermal treatment

be the thermally most stable species among the investigated ones. It has the highest initial redness value (12 units) and the maximum of redness increase was only 24%. Spruce and Scots pine produced similar redness alterations. The redness change course of poplar was completely different to the other species. The intensive increase during the first 3 days of treatment (comparing to the whole change) was completely missing, the trendlines were close to linear. The redness increase was slow at 90 and 110 °C while at 130 °C a quite intensive and enormously high (323%) increase could be observed during the 18-day treatment. Similar data for spruce, Scots pine, black locust and larch are 161, 106, 73 and 21%, respectively. The results show

3.2 Colour Modification by Dry Thermal Treatment

97

that the degradation of hemicelluloses was dominant at 130 °C and it generated the main portion of redness increase. It seems that the extractive content could partly protect the hemicelluloses against thermal degradation. Black locust and larch with the highest extractive content showed the smallest red hue increase while poplar featuring the smallest extractive content produced the greatest red hue increase. In terms of protective effect, the type of the extractives can also play an important role. These assumptions need further chemical investigations. Figures 3.6–3.7 show the yellowness change of the investigated species induced by different dry thermal treatments. The initial values of the b* co-ordinates are 17.4, 22.3, 26.6, 27.7 and 28.1 for poplar, spruce, larch, black locust and Scots pine, respectively. The initial yellowness values were higher than the initial redness values, and the dispersion of the colour co-ordinates was smaller for yellowness than for redness. Most of species showed yellowness increase due to the thermal treatments. The only exception was black locust. Its yellowness decreased during the applied thermal treatments. The high robinetin content of black locust (covering 2–5% of its total mass) causes its unattractive yellow colour. The robinetin type extractives are highly sensitive to thermal treatment. Heat induced degradation of these extractives led to a reduce in the yellowness of black locust even at 90 °C and caused an extremely rapid decrease during the first day already at 130 °C. The investigated species have two types of extractives that are responsible for yellow colour. One of them generated yellowness increases while the other type produced yellowness decrease during thermal treatment. The second type of extractives were found in black locust and in larch. Larch samples produced mainly yellowness decrease at 130 °C similarly to black locust. The final yellowness value at the end of the 130 °C treatment was considerably smaller than the initial one meaning that substantial part of the decrease was induced by the degradation of extractives being

Fig. 3.6 Yellowness change of black locust and poplar samples during dry thermal treatment

98

3 Applications of Colour Measurement in Wood Research

Fig. 3.7 Yellowness change of conifers during dry thermal treatment

originally in larch wood. Treatment at 90 °C generated slow but continuous yellowness change in all investigated species. These changes were relatively fast during the first 3 days of all treatments, then the shift slowed down afterward showing that extractives were involved in the degradation processes of all species. Treatment at 110 °C resulted in considerably greater redness change than that at 90 °C. Conifers produced both yellowness increase and decrease during the treatment at 130 °C. Maximum b* values were reached on the 11th, 6th and 1st day of the treatment of spruce, Scots pine and larch, respectively. This phenomenon shows that the heat generated degradation products underwent secondary degradation at 130 °C. Poplar and spruce (having the smallest extractive content) presented the greatest yellowness increase on the 11th day of the treatment. These increases were 73 and 42%. These data show that hemicelluloses play an important role in the yellowness change during dry thermal treatment. Figures 3.8–3.9 presents the locations of colour dots of black locust, poplar, larch, spruce and Scots pine species generated by dry treatments at 90, 110 and 130 °C. The left end of the trendlines represents the colour dots of untreated samples followed by the dots of treated samples with increasing treatment time. These figures show the change of hue and of chroma. Black locust is a great exception. It suffered great hue value decrease from 78° to 55° at 130 °C changing the greyish-yellow colour to brown tint. The treatment at 90 °C also caused considerable hue value decrease for black locust comparing to the other species. The other species hardly changed their hue at this temperature. Treatment at 110 °C produced moderate chroma increase and small hue decrease. Exception was the hue decrease of black locust. This decrease was 18 units. The chroma values (distance between the colour dot and the origin of the coordinate system) of black locust slightly increased by dry thermal treatment only. The behaviour of larch was partly exceptional as well. It showed small chroma increase and moderate hue value decrease. The other species (poplar, spruce and

3.2 Colour Modification by Dry Thermal Treatment

99

Scots pine) produced great chroma increase and moderate hue value decrease. The maximum chroma increase of poplar and spruce were 16 and 12 units generated by thermal treatment at 130 °C.

Fig. 3.8 Colour dots of black locust and poplar samples on the a*–b* plane during dry thermal treatments. (Dots at the left end of the lines represent the untreated samples.) Constant hue (h*) lines are also indicated

Fig. 3.9 Colour dots of conifer samples on the a*–b* plane during dry thermal treatments. (Dots at the left end of the lines represent the untreated samples.) Constant hue (h*) lines are also indicated

100

3 Applications of Colour Measurement in Wood Research

The results show that dry thermal treatment below 100 °C causes slow colour alteration. Similar results were found by Liu et al. (2016) at 100 °C in semidry condition (air RH = 55%). Above this limit the colour change is increasing rapidly with increasing treatment temperature. Both extractives and hemicelluloses play important role in colour alteration during dry thermal treatment. The investigated temperature range (90–130 °C) has only theoretical importance because of the long treatment time. Higher temperatures (above 150 °C) produce rapid colour change. These treatments have industrial importance not only for the quick and intensive colour change but because of some positive property change of wood. The industrially used temperature interval is 160–260 °C. Stamm and Hansen (1937) heated wood with different gases and reported the decrease of equilibrium moisture content, swelling and shrinkage values. During the next almost 100 years plenty of papers reported more and more results regarding dry thermal treatments. A review paper (Esteves and Pereira 2009) summarised the results of 160 papers. Also, there are several patented processes to produce thermally treated wood material. The most successful in Europe is the ThermoWood patented by Viitaniemi et al. (1997). This study discusses solely the colour change aspects of thermal treatments. One of the main advantages of dry thermal treatment is that it alters wood colour in its whole cross section without using any harmful chemicals. This thermal process can also be applied to modify the colour of wood species having unattractive or highly inhomogeneous initial colour. Moreover, thermally modified dark wood may substitute tropical wood species as well. There are two distinct groups of thermal treatments at high temperature based on the presence of oxygen. One group of treatments is where the heating medium contains oxygen while procedures in the other group exclude oxygen. Chemical reactions are quite different within these two groups since the presence of oxygen allows oxidation. Heat treatments create free radicals which can then react with oxygen forming oxidation products such as quinines (Bekhta and Niemz 2003). Such oxidative reactions are inherently impossible in the absence of oxygen. Chen et al. (2012) studied the colour change of black locust flour caused by thermal treatment in oxygen and nitrogen atmospheric dry and wet conditions. Results showed that the samples suffered greater colour change in terms of all colour parameters (L*, a*, b*) in the presence of oxygen than in nitrogen atmosphere during dry thermal treatment at 120 °C. The decrease in lightness value was, for example, twofold in the presence of oxygen. The most common option to exclude oxygen is the application of oils as heating medium. A review article (Lee et al. 2018) presents 139 papers dealing with the thermal treatment of wood in vegetable oils. Most of the experiments in the literature were carried out in the presence of oxygen (Bekhta and Niemz 2003; Nuopponen et al. 2003; Sundqvist et al. 2006; Windeisen et al.2007; Kaciková et al. 2013; Kamperidou et al. 2013, Zanuncio et al. 2015, Mikleˇci´c and Jirouš-Rajkovi´c 2016; Griebeler et al. 2018; Sikora et al. 2018; Gasparik et al. 2019; Lo Monaco et al. 2020). The colour modification effect of dry thermal treatment is presented here applying wide range of temperatures (90–200 °C). Duration of the heat-up period to reach the

3.2 Colour Modification by Dry Thermal Treatment

101

desired treatment temperature was 10, 15, 20 and 30 min at 90, 130, 160 and 200 °C, respectively. (There was no cooling time. Samples were removed from the chamber right after finishing the treatment.) Test samples for the different temperatures were cut from the same board to minimize the effect of inhomogeneity. The sample size was 100 × 20 × 10 (mm3 ). Deciduous and conifer species with different extractive content were involved in the test to determine the temperature dependence of dry thermal treatment. Black locust heartwood (Robinia pseudoacacia L.) was chosen because of its high extractive content. On the contrary, Poplar (Populus x euramericana cv. Pannonia) was tested due to its low extractive content. Larch heartwood (Larix decidua L.), Scots pine heartwood (Pinus sylvestris L.) and spruce (Picea abies Mill.) were chosen because of their high, medium and low extractive content, respectively. Results presented above show that the colour change generated by thermal treatment is highly temperature dependent. Longer treatment times are required at lower temperatures. Below 150 °C, several days are necessary to obtain a substantial colour change. The treatment time above this temperature limit is rather short, some hours are enough to induce remarkable colour change. For correct comparison, the chosen treatment time was 6 h at all applied temperatures in the present investigation. Colour parameters were measured on the radial surface of the samples to determine the average colour of earlywood and latewood. The lightness change data are presented in Fig. 3.10. Lightness values decreased for all investigated species at all treatment temperatures. The decrease at 90 °C was negligible (almost zero) for all species during the 6-h treatment. At this temperature, the greatest lightness decrease was produced by black locust (0.7 unit). As for the next temperature stage, the treatment at 130 °C induced a relatively large lightness decrease (14 units) for black locust. In contrast, this data was only 0.2 unit for poplar. The elevated temperature (160 °C) doubled the lightness change of black locust compared to the effect of 130 °C. The same lightness decrease was 3, 6, 7 and 54-fold for Scots pine, spruce, larch and poplar, respectively. Comparing the behaviour of black locust and poplar, a highly different nature of changes can be observed. The dominant change of back locust was induced by its extractives. On the contrary, poplar hardly has extractives, thus its darkening was mainly generated by the degradation products of hemicelluloses. Dry thermal treatment at 200 °C generated almost the same lightness decrease for all investigated species showing that the degradation of hemicelluloses is the dominant alteration at this high temperature. Researchers demonstrated that most of the extractives disappeared during thermal treatment at such high temperatures. The thermal stability of extractives in Scots pine was studied by Nuopponen et al. (2003). It was found that fats and waxes were not detectable in the sapwood edges above 160 °C. At temperatures above 200 °C all resin acids disappeared from the wood. Figure 3.11 shows the redness change of the species generated by 6-h dry thermal treatment at various temperatures. The 6-h treatment caused very little redness change at 90 °C. Only larch presented a slight redness decrease. It is important to mention that the redness of larch remained intact during 18-day treatment at 90 °C (see Fig. 3.5). Larch showed very little redness increase at 130 °C as well. Black locust suffered the greatest redness increase at all applied temperatures except 200 °C because of the

102

3 Applications of Colour Measurement in Wood Research

Fig. 3.10 Lightness change of different wood species generated by 6-h dry thermal treatment at different temperatures

high thermal sensitivity of its extractives. The 6-h treatment at 200 °C was too long for black locust. Its redness value started decreasing after 4-h treatment. The other species showed small redness increase at 130 °C and medium increase at 160 °C. The only exception was poplar. Its redness increase was only 1.2 units at 160 °C. These results confirm that the degradation of hemicelluloses is a relatively slow but continuous process. In contrast, the degradation of extractives is rapid at the beginning of thermal treatment and slows down afterward. Obviously, 6 h at 160 °C was too short to induce such a strong modification of hemicelluloses that would be able to lead to a considerable redness increase. However, this short period was enough for the alteration of extractives to generate considerable redness increase in the other investigated species (except poplar). The treatment at 200 °C produced great redness increase for all species. The rate of the redness change was more species dependent than in the case of lightness change. Figure 3.12 shows the yellowness change of the species generated by 6-h dry thermal treatment at various temperatures. Black locust is an exception in terms of yellowness change as well; b* colour coordinate of all investigated species increased except that of black locust. Although, the yellowness decreases of black locust showed similar temperature dependence than that of the other species but in the opposite direction. This behaviour can be interpreted by the high robinetin type extractive content of black locust. These extractives are responsible for the greyishyellow initial colour of black locust. Robinetin type extractives are highly sensitive to thermal degradation and this degradation reduces the yellowness value considerably. The effect of the treatment at 90 °C seems to be neglectable in terms of b* shift similarly to the redness change, for all species. Yellowness change at high temperatures (above 130 °C) was highly species dependent. This phenomenon shows the

3.2 Colour Modification by Dry Thermal Treatment

103

Fig. 3.11 Redness change of different wood species generated by 6-h dry thermal treatment at different temperatures

Fig. 3.12 Yellowness change of different wood species generated by 6-h dry thermal treatment at different temperatures

complexity of the yellowness change. Redistribution and degradation of lignin may cause yellowness change beside the degradation of extractives and hemicelluloses. This kind of lignin related, high temperature induced processes were published in many papers (Tjeerdsma et al. 1998, Wikberg and Maunu 2004, Boonstara and Tjeerdsma 2006; Esteves and Pereira 2009; Kaciková et al. 2013; Esteves et al. 2013; Sikora et al. 2018). As it was introduced in Fig. 3.11, spruce and Scots pine presented similar redness change at all temperatures. Yellowness change of these species was

104

3 Applications of Colour Measurement in Wood Research

also similar except at 200 °C. Spruce produced the greatest yellowness increase at 200 °C; it was almost twice as high as the yellowness increases of Scots pine. This huge difference was probably generated by the different degradation properties of lignin within these species. Figures 3.13 and 3.14 present the Arrhenius plots of redness values of black locust, poplar, larch, spruce and Scots pine species thermally treated for 6 h at different temperatures. The Arrhenius law declares that if the Arrhenius plot is a straight line, the temperature dependence of the studied process is exponential. In our case, all Arrhenius plots are straight lines presenting that both redness and yellowness changes are exponential functions of the dry thermal treatment temperature. The coefficients of determination values are quite high for all investigated wood species. R2 values are above 0.9 for redness and above 0.92 for yellowness. The slope of the trendline for yellowness of black locust is opposite of that of the other lines showing that the changing tendency is also opposite in this case. Some colour dots are relatively fare from the trendline. The reason is laying in the colour inhomogeneity of wood. It is important to mention that all presented dots in the figures belong to different samples having slightly different initial colour. The other inhomogeneity problem is that the radial surface of the specimens was used for colour measurement. It is impossible to guarantee that all individual colour measurement covers the same earlywood-latewood ratio within the measured area. Arrhenius plots of the lightness values do not determine exactly straight lines. Fitting straight lines on the lightness dots the coefficient of determination values are much smaller (between 0.6 and 0.8) than in the case of redness and yellowness suggesting that the lightness change is determined by multiple chemical alterations generating absorption in the visible light region and the individual chemical changes are marked with different temperature dependence.

Fig. 3.13 Arrhenius plots of redness values for black locust, poplar, larch, spruce and Scots pine species thermally treated for 6 h at different temperatures

3.3 Steaming as a Colour Modification Process for Wood

105

Fig. 3.14 Arrhenius plots of yellowness values for black locust, poplar, larch, spruce and Scots pine species thermally treated for 6 h at different temperatures

3.3 Steaming as a Colour Modification Process for Wood Steaming is a thermal treatment where the presence of water molecules amplifies the effect of heat. Mitsui et al. (2004) demonstrated that the increasing air humidity amplifies the darkening of sugi samples during thermal treatment at 90 °C. Steam boosts the oxidizing and decomposing process of extractives. The reflectance spectrum of a material determines its colour. The reflectance spectrum of untreated black locust is presented in Fig. 3.15 together with the reflectance spectra after 36-day dry and wet thermal treatments at 90 °C. (Long thermal treatment duration was chosen because of the slow changes at 90 °C under dry condition.) Black locust presents usual wooden reflectance spectrum, which is almost straight line. Dry treatment reduced the reflection values considerably across the whole visible wavelength interval resulting in the drop of the lightness value. The greatest decrease can be observed in the middle of the spectrum. The reflectance of red region became more dominant compared to the spectrum of natural black locust. This change shifted the colour towards red, though the final colour was brown. The treatment under wet condition (steaming) produced even greater reflectance decrease than dry treatment, but the reflection in the red region remained dominant. The result was dark brown colour. Intersection wavelengths λ0 are 547, 578 and 582 nm for the natural, dry heat treated and steamed samples, respectively. Increasing values prove the red hue shift. In general, steaming increases the chroma values by minimalizing the reflected wavelengths. A more saturated colour is a result of reflection in a small wavelength interval. It is important to mention that steaming always occurs under saturated steam condition. Saturated steam evolves in a closed chamber or pot after a relatively short heat-up period provided that the vessel contains enough liquid water as well. The

106

3 Applications of Colour Measurement in Wood Research

Fig. 3.15 Reflectance spectrum of natural (untreated), and thermally treated black locust under dry and wet (steaming) condition at 90 °C. Treatment duration was 36 days. Intersection wavelengths (λ0 ) are also presented

amount of water in the air will increase with increasing temperature if there is enough liquid water in the vessel. Steaming chambers always must contain water under the wood material to guarantee saturated steam conditions which is of high importance from technological point of view. If the steam is not saturated in the chamber, the air will remove water from the wood resulting in rapid drying of the wood surface which then will generate cracks mainly in the cross section. The presence of saturated steam during steaming is highly important to get proper colour modification. Figure 3.16 presents the effect of water during the thermal treatment of poplar (Populus x euramericana cv. Pannonia) at 110 °C. The alteration of all three colour coordinates (L*, a*, b*) are presented. Dotted lines show the changes generated by dry thermal treatment while the solid lines display the effects of steaming. It is well visible that the presence of water amplifies the effect of thermal treatment. The lightness shift caused by the steaming was 5.2 times greater than the effect of dry thermal treatment after 18-day treatment. The same ratio for redness and yellowness change was 5.1 and 3.4. The presence of saturated steam in the air does not modify the pressure in a closed pot up to 100 °C. Above this temperature, the partial pressure of steam is added to the pressure of air. It means that a steaming chamber or pot must keep overpressure above 100 °C. Usually, cylindric chambers (autoclave) are used for industrial purposes. Internal temperature determines the partial pressure of saturated steam exactly, which is increasing rapidly with increasing temperature. Consequently, absolute pressure in a steaming autoclave will be as high as 1.5, 2.0, 3.0, 6.5 and 16.0 bar at 110, 120, 130, 160 and 200 °C, respectively. The evolving high steam pressure is one reason why steaming temperature above 120 °C is not applied in wood industry. On the other hand, if the pressure is lower than the required one, the steam will not be saturated, only overheated. Unfortunately, in some publications it is not clear and therefore very

3.3 Steaming as a Colour Modification Process for Wood

107

Fig. 3.16 Colour alterations of poplar wood generated by pure thermal treatment under dry condition at 110 °C and by steaming at 110 °C

confusing if the applied steam is saturated or only overheated. Overheated steam can be used for rapid drying of wood material under special conditions. The release of pressure needs special care at the end of steaming. The water in the autoclave is overheated if the pressure is above 1.0 bar. The overheated water starts explosive boiling if the pressure is reduced and it may cause serious accident. The best technology is to cool the water below 100 °C than to release the pressure slowly afterward. Industrial steaming of wood is mainly applied with the aim of colour modification. The recommended maximum temperature is 120 °C. The colour change is extremely fast above this temperature and it is not possible to generate equal temperature distribution within the wood bords in the steaming chamber. The consequence will be inhomogeneous colour on one hand. The inhomogeneous temperature distribution on the other hand generates thermal stresses as well resulting in many microcracks within the wood material. The softening effect of steaming is applied also in wood veneer manufacturing before slicing or peeling the trunk of a tree to prevent the wood from tearing. Steaming reduces the equilibrium moisture content, brings the timber log to a uniform moisture level as well as modifies its mechanical properties. Here, the colour modification effect of steaming will only be discussed. There are two main reasons of wood colour modification. Some wood species have unattractive or neutrally greyish-white colour. Other species have highly inhomogeneous colour resulting in an unfavourable texture especially in case of larger surfaces. Both disadvantages can be reduced by steaming (Varga and van der Zee 2008; Dzurenda 2018). The colour of wood is mainly determined by its extractive content therefore extractives play the main role also during steaming. As there are

108

3 Applications of Colour Measurement in Wood Research

numerous types of extractives in the individual wood species, steaming properties differ species by species. Steaming processes presented in this section were always carried out in a steaming vat with an internal relative humidity of 100%. Prior to and after treatments, specimens were kept under normal laboratory conditions (i.e., 65% RH and T = 21 °C) for a month to ensure equal moisture content of the samples for colour measurement.

3.3.1 Steaming Properties of Black Locust Back locust is originally native to North America but has now spread widely in Europe, China and South Korea. Strength and elastic properties of this hardwood species are better than those of any other species grown in Central Europe. Thanks to its durability, black locust is currently one of the most important raw materials for outdoor furniture. It is also the most frequently steamed species due its unattractive greyish-yellow and highly inhomogeneous colour. Colour change of black locust is highly sensitive to steaming temperature which is not surprising if we consider its high extractive content covering 5–9% of its total mass. The main extractive component is dihydrorobinetin covering 2–5% of the total wood mass (Molnar and Bariska 2002; Sanz et al. 2011). Robinetin type extractives impart the yellow colour for black locust. Steaming properties of black locust are discussed here according to a previous paper (Tolvaj et al. 2010, with written permission of Wood Research). To determine the steaming properties, a series of heartwood samples were steamed at various temperatures with 5 °C steps starting from 75 °C up to 130 °C. (Here, the effect of every second chosen temperature will be presented.) The moisture content of the samples was 25–30% before steaming. Maximum steaming time was 22 days to monitor all of the changes as detailed as possible. At 120 and 130 °C maximum steaming time was only 6 days as there were no real colour changes above this time limit. The effect of chemical changes induced by steaming can be observed with naked eye after a few hours’ treatment. Specimens become visibly darker and the colour changes from the unattractive greenish-yellow to a relatively more pleasant reddish hue. The visual observations described above can be confirmed by objective colour measurement. The lightness change of the samples is presented in Fig. 3.17. The lightness decreased continuously at all applied steaming temperatures and this decrease was more pronounced with increasing temperature. The lightness of black locust diminished proportionally more intensively in the first days of the treatment and levelled off more or less after around 1–10 days depending on the steaming temperature. It can be concluded, that the effective steaming time below 100 °C is 6–10 days, while above 100 °C it is 1–4 days from the viewpoint of lightness change. There is a lower limit of generated lightness value independently of the steaming temperature. As a result of the high sensitivity of the colour change of black locust to the steaming temperature, small (1–2 °C) temperature alteration can result in visible colour difference. For a

3.3 Steaming as a Colour Modification Process for Wood

109

Fig. 3.17 Lightness change of steamed black locust wood as function of steaming time and temperature

better understanding of this temperature sensitivity, lightness values generated by one day steaming are marked in Fig. 3.17. The dotted vertical line demonstrates the large lightness differences produced by the individual steaming temperatures above 100 °C. This difference is more than 13 units between the lightness values generated by110 and 120 °C. Similar large differences can be observed for temperatures below 100 °C after 6–9 days steaming. High steaming temperature sensitivity of black locust might cause difficulties during industrial steaming procedures. If the steaming chamber is unable to keep the temperature within 2 °C tolerance, target wood colour could not be warranted, and the final colour of steamed black locust will not be repeatable. Kollmann et al. (1951) concluded that the value of colour change depends on the temperature and requires high wood moisture content. Mainly this conclusion is the reason of those believes, that only the living wet wood can be steamed properly. To strengthen or refute this statement, not only wet but also dry samples (moisture content was about 10%) were steamed at different temperatures ranging from 85 to 110 °C. Visual observation has already refuted the previous beliefs. Similar colour changes were observed in both dry and wet samples. Objective colour measurement has confirmed the visual observation. Figure 3.18 presents the lightness change of dry (MC = 9%) and wet (MC = 25%) black locust steamed at 85 and 95 °C. The tendency of lightness decreases was similar, however, the intensity of the decrease in the case of dry sample was slower at the beginning than in the case of wet ones at 85 °C. With rising temperature this deviation diminished rapidly. At 95 °C there were no differences between the two curves, and similar results were found above this temperature limit. It shows that dry black locust samples can be steamed properly above 95 °C. It should be emphasized, however, that the 100% relative humidity in

110

3 Applications of Colour Measurement in Wood Research

Fig. 3.18 Lightness change of dry (MC = 9%) and wet (MC = 25%) black locust steamed at 85 and 95 °C

the steaming chest is even more important for proper steaming if the samples are initially dry. The colour shift towards red as a function of steaming time is presented in Fig. 3.19. The red colour represented by a* co-ordinate increased rapidly at the beginning of steaming. This tendency increased with rising temperature. Above 85 °C the curves have a maximum value. (Probably the curves also have a maximum at lower temperatures, but they are out of the examined steaming time range.) The place of maximum shifted towards shorter steaming time with rising temperature. At the meantime, the wastewater in the desiccator became darker. During steaming new chemical components are created containing conjugated double band systems. These structures arise through the degradation of extractives and hemicelluloses of black locust wood. The newly created colouring molecules are easily extractable so that the hot steam leaches them out leading to the decrease of the a* coordinate values. These leached components appear in the wastewater. The place of the curve maxima determines the optimum steaming time depending on the steaming temperature. Figure 3.20 presents the temperature dependence of places for maxima on the redness curves. The optimum steaming time at 120 and 130 °C is as short as 9.6 and 3 h. These intervals are extremely short under industrial conditions. It is impossible to guarantee equal temperature everywhere in a big autoclave within this short period because of the poor heat conductivity of wood. Therefore, steaming temperatures higher than 110 °C are not recommended if black locust is steamed in a large autoclave. Figure 3.21 represents the yellowness values (b* coordinate) of black locust wood generated by steaming at different temperatures. The initial values are quite high. Black locust is one of the most yellow wood species. This unattractive yellow colour is originated from the colour of robinetin type extractives which can be found in

3.3 Steaming as a Colour Modification Process for Wood

111

Fig. 3.19 Redness change of steamed black locust wood as function of steaming time and temperature

Fig. 3.20 The recommended maximum steaming time as function of steaming temperature

high quantity in black locust wood. The yellowness content of the colour decreased continuously during steaming at all investigated temperature. This decrease was slow or zero during the first day of steaming below 110 °C. Above 100 °C the decrease was more pronounced and became rapid at higher temperatures. It is thought that the robinetin is thermally degraded during steaming and its degradation products create brown colour. Our findings refute the second part of this theory. The degradation of robinetin (decreasing yellow coordinate) is slow or

112

3 Applications of Colour Measurement in Wood Research

Fig. 3.21 Yellowness change of steamed black locust wood as function of steaming time and temperature

zero during the first day of steaming. In contrast, the red content (increasing a* coordinate) shows the biggest change during the first day (Fig. 3.19). So, the shift of hue towards red is not originated from the structural changes of robinetin. This finding is reinforced by the pure robinetin deposited on the surface of wood during steaming in some cases. Further chemical analysis is needed to discover the details of the chemical changes. Figure 3.22 presents the locations of colour dots of steamed black locust samples applying various steaming times and temperatures. Starting dots on the upper left corner belong to the unsteamed samples followed by the dots of steamed samples with increasing steaming time. Colour dots of one steaming temperature outline a hoof-shaped curve. At 80 °C even the longest (22 days) steaming was too short for the colour alteration to form this characteristic hoof-shaped curve. The figure shows the dotted lines of some equal hue (h*) values as well. The hue values of steamed black locust cover an extremely large interval between 50 and 85°. Although, at first sight it does not seem to be a large interval of angles, Fig. 2.7 introduced that the hue values of European wood species are typically between 63 and 85°. Hue values below 60° represent rather dark brown colours characteristic to dark exotic species. This finding gives the possibility to substitute dark tropical wood by steamed black locust. Banadics et al. (2016) investigated the colour similarity between dark exotic species and steamed black locust. The results were published in a Hungarian journal. (The paper has English abstract and figure captions and contains photos as well.) It was found that the colour tones of steamed black locust give the same feeling as the dark exotic species, regardless if these colours match exactly those of any exotic species or not. The texture of steamed black locust wood was more attractive than the

3.3 Steaming as a Colour Modification Process for Wood

113

Fig. 3.22 Locations of colour dots for steamed black locust on the a*-b* plane. Constant hue (h*) lines are also indicated

texture of similar exotic wood in some cases. The appropriate steaming parameters were determined for creating the same or similar colour as some exotic species have. Figure 3.22 shows that steaming increases the chroma of wood at the beginning of the treatment. The distance between the origin of the coordinate system and the individual colour dots increases up to the maximum place of a* values. Steaming at 80 °C generates the greatest chroma increase but it needs extremely long steaming time. It is important to consider whether the colour alteration of black locust is generated by a single chemical change during steaming. The Arrhenius plot can give an answer to this question. Figures 3.23 and 3.24 present the Arrhenius plots of redness and yellowness changes. These figures demonstrate the logarithm (Ln) of redness and yellowness values versus reciprocal temperature. Thin dotted straight lines are trendlines. The colour dots follow each other with 5 °C intervals between 75 and 120 °C. The Arrhenius plot of lightness (not presented here) is not a straight line. Calculation of L* coordinate uses the reflectance values of almost the whole visible spectrum. This wavelength interval contains plenty of absorption bands belonging to different chemical groups. It can be stated that lightness change does not give detailed information regarding the chemical changes because of its complexity. The Arrhenius plot of redness (a*) has a breaking point at 105 °C (Fig. 3.23). The trendline is straight before the breaking point and after the breaking point as well demonstrating that the temperature dependence of both changes is exponential. The coefficient of determination of both sections are above 0.91 representing good linear correlations. The breaking point demonstrates that two types of chemical change occurred during steaming. As shown on Fig. 3.19, both increase and decrease of

114

3 Applications of Colour Measurement in Wood Research

Fig. 3.23 Arrhenius plot of redness change of black locust generated by one day steaming

Fig. 3.24 Arrhenius plot of yellowness change of black locust generated by two-day steaming

the redness value could be observed during the steaming of black locust. The steam induced redness increase is attributed to the generation of new chemical groups, which is dominant below 105 °C. On the other hand, the leaching effect of steam is dominant above 105 °C, and results in the decrease of redness. Figure 3.24 shows the Arrhenius plot of yellowness change (b*). This diagram has a breaking point at 90 °C. The trendline is straight both before and after the breaking point, and it is close to horizontal between 75 and 90 °C, representing that these temperatures do not cause yellowness change during the first two days of the steaming. The yellowness decreased exponentially with increasing temperature at temperatures higher than 90 °C. This finding confirms that the yellowness decrease

3.3 Steaming as a Colour Modification Process for Wood

115

Fig. 3.25 Colour dots of different black locust specimens presented on the a*–b* plane

of black locust is determined by the degradation of robinetin type extractives during steaming. Colour of black locust is highly inhomogeneous. This inhomogeneity can be observed between specimens and within a specimen as well. Figure 3.25 presents the colour dots distribution of many black locust specimens. Visually different specimens were chosen to present the inhomogeneity. The colour dots occupy a large area on the a*–b* plane. The redness values are between 1 and 7 units and the yellowness values cover the 24–35 interval. It is interesting to mention that increasing redness values are associated with decreasing yellowness values. The relatively large redness values belong to so-called iron-stained specimens. This type of black locust wood is not usual it can be found only in some special cases. This large colour inhomogeneity of black locust specimens makes it difficult to correctly repeat black locust tests. During the last three decades, a lot of steaming and photodegradation experiments have been carried out with black locust. The results are quite diverse depending on the initial colour parameter of the specimens. Steaming can reduce the inhomogeneity between and within specimens. Figure 3.26 shows the colour dots distribution of black locust specimens before steaming (colour dots marked as Target) and their shift on the a*–b* plane during steaming at 100 °C from day to day up to the 6th day. During the first day of the treatment, only the redness values changed considerably. The average value of yellowness did not change, however, dispersion of the colour dots became apparently smaller after one day steaming already. Then, the redness values hardly changed from the 2nd to the 6th day of the steaming while the yellowness values decreased continuously, so that the occupied area did not change eventually. Results show that the colour homogenisation occurred mainly during the first day of steaming. The standard deviation values show more precisely the colour homogenisation effect. Figure 3.27 shows the standard deviation values for L*, a* and b* as a function of steaming time at 100 °C. Initial standard deviation values were 2.26, 1.3 and 0.7

116

3 Applications of Colour Measurement in Wood Research

Fig. 3.26 Change of colour dots distribution during the steaming of black locust at 100 °C

for L*, b* and a*, respectively. These values represent the colour inhomogeneity of natural black locust. The colour homogenization effect of steaming is precisely reflected in the reduced standard deviation values of all three colour coordinates. Most of this reduction occurred on the first day of steaming. The standard deviation values then remained almost constant. This finding is important for the industry if the purpose of steaming is colour homogenisation. The most important finding of the laboratory steaming tests is that the colour change of black locust is highly temperature dependent. Colour change during the

Fig. 3.27 Change of standard deviation values during steaming of back locust at 100 °C

3.3 Steaming as a Colour Modification Process for Wood

117

laboratory steaming experiments was evenly distributed over the entire cross section of the specimens irrespective of the tested sample thickness. It is important to mention that equal temperature is strictly required in the whole cross section of wood during steaming to get homogeneous colour. Since the heat conductivity of wood is poor it is recommended to interrupt the increase of the temperature in the chamber for some hours when reaching 70–75 °C, if thick boards are treated. Keeping the temperature constant is important in order to ensure equal temperature and uniform colour modification in the entire cross section. For steaming above 110 °C, it is recommended to add two constant temperature periods. Uniformity of the colour modification achieved by steaming is considered to be a major advantage compared to surface treatment by pigmented stains or finishes. Reprocessing of manufactured products by sanding (for instance floors) or other processes does not bear any risk of jeopardizing the reselected colour of the wood surface. Another advantage of colour modification by steaming is that it is environmentally friendly compared to finishing with chemicals.

3.3.2 Steaming Properties of Beech and Turkey Oak Beech is primarily European species. It has medium density and moderate mechanical properties. It is used for manufacturing valuable and strong plywood panels and special plywood products. Beech is one of the most popular timber species of the furniture industry (Molnar and Bariska 2002). Beech is usually steamed to turn its white–grey natural colour into a more attractive reddish tint (Dzurenda 2013, 2022; Dzurenda and Dudiak 2020). Another purpose of the heat treatment could be to achieve colour homogenisation (Tolvaj et al 2009). Pre-treatment also helps the drying process of beech wood by softening of the wooden tissues which then become more elastic, allowing an easier migration of water through the capillaries. In addition, the inherent growth stresses in the wood are partially relieved during the steaming process, reducing the risk of cracking during drying (Timar et al.2016). The increasing red heartwood portion of beech is a big challenge for the timber industry nowadays. There is a great colour difference between white and red heart portions. The formation of red heart does not follow the boundary of the annual ring. The reason of red heart formation in logs is still only partly discovered. It was found that the increase of the pH value experienced at the boundary of red heartwood is indispensable for the enzymatic processes to take place. At these increased pH values, both enzymes (peroxidase and polyphenol-oxidase), proved to be responsible for the oxidation of phenolic compounds, have a very high activity. The sharp decrease of the concentration as well as the change of the composition of phenolic compounds can be experienced at the red heartwood boundary. The chromophores of the red heartwood are formed in front of the colour boundary in a narrow tissue range by the hydrolysis and the oxidative polymerisation of beech polyphenols (Albert et al. 2003; Hofmann et al. 2004, 2008). Radial variation in the polyphenol levels indicated that the concentration of most of the compounds increased in the transition

118

3 Applications of Colour Measurement in Wood Research

zone in front of the colour boundary, and decreased sharply behind it. In the red heartwood, only free aglycones could be evidenced in low amounts. According to the results, beech sapwood polyphenols behave differently during the red heartwood formation process. Based on their structure, function, and reactivity, some polyphenols undergo hydrolysis and accumulate as free aglycones or as their metabolites in the red heartwood tissues, while other compounds are bound to the cell wall structure as non-extractable polyphenols and contribute to the colour and resistance of red heartwood tissues (Hofmann et al. 2022) The total colour difference between white and false heartwood portions counted more than 18 units according to the results of Dzurenda (2023). The colour difference between white and red heart portions can be reduced by steaming (Tolvaj et al. 2009). For determining the steaming properties of beech wood, wet specimens (MC≈46%) were prepared containing white tissue and red heart separately. The applied steaming temperature were: 80, 90 and 100 °C. Based on the objective colour measurements, the lightness coordinate showed the greatest alteration (Fig. 3.28) since it decreased rapidly during the first 12 h of steaming. After this period, the lightness changed moderately or it remained constant. The main lightness change occurred during the first day of steaming below 100 °C. Both white and red heart presented continuous lightness decrease during the examined treatment period. Red heartwood presented slight lightness increase after 3-day steaming at 80 and 90 °C. The lightness decrease was independent of the steaming temperature below 100 °C. Comparing the lightness change of black locust and beech (Figs. 3.17 and 3.28), great differences are visible. While the lightness change of black locust is highly temperature dependent, beech does not show temperature dependence during the first 12 h of steaming. This phenomenon caused considerable difficulties in the industrial steaming practice from the 1960s until the 1980s since steaming chambers were not able to maintain constant temperature at that time. Due to the lack of seamless temperature control, it was impossible to produce the desired colour of black locust by the series of steaming. However, it was not a problem during steaming of beech. Results show that the extractives in beech are highly susceptible to steaming. Below 100 °C only the extractives are sensitive to thermal degradation, main chemical components (polyoses and lignin) remain stable. This is well demonstrated by the steam induced lightness change of beech. Extractives degraded during the first day of steaming already and no alteration could be observed afterward below 100 °C. However, the degradation of hemicelluloses generated continuous lightness decrease at 100 °C. As for the two different investigated wood parts, the lightness of white beech decreased more rapidly than that of red heartwood. This phenomenon gives the possibility to equalize the colour difference between the two wood parts. The homogenisation happened during the first 12-h of steaming, and the homogenisation effect was not temperature dependent. The redness change showed opposite tendency to that of lightness change (Fig. 3.29). The curves of red colour have a maximum. The increase before the peak was extremely intensive, while the decrease after that was moderate. Peak positions shifted towards shorter steaming time values with rising temperature. The peak values and positions were almost the same for both white and red heartwood. The

3.3 Steaming as a Colour Modification Process for Wood

119

Fig. 3.28 Lightness change of white (W) and red (R) heartwood of beech as a function of steaming time and temperature

redness decrease presented that the steam was partly able to leach out the degradation products of extractives reducing the redness value. The place of maximum determines the optimum steaming time to achieve the greatest redness increase. To get the most attractive red colour, the preferred steaming times are 2 days at 80–90 °C and 1 day at 100 °C. The redness coordinate showed rapid homogenisation between white and red heart material. The redness value of red heart was 1.4 times higher than that of white tissue. This difference disappeared during the first 6 h of steaming independently of the steaming temperature.

Fig. 3.29 Redness change of white (W) and red (R) heartwood of beech as a function of steaming time and temperature

120

3 Applications of Colour Measurement in Wood Research

Fig. 3.30 Yellowness change of white (W) and red (R) heartwood of beech as a function of steaming time and temperature

Yellowness did not present as intensive change as redness did, and there was no steaming temperature dependence either (Fig. 3.30). The yellowness of white beech decreased slowly during the 6-day steaming and the decrease was only 2–3 unit. The yellowness of red heart lessened rapidly during the first 12-h of steaming, but the decrease was only 2 units. After this initial phase, yellowness values did not change at any of the investigated steaming temperatures. Figure 3.31 presents the locations of colour dots of steamed white and red heart of beech wood on the a*–b* plane generated by various steaming times and temperatures. Starting dots on the left side of the lines belong to the untreated samples followed by the dots of steamed samples with increasing steaming time. The figure shows the dotted straight lines of some relevant equal hue (h*) values as well. Colour dots belonging to the same steaming temperature create a hoof-shaped curve representing that there was both redness value increase and decrease during the procedure. The occupied hue interval is relatively small, 15° for white tissues (between 58° and 73°) and only 10° for red heart (between 56° and 66°). For comparison, the same hue interval for steamed black locust was much greater, 35°. The last dots (results of 6-day steaming) are located between 60° and 62° for both white and red heart, confirming our findings in terms of the colour homogenisation effect of steaming. The chroma (h*) of individual colours (distance between the colour dot and the origin of the a*–b* coordinate system) increased at the beginning of steaming. As the change of chroma values was mainly determined by the redness change, the chroma increased up to the maximum place of redness values (Fig. 3.29). The increase of chroma gives more pleasant vivid colour tint. It can be concluded that steaming of beech wood does not need special care, the process is hardly temperature dependent. Steaming modifies the greyish-white colour of beech to a more pleasant reddish-brown tint. The chroma of both white and red

3.3 Steaming as a Colour Modification Process for Wood

121

Fig. 3.31 Colour dots of steamed white (W) and red heart (R) beech on the a*–b* plane. Constant hue (h*) lines are also indicated (dotted lines)

heart can be increased with properly chosen steaming parameters, considerably. One of the main advantages of the steaming of beech timber is the colour homogenisation between white and red heart. The great colour deviation does not disappear but the difference will be acceptable. Turkey oak is an East-Mediterranean species, but nowadays it is spreading northwards. Turkey oak develops a wide, light-grey sapwood, distinctly different from the dark reddish-brownish heartwood in colour. The border between dark and white portions is usually extremely sharp. Cold winter can produce longitudinal frost cracks and the tree try to isolate the defect by producing irregular dark tissue. The tree produces tick year ring portion around the crack during the next some years. The frequently occurring longitudinal frost ribs may be clearly visible on the stem surface. The cross section of a trunk presents the extreme colour inhomogeneity of Turkey oak timber (Fig. 3.32). The great colour differences are visible even in black and white. The frost cracks and their repair are well visible as frost rib. When traded, Turkey oak is sold in ‘white’ and ‘red’ bundles. After sawing a trunk, most of the boards contain both sapwood and irregular heartwood. This is the reason why turkey oak timber is mostly used as firewood. The great colour difference between Turkey oak sapwood and heartwood can be reduced by steaming (Molnar et al. 2006; Ferrari et al. 2013). The steaming properties of Turkey oak are presented here according to previous papers (Molnar et al. 2006 with written permission of Holztechnologie, Tolvaj and Molnar 2006 with written permission of Acta Silvatica et Lignaria Hungarica). White sapwood and dark regular heartwood (irregular dark heartwood was excluded) samples were

122

3 Applications of Colour Measurement in Wood Research

Fig. 3.32 Picture of the cross section for Turkey oak trunk containing frost cracks and irregular heartwood. (Photo: Zoltan Börcsök)

prepared with MC ≈ 55% for the test and 80, 95 and 110 °C steaming temperatures were applied. Turkey oak showed similar steaming properties as beech. The lightness change was independent of the temperature below 100 °C during the first one and two days of steaming for heartwood and sapwood (Fig. 3.33). Steaming at 80 °C did not generate lightness alteration after this period. Steaming at 95 °C resulted in slow but continuous lightness decrease. Steaming at 110 °C produced rapid lightness decrease during the first two-day treatment and the change remained moderate afterward for both sapwood and heartwood. The lightness of sapwood and heartwood was slightly equalized during the first day of steaming. Alteration of the red colour coordinate was different in sapwood and heartwood (Fig. 3.34). The changes were independent of the temperature below 100 °C. Sapwood produced rapid redness value increase at 80 and 95 °C during the first day of steaming. The redness increase was moderate but continuous afterward. Heartwood showed similar redness increase as sapwood but the intensity of change was moderate during the first three days of steaming below 100 °C. Steaming at 110 °C generated significant redness increase during the first 12 h of the treatment. Steaming redoubled the redness value of sapwood during the first 6 h followed by constant value

3.3 Steaming as a Colour Modification Process for Wood

123

Fig. 3.33 Lightness change of sapwood (S) and heartwood (H) of Turkey oak as a function of steaming time and temperature

between 0.5 and 3 days and slowly decreased afterward. The redness value increase of heartwood generated by 6-h steaming at 110 °C was half of the initial value. The redness homogenisation occurred during the first 12 h of steaming below 100 °C and during the first 6 h of steaming at 110 °C. Surprisingly, further steam treatment resulted in redness dehomogenisation after the initial homogenisation period. The redness value increase was greater in sapwood than in heartwood resulting in a higher redness value in sapwood than in heartwood. The redness difference between sapwood and heartwood was smaller at the end of the steaming than at the beginning

Fig. 3.34 Redness change of sapwood (S) and heartwood (H) of Turkey oak as a function of steaming time and temperature

124

3 Applications of Colour Measurement in Wood Research

Fig. 3.35 Yellowness change of sapwood (S) and heartwood (H) of Turkey oak as a function of steaming time and temperature

but the direction of difference was the opposite. The difference before steaming was 51–66% and it was 20–22% after 6-day steaming. These numbers show that the dehomogenisation was relatively small. It is important to mention that extremely dark irregular heartwood was excluded from the test. The steam induced yellowness change is presented in Fig. 3.35. Both sapwood and heartwood demonstrated slow but continuous yellowness increase which was independent of the temperature. Steaming produced small dehomogenisation in all examined cases. The yellowness change of Turkey oak due to the steaming was similar to that of beech. Figure 3.36 shows the colour dots of untreated and steamed sapwood and heartwood of Turkey oak on the a*–b* plane, generated by various steaming times and temperatures. Starting dots on the left side of the lines belong to the unsteamed samples followed by the dots of steamed samples with increasing steaming time. The figure also shows the dotted straight lines of some relevant equal hue (h*) values. The occupied hue interval is relatively small, 16° for sapwood (between 63° and 79°) and only 7° for heartwood (between 61° and 68°). For comparison, similar hue interval for steamed black locust was much greater (35°) while beech presented similar data. The final dots (results of 6-day steaming) are located between 65° and 68° for both white and red heart, representing the colour homogenisation. In terms of chroma, steaming produced great increase in sapwood and moderate increase in heartwood. Results show that steaming modified the greyish-white colour of sapwood to an attractive highly saturated brown tint. The greyish-brown initial tone of heartwood turned to a more saturated (less grey) brown tint. It can be concluded that steaming below 100 °C is suitable to modify the colour of Turkey oak to create attractive and highly saturated brown colour for both sapwood

3.3 Steaming as a Colour Modification Process for Wood

125

Fig. 3.36 Locations of colour dots of sapwood (S) and heartwood (R) of Turkey oak on the a*–b* plane. Constant hue (h*) lines are also indicated (dotted lines)

and heartwood. For colour homogenisation, a 12-h steam treatment proved to be long enough. The initial great colour difference between heartwood and sapwood would not completely disappear but it would be acceptable.

3.3.3 Steaming Properties of Poplar There are 35 species of the Populus genus. They are native to the temperate zone of the northern hemisphere, but nowadays they are planted in South America as well. In addition to the various species, cultivated varieties have a dominant role in plantation forestry. Good quality poplar timber is utilised for plywood and blockboard. Peeled poplar is suitable raw material for LVL beams. Poplars are the main rough materials of matches (Molnar and Bariska 2002). Poplar wood has initially greyish white colour. The colour difference between earlywood and latewood is hardly visible generating characterless texture. This colour appearance of poplar wood can be improved by steaming. Dry (with 8–12% moisture content) poplar timber (Populus x euramericana cv. Pannonia) was used for sample preparation to determine the steaming properties. The chosen steaming temperatures were 90, 100 and 110 °C. Heartwood and sapwood samples were steamed separately. The results are presented here according to a previous paper (Banadics and Tolvaj 2019). Based on the visual observation, the colour turned to much more pleasant brown compared to the initial colour. The lightness decreased during all of the applied

126

3 Applications of Colour Measurement in Wood Research

treatments and the colour hue shifted towards brown tone. The colour difference between earlywood and latewood appeared and the texture of wood became more pronounced. The colour of steamed wood was homogeneous throughout the whole cross section. As always, the objective colour measurement provided more detailed information regarding the colour alteration, than the visual observation. Figure 3.37 shows the lightness change due the thermal treatment for all temperature settings. Lightness was reduced significantly in all cases. The initial colour of heartwood was somewhat darker (3 units) than that of sapwood. This lightness difference remained small between sapwood and heartwood during the first 5 days of steaming. In the remaining period steaming resulted in a wider scattering of the lightness values. The only exception was the treatment at 90 °C, during which the lightness values of sapwood and heartwood shifted almost equally. The greatest lightness decrease was 28 units generated by 20 days steaming at 100 °C. The results strengthened that steaming is suitable to produce dark colour for poplar wood. Different physical and chemical processes generated the lightness decreases. The coloured degradation products of extractives and hemicelluloses reduced while leaching effect of high temperature steam increased the value of lightness. These processes are temperature dependent in different way. That is why the order of curves did not follow exactly the temperature range for sapwood but followed for heartwood. Lightness is not the proper parameter for tracking the chemical changes. Colour coordinates (a* and b*) are more important in a chemical sense while lightness is more relevant from the industrial point of view. Figure 3.38 shows the redness change due to the humid thermal treatment. The extractives with the hemicelluloses contribute mostly to temperature-induced colour

Fig. 3.37 Lightness change of sapwood (S) and heartwood (H) of poplar as a function of steaming time and temperature

3.3 Steaming as a Colour Modification Process for Wood

127

changes of wood by formation of various conjugated double bonds, carbonyl functionalities and quinoid structures (Xin et al. 2017). The initial redness of the specimens was low, 2.3 and 2.5 for heartwood and sapwood, respectively. This is because the extractive content of poplar is low. It means that extractives were not the main generators of redness increase. The redness value increased continuously for all specimens by elapsed steaming time. This increase was continuous during the whole steaming period. The initial values were multiplied 2.5–3 times during the first 5day period. This change was followed by moderate redness increase. In contrast, black locust which has a very high extractive content showed great redness increase within the first 2 days of steaming at temperatures above 90 °C (Fig. 3.19). After reaching the maximum the redness value of black locust decreased continuously above 90 °C. The reason is that the newly generated chromophore degradation products of extractive were not enough stabile. The high steam temperature may degrade them, reducing the redness value (Tolvaj et al. 2010; Csanady et al. 2015). In the case of poplar, however, continuous redness increase demonstrate clearly that the degradation products of hemicelluloses play dominant role in redness change during steaming. The a* values of sapwood and heartwood were close to each other during steaming at 90 °C. Rising steaming temperatures elevated the redness values for sapwood, and only the 110 °C did it for heartwood. From an industrial point of view, steaming at 110 °C is the most effective treatment to produce brown poplar timber. The highest redness value, which can be generated at 110 °C by 20 days steaming, is 10.55 units. This is 4.5 times higher than the initial redness value. Choosing the proper steaming time, the desired colour can be generated.

Fig. 3.38 Redness change of sapwood (S) and heartwood (H) of poplar as a function of steaming time and temperature

128

3 Applications of Colour Measurement in Wood Research

Figure 3.39 represents the yellowness change of poplar caused by steaming at 90, 100 and 110 °C. The increase of the yellow colour coordinate (b*) was similar to the a* value increase, however, the yellowness change was less temperature dependent. At 90 and 100 °C, the samples showed similar behaviour; the curves of sapwood and heartwood were close to each other independently of the steaming temperature. 110 °C generated significant yellowness increase for sapwood and small increase for heartwood, compared to the applied lower temperatures. The intensive increase was during the first 5 days of steaming. The initial yellowness values were multiplied 1.3 and 1.57 times during this period for heartwood and sapwood, respectively. This change was followed by moderate yellowness increase. The highest yellowness value, generated at 110 °C by 20 days steaming, was 33.25 units. This is almost twice as high as the initial yellowness value. In contrast, the highest redness value was 4.5 times greater than the initial value, showing that the redness change was much more dominant than the yellowness change. The steaming at 110 °C produced greater yellowness increase for sapwood than for heartwood generating visible colour difference. It is important to mention that steaming caused colour dehomogenisation of sapwood and heartwood of poplar. The chromophore molecules generated by steaming were stable against the leaching effect of steam. These steaming properties of poplar wood are the opposite of the properties of the previously discussed wood species (black locust, beech and Turkey oak). Figure 3.40 represents the colour dots of steamed sapwood and heartwood of poplar wood on the a*-b* plane generated by various steaming parameters. The starting dots on the left side of the trend lines belong to the unsteamed samples followed by the dots of steamed samples with increasing steaming time. The figure shows the dotted straight lines of some relevant equal hue (h*) values as well.

Fig. 3.39 Yellowness change of sapwood (S) and heartwood (H) of poplar as a function of steaming time and temperature

3.3 Steaming as a Colour Modification Process for Wood

129

Fig. 3.40 Colour dots of poplar sapwood (S) and heartwood (H) on the a*–b* plane before and after steam treatment. Constant hue (h*) lines are also indicated (dotted lines)

The occupied hue interval is relatively small, 14° for sapwood (between 68° and 82°) and 13° for heartwood (between 70° and 83°). Steaming produced continuous hue decrease independently of the steaming temperatures, representing the monotone colour alteration towards brown tint. Straight trendlines picture that the colour changes were generated by one main chemical change. Comparing the steaming properties of other hardwoods and conifers investigated in this work, poplar is the only exception presenting continuous and linear hue value decrease. This phenomenon supports the hypothesis that the colour change of poplar wood is mainly determined by the alteration of hemicelluloses during steaming. The initial chroma values of both sapwood and heartwood were relatively small (17.7 and 18.3 units for sapwood and heartwood) representing that natural poplar wood has colourless white-greyish tone. Steaming continuously increased the chroma values over treatment time for both sapwood and heartwood. Steaming at 110 °C generated the highest chroma values, and these were 35 for sapwood and 31 units for heartwood after the 20-day treatment. The great chroma increase is one of the most important advantages of steaming beside creating attractive brown colour. The experimental results showed that steaming is a proper method to produce brown colour for poplar wood. The colour chroma can be doubled by steaming. All these steam induced colour alterations in poplar wood lead to a favourable texture thus provide an added value in terms of aesthetic appearance, so that steam treated poplar could be a good choice for indoor wooden applications, especially claddings.

130

3 Applications of Colour Measurement in Wood Research

3.3.4 Steaming Properties of Scots Pine and Spruce Scots pine is one of the most adaptable tree species in the world. Its extractive content is about 2.2%. One of the most important extractives in Scots pine is resin to be found mainly in heartwood. Scots pine is used mainly for carpentry structures. It provides a good rough material for decorative, durable doors and windows. Spruce and Scots pine are frequently used for paper making (Molnar and Bariska 2002). Scots pine and spruce timbers are usually steamed to create colour similar to aged furniture and indoor wooden structures. The properly steamed softwood may be used to repair historical artefacts and relic furniture. Besides restoration, steamed stocks are excellent sources for manufacture of periodical furniture, where the aged appearance has aesthetical value. Scots pine sapwood and spruce specimens with dimensions 100 × 30 × 10 (mm) were prepared for the steaming test. The radially cut in-plane (i.e. 100 × 30 mm) surfaces contained earlywood and latewood regions as well. These radial surfaces were used for colour measurement Thus the measured colour values represented the average colour of earlywood and latewood. The initial moisture content of the samples was between 10 and 12%. Machined wood surfaces were slightly sanded. The treatment was carried out in a steaming vat with a 100% relative humidity condition at pre-set temperatures of 70, 80, 90, 95, and 100 °C. Although, steaming at 70 °C is an extremely slow colour modification process therefore it is never used in industrial conditions, this temperature produced considerable colour coordinate (a*, b*) alterations for the investigated conifers. Steaming properties of Scots pine and spruce timbers are similar. The results are presented here according to a previously published paper (Tolvaj et al. 2012). The specimens became visibly darker during steaming, and the colour hue turned towards brown. These changes were sensitive to the steaming temperature and to the steaming time. The observed colour change can be attributed to the alteration of the conjugated double bond systems found mainly in extractives. Conifers generally have moderate extractive content that is sufficient for colour modification by steaming. The main chemical components of conifers (polyoses and lignin) are usually intact by steaming in atmospheric conditions. The lightness decreases of the examined two conifers was similar. Here the lightness data of Scots pine are presented only (Fig. 3.41). The lightness of spruce and Scots pine samples decreased in the whole time-interval with increasing treatment time. Elevated temperature resulted in rapid lightness decrease during the first 4 days of treatment for both species, however, at 70 °C this trend could not be observed. The only difference between the two species was that the lightness of Scots pine samples hardly altered beyond the 18th day of steaming while spruce samples exhibited a continuous lightness decrease during the 22-day treatment. The redness change is presented here by the data of Scots pine (Fig. 3.42). The initial redness value of Scots pine was one unit higher than the initial redness value of spruce. The steam induced changes of the red colour co-ordinates showed more distinct deviations between the examined species than lightness did. Steaming at 70 °C produced slow but continuous redness increase and the exact same redness

3.3 Steaming as a Colour Modification Process for Wood

131

Fig. 3.41 Lightness change of Scots pine generated by steaming at different temperatures

alteration could be observed during the treatment of Scots pine at 80 °C. In comparison, spruce suffered much greater red colour change at 80 °C. Up to the 9th day of the treatments, higher temperatures caused higher redness increase for both species, except at 100 °C. Steaming at 100 °C was the only treatment during which the redness decreased after 12 days for spruce and after 9 days for Scots pine. Interestingly, redness values generated at 100 °C were smaller than those generated by the 95 °C treatment in the whole investigated time interval for both examined conifers. The reason is that the hot steam at 100 °C can leach out the new chromophores generated by the steam. (Slight leaching was observed also at 95 °C after 9-day steaming.) The new chromophore generation was dominant during the first 6 days of steaming at 100 °C for Scots pine and the leaching effect was dominant afterward. Figure 3.43 represents the yellowness change of spruce generated by steaming. The average initial yellowness value was 20.1 for spruce and 23.8 units for Scots pine. The difference was generated by the more yellow latewood of Scots pine. The treatment at 70 °C produced continuous yellowness increase during the whole examined treatment period for both investigated conifers. This temperature created the highest yellowness values during the 22-day steaming: 26 units for spruce and 29 units for Scots pine. The other trend lines of spruce yellowness change have a distinct maximum value at the 15th, 6th, 4th, and 2nd day of the treatment at 80, 90, 95 and 100 °C, respectively. Scots pine showed similar yellowness change properties as spruce, only the trendline of 80 °C did not have maximum. It appears that the steam leached out almost all of chromophore groups that were formed during steaming at 100 °C. Steaming time values belonging to the maxima determine the optimum steaming time except if a faded tone is needed for restoration of old wooden artefact. Steaming above 90 °C is recommended for this purpose. Figure 3.44 shows the colour dots of untreated and steamed spruce wood on the a*-b* plane generated by various steaming times and temperatures. The starting

132

3 Applications of Colour Measurement in Wood Research

Fig. 3.42 Redness change of Scots pine generated by steaming at different temperatures

Fig. 3.43 Yellowness change of spruce generated by steaming at different temperatures

dots on the left side of the trend lines belong to the unsteamed samples followed by the dots of steamed samples with increasing steaming time. The figure shows the dotted straight lines of some relevant equal hue (h*) values as well. The initial hue values were between 79° and 81°. All steaming temperatures generated continuous hue decrease towards brown tint during the whole treatment time. The shift of tint towards brown was even more intensive when the yellowness values decreased. The final hue values generated by the different steaming temperatures cover relatively large hue interval between 66 and 75°. The chroma of the generated colours also increased by elapsed steaming time up to the maximum places of yellowness change

3.3 Steaming as a Colour Modification Process for Wood

133

Fig. 3.44 Colour dots of steamed spruce on the a*–b* plane. Constant hue (h*) lines are also indicated (dotted lines)

(Fig. 3.44). The initial value of chroma was around 20 units. The highest chroma value (27 units) was produced by the 22-day steaming at 80 °C. It can be concluded that steaming can generate large variety of colours between the initial hues and brownish tone for conifers. The developed database can be used to identify optimum steaming parameters. Once a desired set of colour coordinates has been established, by selecting the appropriate steaming parameters, the current expensive trial and error practices may be avoided or at least reduced. Furthermore, steam-treated softwoods may be used to manufacture replicated historical furniture, joinery constructions, and other artefacts.

3.3.5 Steaming Properties of Larch and Sugi Larch is typical sub-alpine species, living on the northern hemisphere. The heartwood of larch is highly durable and weather resistant thanks to its resin and tannin content and high density. Larch timber is one of the most durable wooden materials for outdoor usage. It is applied in mining, bridge building, shipbuilding and rail ties. Its elegant, characteristic texture is advantageous in veneer manufacture (Molnar and Bariska 2002). The colour of larch timber is highly inhomogeneous because the heartwood is much darker and redder than the sapwood. Also, there are great differences between the colour of earlywood and latewood. Research was carried out to clarify the colour homogenisation possibilities of larch and to find out all the possible colours to be created by steaming. The hydrothermal treatment was carried out in a steam chest at 100% relative humidity at 90 and 110 °C. The sample size was 120 × 20 × 10 (mm). The largest surface contained only earlywood or latewood (tangential surface).

134

3 Applications of Colour Measurement in Wood Research

Half of each sample was sapwood while the other half was heartwood. The average moisture content of the samples was 8.5% before the steaming process. The results are presented here according to a previously published paper (Preklet et al. 2019 with written permission of Wood Research). Based on the objective colour measurements, the investigated four types of larch tissues have highly different colour hue and lightness. The initial average colour data of the investigated samples are presented in Table 3.1. The heartwood proved to be much darker than the sapwood. Especially, the earlywood in sapwood was much lighter than all other tissues. The latewood in sapwood had almost the same lightness as earlywood in heartwood. The darkest tissue was the latewood in heartwood. The border between the dark and light portions was sharp. The redness of larch wood showed the greatest diversity among the tissues. The latewood in sapwood and the earlywood in heartwood were more than two times redder (a* value), the latewood in heartwood almost four times redder than the earlywood in sapwood. There were moderate differences among the tissues in the yellow hue colour coordinate. These great colour differences can be diminished by steaming. The standard deviation (SD) values were small representing the relatively high colour homogeneity of earlywood and latewood separately. The latewood in heartwood had the highest SD values. This tissue has the greatest extractive content generating the colour of the tissue. The individual samples might contain extractives in different quantity establishing the colour inhomogeneity of the latewood in heartwood. Figure 3.45 shows the redness change generated by steaming at 90 °C. The colour dots of all tissues moved toward each other with elapsed steaming time representing colour homogenisation. The colour homogenisation continued during the first 9 days of steaming. During this steaming period, the initial redness value difference among the tissues (13.6 units) was reduced to 4.3 units. The redness of the two earlywood types was almost equal after 20 days of steaming, and the same happened for the two latewood parts as well. Nevertheless, the end values were different for earlywood and latewood. The steaming at 110 °C produced similar redness change as the treatment at 90 °C but this alteration was more intensive during the first five days of steaming (Fig. 3.46). Applying this higher temperature, the homogenisation was completed during the first 9 days of steaming, showing that the maximum of effective steaming time is 9 days if the aim is to homogenize the colour of larch. For industrial practice, Table 3.1 Initial colour data of different larch wood tissues. S = sapwood, H = heartwood, E = earlywood, L = latewood SD = standard deviation SE

L*

SD (L*)

79.29

1.25

a* 4.80

SD (a*)

b*

SD (b*)

0.54

23.37

1.17

SL

70.53

1.59

11.20

0.81

30.77

1.77

HE

71.92

2.80

13.08

1.94

29.65

1.79

HL

61.40

3.40

18.39

2.07

34.12

2.26

3.3 Steaming as a Colour Modification Process for Wood

135

Fig. 3.45 Redness change of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of larch generated by steaming at 90 °C

however, the proper steaming time would be less than 9 days since it is better not to completely homogenize the colour. This is because the unique colour harmony of the wood texture is generated by the moderate colour difference between earlywood and latewood. The redness difference can be altered between 14 and 2 units by choosing the proper steaming time (indicated by Fig. 3.46) to achieve the desired colour harmony.

Fig. 3.46 Redness change of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of larch generated by steaming at 110 °C

136

3 Applications of Colour Measurement in Wood Research

Redness values were almost constant in the second part of the applied steaming period (between 9 and 20 days). It means that the chromophore chemical groups generating the redness were stabile to the thermal treatment at 110 °C. Consequently, the colour of steamed larch wood is even more stable at ambient temperatures during the everyday usage. As discussed in previous chapters, steaming results of other wood species such as black locust, beech, Turkey oak, Scots pine, and spruce indicated that the red colour created by steaming was not stabile above 100 °C (Tolvaj et al. 2009, 2010, 2012; Tolvaj and Molnar 2006). The high temperature degraded the newly generated chromophore molecules and the steam leached out partly these coloured chemical compounds from the samples resulting in the decrease of a* values. In contrast, current experiments showed that the chromophore groups of larch were stable during the 20-day steaming at 110 °C. The colour stability of larch wood is an important advantage of steaming. Figure 3.47 represents the yellowness change caused by steaming at 90 °C. The colour dots of all tissues moved toward each other during the first 9 days of steaming representing the colour homogenisation. During this steaming period, the initial yellowness value difference (10.8 units) was reduced to as low as 1.8 units. Yellow hue values decreased continuously after the second day of steaming. The only exception was the earlywood in sapwood. Its yellowness has not changed after the fifth day of steaming. Only a small yellowness decrease was observed after the fifteenth day of steaming. The yellowness value of the two types of tissue (earlywood and latewood) of both sapwood and heartwood samples were close to each other after 14-day steaming, however, the end values were slightly different for sapwood and heartwood. The change in yellowness at 110 °C (not presented here) was only slightly different from the change at 90 °C. Determinative part of the yellowness change occurred

Fig. 3.47 Yellowness change of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of larch generated by steaming at 90 °C

3.3 Steaming as a Colour Modification Process for Wood

137

Fig. 3.48 Lightness change of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of larch generated by steaming at 90 °C

during the first two days of steaming. The yellowness of earlywood in heartwood hardly changed after this period. However, the yellowness value all other tissues slightly decreased during the further steaming process. The average value of yellowness after 20 days of steaming at 110 °C (33.7 units) was higher than at 90 °C (30.7 units). It shows that increasing temperature creates higher yellowness values. The lightness difference among the tissues was 17.9 units at the beginning of steaming (Fig. 3.48). The lightness values decreased continuously with elapsed steaming time at both temperatures (only the data of steaming at 90 °C are presented here). The difference between the effects of two temperatures was negligible during the whole treatment process. The homogenisation effect of steaming can be visualised if the colour dots are presented on the a*–b* plane as shown in Figs. 3.49 and 3.50 for 90 °C and 110 °C steaming, respectively. It is well visible that the initial colour dots (filled marks) representing untreated samples are fare from each other. The hue values are between 61° and 78°. The colour dots moved toward each other and formed a hoof-shaped trend line later on generated mainly by the increase and decrease of yellowness values. The coordinates of a centre dot (X) were calculated as the average of coordinates for the dots of 20-h steaming time (enlarged symbols). The coordinates of this centre dot are: a* = 14.2 and b* = 30.7 units (Fig. 3.49). This centre dot is located close to the centre of the initial colour dots (filled symbols). This fact shows that steaming at 90 °C does not created great colour alteration (except SE). The hue value of the centre dot is 65.2°. The initial hue values were 78 for earlywood and 70° for latewood in sapwood. Similar data for earlywood and latewood in heartwood were 66° and 61°. The distances among the corresponding colour dots (equal steaming time) decreased due to the steaming representing the homogenisation effect of the treatment. The

138

3 Applications of Colour Measurement in Wood Research

Fig. 3.49 Colour dots of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of larch generated by steaming at 90 °C. Constant hue (h*) lines are also indicated (dotted lines)

chroma value of earlywood in sapwood increased a lot. The chroma value of other tissues showed both increase and decrease. The high chroma value of latewood in heartwood mainly decreased during steaming. It can be concluded that the chroma values of all tissues moved towards homogenisation as well. The colour dots of steaming at 110 °C are presented in Fig. 3.50. The changes of both colour coordinates (a* and b*) were large during the first 2 days of steaming and the alteration was small afterward. The result was that most of colour dots were located only within a small area after 2-day steaming, representing excellent colour homogenisation. The coordinates of a centre dot (X) were calculated as the average of the coordinates for the dots of 20-h steaming time (enlarged symbols). The coordinates of this centre dot are: a* = 13 and b* = 33.7 units. This calculated centre dot is considerably fare from the centre point (a* = 11.8; b* = 29.6) of the initial colour dots (filled symbols). This deviation presents that steaming at 110 °C produced greater yellowness value increase than steaming at 90 °C. The hue values of centre dots (before steaming and at the end of steaming) were equal (≈ 68°). The experimental results showed that the colour of larch wood is highly inhomogeneous. It has light earlywood in sapwood. The colour of latewood in sapwood has similar colour as the earlywood in heartwood. This colour is darker and much redder than the colour of earlywood in sapwood. The latewood in heartwood has the darkest and most reddish colour. Steaming can reduce these large colour differences. Wide range of colours was created by steaming between the initial colour and brown colour depending on the steaming time and temperature. The colour homogenisation was so successful that it was difficult to differentiate sapwood and heartwood by naked eye at the end of the steaming process at 110 °C. The effective steaming time

3.3 Steaming as a Colour Modification Process for Wood

139

Fig. 3.50 Colour dots of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of larch generated by steaming at 110 °C. Constant hue (h*) lines are also indicated (dotted lines)

for colour homogenisation was 5 and 2 days at 90 °C and 110 °C, respectively. The chroma of most tissues increased considerably (except latewood in heartwood) due to the steaming at both applied temperatures, representing that steaming can generate saturated colour. Sugi is the national tree of Japan. It is also the most common commercial softwood material in Japan. It is lightweight but strong timber and easy to saw. Sugi is mainly used as building material, pillar, ceiling board, panelling, fences, furniture, packaging material, craft products, barrels and chopsticks. Sugi is waterproof and resistant to decay. Sapwood of sugi is straw coloured and clearly demarcated from the heartwood which is reddish brown. Latewood in both sapwood and heartwood is considerably darker than earlywood. Sapwood and heartwood samples were prepared containing only earlywood or latewood (tangential surface) on the measured surfaces for the steaming procedure. The average moisture content of the samples was 9.1% before the steaming process. Steaming was carried out at 90 and 110 °C. Colour appearance and inhomogeneity of sugi timber is similar to the colour appearance of larch timber. Steaming properties of sugi are also similar to that of larch. Because of the similarity, only the a* and b* parameters of sugi are presented here. (Detailed steaming properties of sugi were published in a previous paper (Tolvaj et al. 2019).) The homogenisation effect of steaming can be visualised by plotting the colour dots on the a* -b* plane. Figures 3.51 and 3.52 show the colours generated by steaming at 90 and 110 °C, respectively. Filled marks represent the colour dots of tissues before steaming. The initial colour dots are evidently far from each other because of the high colour inhomogeneity of sugi wood. The colour dots converged

140

3 Applications of Colour Measurement in Wood Research

Fig. 3.51 Colour dots of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of sugi generated by steaming at 90 °C. Constant hue (h*) lines are also indicated (dotted lines)

towards a centre dot (X) during the steaming process (Fig. 3.51). The coordinates of this centre dot are a* = 12.1 and b* = 27.4 units. The distances among the individual colour dots decreased with elapsed steaming time representing the colour homogenisation effect of steaming. The croma of the investigated surfaces changed in different ways. Steaming generated continuous croma value increase for earlywood in sapwood and continuous croma value decrease for latewood is heartwood. The other tissues showed small croma value increase and decrease as well. Similar changes were monitored when sugi wood was steamed at 110 °C (Fig. 3.52). There were great changes during the first two days of steaming. The direction of changes is different compared to the effect of steaming at 90 °C. The colour dots moved towards a common point, but this point is not in a central position. (The coordinates of centre point for initial colours of the different tissues were a* = 11.1 and b* = 25.6.) This happened because the yellowness values increased considerably, elevating the colour dots on the chart during the steaming. The middle point of the final colour dots (X) is located on the top of the diagram; its coordinates are a* = 12,8 and b* = 33,5 units. Most of the change happened within the first two days of steaming. The colour dots were close to each other after the 5th day of steaming, representing colour homogenisation. All tissues became more saturated in colour during the steaming at 110 °C. The chroma of earlywood increased much more than the chroma of latewood in sapwood. The chroma has hardly changed after the second day of steaming. Only the tissues of heartwood showed some increase in chroma after the second day of steaming. This result is important for the wood industry, because a more saturated colour is generally more acceptable for the end users than a dull colour.

3.4 Effect of Wetting and Finishing on Wood Colour

141

Fig. 3.52 Colour dots of earlywood (E) and latewood (L) in sapwood (S) and heartwood (H) of sugi generated by steaming at 110 °C. Constant hue (h*) lines are also indicated (dotted lines)

The measured data showed that the colour of sugi wood is highly inhomogeneous. It has light earlywood in sapwood. The colour of latewood in sapwood has similar colour as the earlywood in heartwood. This colour is darker and much redder than the colour of earlywood in sapwood. The latewood in heartwood has the darkest and most reddish colour. Steaming was able to reduce these large colour differences. A wide range of colours was created by steaming between the initial colour and brown colour depending on the steaming time and temperature. The colour homogenisation was so successful that it was difficult to differentiate between the sapwood and the heartwood by naked eye at the end of the steaming process. The effective steaming time for colour homogenisation was nine and two days at 90 °C and 110 °C, respectively.

3.4 Effect of Wetting and Finishing on Wood Colour The light beam travelling in a material is partly absorbed and partly scattered. Scattered photons are partly absorbed while others return and leave the material surface together with the photons reflected directly by the surface. The increasing depth of light penetration amplifies the absorption of scattered photons and reduces the reflected intensity. Consequently, the lightness of the visible colour is decreasing if the penetration depth is increasing. This phenomenon is visible if the wood surface is wetted. Water conducts the light into deeper layers of the wood increasing the absorption of the scattered photons. Surface finishing such as transparent lacquer can conduct the light into even deeper layers than the water. Figure 3.53 presents

142

3 Applications of Colour Measurement in Wood Research

Fig. 3.53 Reflectance spectra of black locust wood under different conditions

the reflectance spectra of natural, wetted and lacquered black locust samples. The wet surface was generated by pouring distilled water on the surface of the specimen. Residual water was removed from the surface after half an hour, then the colour was measured immediately. The lacquered surface was configured with two transparent layers. The reflectance curve of wetted specimens is located under the curve of natural black locust specimens. This deviation proves that the wetted wooden surface is darker than the natural one. The difference among the curves is only slightly increasing with decreasing wavelength, reflectance curves of the three investigated surfaces run almost parallel indicating that wetting does not generate significant hue value change. The figure presents that the reflected intensities were a little smaller for lacquered surface than for wetted surface. This phenomenon shows that the penetration of light was slightly deeper on lacquered surface than on wetted surface. The effect of wetting is interesting on the surface of steamed timber. Figure 3.54 shows the reflectance spectra of beech wood created by different treatments. Both natural and steamed beech samples were wetted. Steaming was performed at 95°C for one day. The wet surface was generated by pouring distilled water on the surface of the specimen. Residual water was removed from the surface after half an hour, then the colour was measured immediately. The effect of wetting for beech is quite similar to the effect of wetting for black locust (Fig. 3.53). The reflectance values of natural beech were mainly reduced in the 400–600 nm wavelength range and the colour hue changed very little by wetting. Steaming, however, changed the colour hue considerably and the course of the reflectance curve of steamed beech also differs from that of the unsteamed one. An additional wetting further stressed the reflectance reducing effect of steaming and diminished the reflection almost evenly in the entire spectrum while the course of the curve did not change much compared to that of the steamed sample.

3.4 Effect of Wetting and Finishing on Wood Colour

143

Fig. 3.54 Reflectance spectra of beech wood under different conditions

Calculation of the difference spectra (derived by subtracting the reflectance spectrum of the initial surface from that of the wet surface) show clearly the wetting induced changes (Fig. 3.55). Wetting of the unsteamed samples resulted in almost linear reflectance value decrease in the whole visible region. There is only a little deviation in the 400–410 nm region. The wetting of steamed samples produced much greater reflectance decrease than the wetting of untreated samples and the tendency of change was completely different. The maximum difference is at 600 nm and its value is 20 units. This finding shows that steaming modifies the structure of beech wood and this structural change alters the wetting properties of the wood material as well. Steamed beech was able to adsorb three times more water during the half-hour wetting period than unsteamed beech. (Details can be found in Table 3.3.) To clarify the species dependence of wetting, the following European species were chosen to investigate the colour modification effect of wetting: alder (Alnus glutinosa L.), black locust heartwood (Robinia pseudoacacia L), beech (Fagus silvatica L.), steamed beech (steaming temperature: 95 °C, steaming time: 1 day), birch (Betula pendula Roth), linden (Tilia cordata Mill.), poplar (Populus x euramericana cv. Pannonia), oak heartwood (Quercus petraea), larch heartwood (Larix decidua L.), Scots pine heartwood (Pinus sylvestris L.) and spruce (Picea abies Mill.). Steamed beech was chosen because beech timber is often utilized in steamed condition due to the improved colour appearance. The wet surface was generated by pouring distilled water on the surface of the specimen. The remaining water was removed after half an hour and the colour was measured immediately afterward. Colour measurement was performed on the radial surface of the samples. The results are presented here according to a previous paper (Tolvaj and Preklet 2015). The lightness of the investigated species before and after wetting is presented in Fig. 3.56. Wetting reduced the lightness values of all species. Similar results were published by Teischinger et al. (2012) and Meints et al (2017). Lightness value is

144

3 Applications of Colour Measurement in Wood Research 0

Beech diff. wetting

Reflectance difference

Steamed b. diff. wetting -5

-10

-15

-20 400

450

500

550

600

650

700

Wavelength (nm) Fig. 3.55 Reflectance differences of beech and steamed beech generated by wetting

proportional to the intensity of the reflected light. Reflectance values are determined by the reflectivity of the surface and by the absorption properties of the material. Photons travelling in a material are partly absorbed and partly scattered. (The travelling direction changes during scattering.) Some part of the scattered photons returns and leave the material increasing the reflected intensity. Wetting increases the depth of light penetration thus increasing the possibility of absorption. Less scattered photon returns from the wetted wood than from the dry wood. The partial density of water in the penetration layer is also an important parameter to increase the light conductivity of wood. This parameter is highly species dependent. Steaming highly modified the water uptake properties of beech. During the 30-min wetting, water uptake of steamed beech was more than three times greater (in g/cm2 ) than the same data of unsteamed beech. This parameter shows that the partial water density in the wetted layer is much greater in steamed beech than in natural beech. The wetting induced lightness decrease of unsteamed beech was 4.3%. In contrast, steamed beech produced 38.6% lightness decrease. This change was the greatest among the investigated species. Conifers produced smaller lightness decrease than most of the deciduous species (Table. 3.2). The smallest lightness decrease was produced by poplar (2.3%) together with linden (3%) and spruce (3.4%). These species have the smallest extractive content among the investigated species which is important in terms light absorption. Most of the extractives are intensive light absorber together with lignin. Low extractive content generated small absorption and most of scattered photons returned from the internal layers of wood increasing the reflectance values. These tree species had the greatest lightness values (above 80 units) in dry condition as well because of the low absorption rate of scattered photons. Redness was the most sensitive colour parameter to wetting. All investigated species produced redness increase due to the wetting (Fig. 3.57). There are great

3.4 Effect of Wetting and Finishing on Wood Colour

145

Fig. 3.56 Lightness values of different European wood species before and after wetting

Table 3.2 Lightness, redness and yellowness changes of different European wood species in percentage and the total colour difference (ΔE*) generated by half an hour wetting Order of change

L* Name

a* Change (%)

Name

ΔE*

b* Change (%)

Name

Change (%)

Name

Value

1

St. beech

39

Alder

97

St. beech

82

St. beech

32.9

2

Oak

17

Birch

97

Birch

81

Birch

20.6

16

62

3

Birch

St. beech

92

Alder

Oak

19.4

4

Bl. locust 14

Oak

89

Bl. locust 53

Alder

17.9

5

Sc. pine

Sc. pine

80

Oak

Bl. locust 17.7

11

52

6

Alder

11

Bl. locust 67

Spruce

47

Sc. pine

15.6

7

Larch

7

Beech

43

Sc. pine

45

Larch

13.4

8

Beech

4

Spruce

36

Beech

39

Beech

10.1

9

Spruce

3

Larch

34

Larch

37

Spruce

10.0

10

Linden

3

Linden

30

Poplar

32

Linden

7.9

11

Poplar

2

Poplar

25

Linden

29

Poplar

6.5

differences among the species in terms of wetting induced redness increase, ranging from 25 to 97%. These large differences suggest that the extractives that contribute greatly to the redness values are found in very different amounts in the different species. Wetting almost doubled the redness value of some species, including alder, birch and steamed beech producing 97.3, 96.6 and 92% redness increase, respectively. The smallest redness increase was produced by poplar (25%). Species with low initial

146

3 Applications of Colour Measurement in Wood Research

Table 3.3 Water uptake data of different European wood species generated by half an hour wetting Order of change

According to the sum of position’s number in L*, a* and b* lists (Table 3.2)

Water uptake/unit area (g/cm2 )

1

Steamed beech

Steamed beech

0.143

2

Birch

Alder

0.141

3

Alder

Birch

0.092

4

Oak

Linden

0.056

5

Black locust

Beech

0.041

6

Scots pine

Poplar

0.04

7

Beech

Spruce

0.033

8

Spruce

Larch

0.029

9

Larch

Scots pine

0.024

10

Linden

Oak

0.023

11

Poplar

Black locust

0.013

redness value (below 5 units) are poplar, linden and spruce. These species have also the lowest extractive content thus they produced only a little redness increase during wetting (Table 3.2). The redness value of steamed and wetted beech stands out among the investigated species. Steamed beech has relatively high redness and this value was almost doubled by wetting. The wetting induced yellowness change of the investigated species (Fig. 3.58) was more balanced among the species than the redness change. All species presented

Fig. 3.57 Redness values of different European wood species before and after wetting

3.4 Effect of Wetting and Finishing on Wood Colour

147

Fig. 3.58 Yellowness values of different European wood species before and after wetting

yellowness increase due to the wetting. The percentage of changes were between 29 and 82%. The greatest yellowness increase was presented by steamed beech (81.7%) and by birch (80,6%). Linden was the least sensitive species. Its yellowness increased by 29%. Figure 3.59 shows the alteration of hue values generated by wetting. In spite of the great redness and yellowness changes, the hue values hardly changed. Small hue value increases and decreases also happened. It means that the visual colour tint was not affected by wetting but the values of individual colour coordinates (a* and b*) increased. The perceived colours were more vivid after wetting. This phenomenon indicates that colour chroma increased due to wetting. The change of chroma is presented in Fig. 3.60. Wetting increased the chroma value of all investigated species. As chroma is determined by the a* and b* coordinates the order of changes is the same as the integrated order according to a* and b*. Steamed black locust (84.2%) and birch (82.8%) showed the greatest chroma increase while linden (29%) and poplar (31%) produced the smallest chroma change. The chroma represents the distance between the colour dot and the origin of the a*b* coordinate system. While small chroma value represents gray colour tone, high chroma value characterises vivid colour tint. Wetting always turns the colour towards more intensive tint. Similar results were published by Meints et al (2017). Table 3.2 presents the order and the percentage of changes for the species according to the alteration of the colour coordinates. The lightness change was moderate compared to the alteration of redness and yellowness. Most of the investigated species produced greater redness increase than yellowness increases except larch, spruce and poplar, although the differences were small for these species. The greatest differences between the changes of a* and b* were produced by oak (37%)

148

3 Applications of Colour Measurement in Wood Research

Fig. 3.59 Hue values of different European wood species before and after wetting

Fig. 3.60 Chroma values of different European wood species before and after wetting

and by alder (35%). The column of total colour change (ΔE*) provides the order of specimens in terms of the integrated change generated by wetting. Position of the species in the ΔE* list is determined mainly by the lightness change (see oak, alder, larch, spruce). The sum of the sequential numbers in the L*, a*, b* lists also provide an order (Table 3.3). These two orders of positions are similar. Only larch shows

3.4 Effect of Wetting and Finishing on Wood Colour

149

two positions difference. Table 3.3 presents the water uptake data as well. The water uptake data were determined by wetting the surface of the samples for 30 min. The mass of specimens was measured before and after wetting to determine the absolute water uptake. A water uptake parameter was calculated by the quotient of adsorbed water and the adsorbing surface in g/cm2 . This parameter gives the mass of adsorbed water per one cm2 area. The water uptake of steamed beech was 3.5 times greater than the water uptake of unsteamed beech. Beech specimens underwent swelling during steaming and shrinkage during drying but the shrinkage was smaller than the swelling. There was remained swelling and it counted 1.0, 2.7 and 5.4% in longitudinal, radial and tangential directions. This remained swelling multiplied the volume of water uptake of steamed beech compared to unsteamed beech. The velocity of water penetration is an important parameter regarding to the colour change by wetting, although it is difficult to measure directly. The change of reflectance values can give information regarding to the water penetration properties of the specimens. The changes of reflectance values are determined by the penetration depth and the partial density of water in the penetration layer. Time dependence of wetting was monitored by reflectance measurement. The reflectance spectrum was measured after 10 s, 1-, 5-, 20- and 30-min wetting. New sample was used for all individual measurement. Three species (steamed beech, birch and larch) were chosen to demonstrate the different water uptake properties of the species. Steamed beech was chosen because steaming modified the cell wall structure. Birch showed the greatest colour alterations during wetting. Larch represents conifers having high resin content and it showed the smallest colour change by wetting among the investigated conifers (Table 3.3). Figures 3.61, 3.62 and 3.63 show the reflectance spectra of the chosen species after different wetting durations. Steamed beech presented the greatest and larch the smallest time dependence of reflectance change by wetting. The greatest shift of the reflectance spectrum happened during the first 10 s of wetting for all species. Visible light does not penetrate deep into wood. Hon and Ifju (1978) determined the penetration depth of UV and visible light by detecting the free radicals generated by the photodegradation. The measured penetration depth was 75 µm for UV light and 200 µm for visible light in case of dry wood. It seems, that 10 s wetting was enough for rapid penetration increase for light. Further water uptake increased the partial water density in the penetration layer and the partially more water conducted the light into deeper layers of wood. This process was relatively slow and stopped after 30 min wetting. Probably, the water penetration continued after 30-min wetting but the photons of light were unable to reach this region because of scattering and strong absorption. Great differences were found among the species regarding to the time dependence of reflectance decrease. Steamed beech presented relatively intensive reflectance decrease up to twenty minutes (Fig. 3.61). Similar type of decrease continued only up to five minutes for birch (Fig. 3.62). The reflectance decrease was negligible for birch afterward. Beech, black locust and oak presents similar reflectance decrease during the first 10-min wetting as birch, but the reflectance decrease was negligible afterward.

150

3 Applications of Colour Measurement in Wood Research

Fig. 3.61 Reflectance spectra of steamed beech specimens before and after 10 s, 1-, 5-,10-, 20and 30-min wetting

Fig. 3.62 Reflectance spectra of birch specimens before and after 10 s, 1-, 5-, 20- and 30-min wetting

Larch showed completely different water uptake properties (Fig. 3.63) than birch. Wetting caused only a small reflectance decrease and this small decrease occurred during the first 10 s of wetting. Scots pine, spruce and poplar showed similar wetting properties as larch regarding the change of reflectance spectra. For investigating the colour modification effect of lacquering, the same species were chosen as for the wetting tests. Two transparent lacquer layers were applied and the colour parameters were determined after the hardening of the first and second layers.

3.4 Effect of Wetting and Finishing on Wood Colour

151

Fig. 3.63 Reflectance spectra of larch specimens before and after 10 s, 1-, 5- and 30-min wetting

Figure 3.53 shows the reflectance spectra of lacquered and wetted black locust samples. Both treatments produced considerable reflectance value decrease in the whole visible wavelength interval. There is only a small difference between the spectra of wetted and lacquered surfaces. The lacquer layer generated a little greater reflectance decrease than wetting. The other species showed similar properties, only birch samples presented equal reflectance values after both treatments. These results suggest that wetting and lacquering of wooden surfaces generate similar colour changes. The measured and calculated colour parameters strengthened this hypothesis. The lightness decrease was similar generated by both treatments. The only exception was the steamed beech, which produced considerably smaller lightness decrease after lacquering than after wetting. The same happened for redness and yellowness change as well. Steaming modified the cell-wall structure of beech by generating reminder swelling. The small water molecules were able to penetrate deeper than the much bigger molecules of lacquer and the deeper penetration produced darker colour. Redness values generated by one and two lacquer layers are presented in Fig. 3.64. All investigated species produced redness increase due to the lacquering. There were great differences among the redness increase of the species. The second layer increased the redness values even more for all species. The smallest and the greatest effect of the second layer was found on poplar (only 0.2 units increase) and on black locust (3.1 units) surfaces, respectively. The redness increase caused by the two-layer lacquering is between 48 and 128%. This interval is between 25 and 97% in case of wetting. These results show that lacquering produced greater redness increase than wetting (see Fig. 3.55) in general, however, there are some exceptions. Steamed beech produced smaller redness increase by lacquering than by wetting, which can be interpreted by the penetration depth (see the explanation of lightness change above). The other exception was the Scots pine. This deviation needs further investigation.

152

3 Applications of Colour Measurement in Wood Research

Fig. 3.64 Redness values of different European wood species before and after lacquering

The investigated lacquered wood surface produced considerably greater yellowness increase than the wetted surface (Figs. 3.58 and 3.65). The only exception was the steamed beech. The second layer considerably elevated the yellowness values. The second layer generated the greatest difference for larch (10 units) and the smallest for birch (6 units). These differences are much greater than the differences in case of redness changes. The reason can be that the lacquer itself showed a blanch yellow tone. The greatest and the smallest yellowness increase due to the lacquering was observed on poplar (126%) and on steamed beech (70%), respectively. The chroma values were considerably greater for lacquered surfaces than for wetted surfaces. The only exception was the steamed beech. The hue values were hardly affected by lacquering, similarly to the wetting. The experimental results showed that wetting highly modifies the colour parameters of wood. The rate of modification is highly species dependent. The modification effect of the moisture must be kept in mind in all treatments and usages where wood moisture content is affected during the procedure. The best example of this is the outdoor weathering of wood, where the moisture content of the specimens is always changing due to the rain and the varying relative humidity of the air. Unfortunately, some researchers ignore this phenomenon, and the measured chaotic colour data prevents correct evaluation.

References

153

Fig. 3.65 Yellowness values of different European wood species before and after lacquering

References Albert, L., Hofmann, T., Németh, Zs., Rétfalvi, T., Koloszár, J., Varga, Sz., Csepregi, I.: Radial variation of total phenol content in beech (Fagus sylvatica L.) wood with and without red heartwood. Holz Roh Werkstoff 61(3), 227–230 (2003). https://doi.org/10.1007/s00107-0030381-x Banadics, E.A., Gálos, B., Tolvaj, L.: A sötét egzóta faanyagok helyettesítése g˝ozölt akác faanyaggal. [Substitution of dark exotic wood species by steamed black locust.] Faipar 64(1), 22–28 (2016) Banadics, E.A., Tolvaj, L.: Colour modification of poplar wood by steaming for brown colour. Eur. J. Wood Prod. 77, 717–719 (2019). https://doi.org/10.1007/s00107-019-01397-9 Bekhta, P., Niemz, P.: Effect of high temperature on the change in color, dimensional stability and mechanical properties of spruce wood. Holzforschung 57, 539–546 (2003). https://doi.org/10. 1515/HF.2003.080 Boonstara, M.J., Tjeerdsma, B.: Chemical analysis of heat treated softwoods. Holz Roh Werkstoff 64, 204–211 (2006). https://doi.org/10.1007/s00107-005-0078-4 Chen, Y., Fan, Y., Gao, J., Stark, M.S.: The effect of heat treatment on the chemical and color change of black locust (Robinia pseudoacacia) wood flour. BioResources 7(1), 1157–1170 (2012) Csanady, E., Magoss, E., Tolvaj, L.: Quality of Machined Wood Surfaces. Springer, Heidlberg New York London (2015). https://doi.org/10.1007/978-3-319-22419-0 Dzurenda, L.: Modification of wood colour of Fagus sylvatica L. to a brown-pink shade caused by thermal treatment. Wood Res. 58(3), 475–482 (2013) Dzurenda, L.: Colour modification of Robinia pseudoacacia L. during the processes of heat treatment with saturated water steam. Acta Facultatis Xylologiae Zvolen 60(1), 61−70 (2018). https://doi.org/10.17423/afx.2018.60.1.07 Dzurenda, L., Dudiak, M.: Changes in wood tree species Fagus sylvatica L. and characteristics of the thermal process of modifying its color with saturated water steam. Adv. Ecol. Environm. Res. 5(4), 142–156 (2020) Dzurenda, L.: Range of color change of Beech Wood in the steaming process. BioResources 17(1), 1690–1702 (2022)

154

3 Applications of Colour Measurement in Wood Research

Dzurenda, L.: Natural variability of the color of Beech Wood in the color space CIE L*a*b*. Forests 14(6), ID 1103 (2023). https://doi.org/10.3390/f14061103 Esteves, B.M., Pereira, H.M.: Wood modification by heat treatment: a review. BioResources 4(1), 371–404 (2009) Esteves, B., Marques, A.V., Domingos, I., Pereira, H.: Chemical changes of heat treated pine and eucalypt wood monitored by FTIR. Maderas, Sciencia Technol. 15(2), 245–258 (2013). https:// doi.org/10.4067/S0718-221X2013005000020 Ferrari, S., Allegretti, O., Cuccui, I., Moretti, N., Marra, M., Todaro, L.: A revaluation of Turkey oak wood (Quercus cerris L.) through combined steaming and thermo-vacuum treatments. BioResources 8(4), 5051–5066 (2013) Gasparik, M., Gaff, M., Kacik, F., Sikora, A.: Colour and chemical changes in teak (Tectonia grandis L.f.) and Meranti (Shorea spp.) wood after thermal treatment. BioResources 14(2), 2667–2683 (2019) Gierlinger, N., Jacques, D., Grabner, M., Wimmer, R.: Colour of larch heartwood and relationships to extractives and brown-rot decay resistance. Trees 18(1), 102–108 (2004). https://doi.org/10. 1007/s00468-003-0290-y Griebeler, C., Tondi, G., Schnable, T., Iglesias, C., Ruiz, S.: Reduction of the surface colour variability of thermally modified Eucalyptus globulus wood by colour pre-grading and homogeneity thermal treatment. Eur. J. Wood Prod. 76, 1495–1504 (2018). https://doi.org/10.1007/s00107018-1310-3 Hofmann, T., Albert, L., Rétfalvi, T., Bányai, É., Visi-Rajczi, E., Börcsök, E., Németh, Zs.: Quantitative TLC analysis of (+)-Catechin and (−)-epicatecin from Fagus sylvatica L. with and WITHOUT Red Heartwood. J. Planar Chromatogr. 17, 350–354 (2004). https://doi.org/10.1556/ JPC.17.2004.5.5 Hofmann, T., Albert, L., Rétfalvi, T., Visi-Rajczi, E., Brolly, G.: TLC analysis of the in vitro reaction of Beech (Fagus sylvatica L.) wood enzyme extract with catechins. J. Planar Chromatogr. 21, 83–88 (2008). https://doi.org/10.1556/jpc.21.2008.2.2 Hofmann, T., Guran, R., Zitka, O., Visi-Rajczi, E., Albert, L.: Liquid chromatographic/mass spectrometric study on the role of Beech (Fagus sylvatica L.) Wood Polyphenols in Red Heartwood Formation. Forest 13, ID 10 (2022). https://doi.org/10.3390/f13010010 Hon, D.N.S., Ifju, G.: Measuring penetration of light into wood by detection of photo-induced free radicals. Wood Sci. 11(2), 118–127 (1978) Kacikova, D., Kacik, F., Cabalová, I., Durkovic, J.: Effects of thermal treatment on chemical, mechanical and colour traits in Norway spruce wood. Biores. Technol. 144, 669–674 (2013). https://doi.org/10.1016/j.biortech.2013.06.110 Kamperidou, V., Barboutis, I., Vasileiou, V.: Response of colour and hygroscopic properties of Scots pine wood to thermal treatment. J. Forest. Res. 24(3), 571–575 (2013). https://doi.org/10. 1007/s11676-013-0389-y Kollmann, F., Keylwerth, R., Kübler, H.: Verfaerbungen des Vollholzes und der Furniere bei der künstlichen Holzrackung. Holz Roh Werkstoff 9(10), 382–391 (1951) Lee, S.H., Ashaari, Z., Lum, W.C., Halip, J.A., Ang, A.F., Tan, L.P., Chin, K.L., Tahir, P.M.: Thermal treatment of wood using vegetable oils: a review. Const. Bulid. Mat. 181, 408–419 (2018). https:// doi.org/10.1016/j.conbuildmat.2018.06.058 Liu, X.Y., Timar, M.C., Varodi, A.M., Sawyer, G.: An investigation of accelerated temperatureinduced ageing of four wood species: colour and FTIR. Wood Sci. Technol. 21, 357–378 (2016). https://doi.org/10.1007/s00226-016-0867-4 Lo Monaco, A., Pelosi, C., Agresti, G., Picchio, R., Rubino, G.: Influence of thermal treatment on selected properties of chestnut wood and full range of its visual features. Drewno-Wood 63, 5–24 (2020). https://doi.org/10.12841/wood.1644-3985.344.10 Meints, T., Teischinger, A., Stingl, R., Hansmann, C.: Wood colour of central European wood species: CIELAB characterisation and colour intensification. Eur. J. Wood Prod. 75, 499–509 (2017). https://doi.org/10.1007/s00107-016-1108-0

References

155

Mikleˇci´c, J., Jirouš-Rajkovi´c, V.: Influence of thermal modification on surface properties and chemical composition of beech wood (Fagus sylvatica L.). Drvna Industrija 67(1), 65–71 (2016). https://doi.org/10.5552/drind.2016.1520 Mitsui, K.: Changes in the properties of light-irradiated wood with heat treatment. Part 2. Effect of light-irradiation time and wavelength. Holz Roh Werkstoff 62, 23–30 (2004). https://doi.org/ 10.1007/s00107-003-0436-z Molnar, S., Bariska, M.: Wood Species of Hungary. Szaktudás Kiadó Ház, Budapest (2002) Molnar, S., Tolvaj, L., Nemeth, R.: Holzqualität und Homogenisierung der Farbe von Zerreiche (Quercus cerris L.) mit Dämpfung. Holztechnologie 47(5), 20–23 (2006) Nuopponen, M., Vuorinen, T., Jamsa, S., Viitaniemi, P.: The effects of a heat treatment on the behaviour of extractives in softwood studied by FTIR spectroscopic methods. Wood Sci. Technol. 37, 109–115 (2003). https://doi.org/10.1007/s00226-003-0178-4 Preklet, E., Tolvaj, L., Banadics, E.A., Alpar, T., Varga, D.: Colour modification and homogenisation of larch wood by steaming. Wood Res. 64(5), 811–820 (2019) Preklet, E., Tolvaj, L., Tsuchikawa, S., Varga, D.: Colour modification of wood by dry thermal treatment between 90 °C and 200 °C. Acta Silv. Lign. Hung. 19(1), 9–20 (2023). https://doi. org/10.37045/aslh-2023-0001 Sanz, M., Fernández de Simon, B., Esteruelas, E., Munoz, A.M., Cadahia, E., Hernández, T., Estrella, I., Pinto, E.: Effect of toasting intensity at cooperage on phenolic compounds in Acacia (Robinia pseudoacacia) Heartwood. J. Agric. Food Chem. 59, 3135–3145 (2011). https://doi. org/10.1021/jf1042932 Sikora, A., Kacik, F., Gaff, M., Vongrová, V., Budeniková, T., Kubovsky, I.: Impact of thermal modification on color and chemical changes of spruce and oak wood. J. Wood Sci. 64, 406–416 (2018). https://doi.org/10.1007/s10086-018-1721-0 Sundqvist, B., Karlsson, O., Westermark, U.: Determination of formic-acid and acetic acid concentrations formed during hydrothermal treatment of birch wood and its relation to colour, strength and hardness. Wood Sci. Technol. 40, 549–561 (2006). https://doi.org/10.1007/s00226-0060071-z Stamm, A., Hansen, L.: Minimizing wood shrinkage and swelling: effect of heating in various gases. Ind. Eng. Chem. 29(7), 831–833 (1937) Teischinger, A., Zukal, ML., Meints, T., Hansmann, C., Stingl, R.: Colour characterization of various hardwoods. In: The 5th Conference on Hardwood Research and Utilization in Europe, Sopron, pp. 180–188, 10–11 Sept 2012 Tieman, H.D.: The effect of different methods of drying on the strength of wood. Lumber World Rev. 28(7), 19–20 (1915) Timar, M.C., Varodi, A., Hacibektasoglu, M., Campean, M.: Color and FTIR analysis of chemical changes in Beech Wood (Fagus sylvatica L.) after Light steaming and heat treatment in two different environments. BioResources 11(4), 8325–8343 (2016) Tjeerdsma, M., Boonstra, M., Pizzi, A., Tekely, P., Millitz, H.: Characterisation of thermally modified wood: molecular reasons for wood performance improvement. Holz Roh Wekstoff 56, 149–153 (1998) Tolvaj, L., Nemeth, R., Varga, D., Molnar, S.: Colour homogenisation of beech wood by steam treatment. Drewno-Wood 52, 5–17 (2009) Tolvaj, L., Molnar, S., Nemeth, R., Varga, D.: Color modification of black locust depending on the steaming parameters. Wood Res. 55(2), 81–88 (2010) Tolvaj, L., Papp, G., Varga, D., Lang, E.: Effect of steaming on the colour change of softwoods. BioResources 7(3), 2799–2808 (2012) Tolvaj, L., Preklet, E.: A faanyag színváltozása nedvesítés hatására. [Colour change of wood by wetting.] Faipar 63(1), 41–46 (2015) Tolvaj, L., Banadics, EA., Tsuchikawa, S., Mitsui, K., Preklet, E.: Color modification and homogenization of sugi wood by steaming. Asian J. Fores. 3(1), 20–24 (2019). https://doi.org/10.13057/ asianjfor/r030103

156

3 Applications of Colour Measurement in Wood Research

Varga, D., van der Zee, M.E.: Influence of steaming on selected wood properties of four hardwood species. Holz Roh Werkstoff 66(1), 11–18 (2008). https://doi.org/10.1007/s00107-007-0205-5 Viitaniemi, P., Jämsä, S., Viitanen, H.: Method for improving biodegradation resistance and dimensional stability of cellulosic products. United States Patent Nº5678324 (1997) Windeisen, E., Stroble, C., Wegener, G.: Chemical changes during the production of thermotreated beech wood. Wood Sci. Technol. 41, 523–536 (2007). https://doi.org/10.1007/s00226007-0146-5 Wikberg, H., Maunu, S.L.: Characterisation of thermally modified hard- and softwoods by CPMAS NMR. Carbohydr. Polym. 58, 461–466 (2004). https://doi.org/10.1016/j.carbpol.2004.08.008 Xin, Y.L., Timar, M.C., Varodi, A.M., Sawyer, G.: An investigation of accelerated temperatureinduced ageing of four wood species: colour and FTIR. Wood Sci. Technol. 51, 357–378 (2017). https://doi.org/10.1007/s00226-016-0867-4 Zanuncio, A.J.V., Carvalho, A.G., de Souza, M.T., Jardim, C.M., Carneiro, A.C.O., Colodette, J.L.: Effect of extractives on wood color of heat treated Pinus radiata and Eucalyptus pellita. Maderas, Ciencia Technol. 17(4), 857–864 (2015). https://doi.org/10.4067/S0718-221X2015005000074

Chapter 4

Monitoring of Wood Photodegradation by Colour Measurement

Abstract The chapter presents how effective the colour measurement is in monitoring the sensitivity of different wood species to photodegradation. The colour measurement mistakes during outdoor weathering test are discussed. Effects of outdoor and indoor sun radiation tests are compared. Recommendations are given regarding the use of artificial light sources to study the photodegradation properties of wood. It is demonstrated that colour is an excellent parameter to monitor the effect of air temperature and humidity during photodegradation. Experimental results show that photodegradation is a highly temperature-dependent process. The effect of water leaching on photodegraded wood can be followed by colour change monitoring. Colour measurement proved to be a proper method for determining the photodegradation properties of thermally modified wood. Keywords Wood · Weathering · Sun radiation · Yellowing · Hemicelluloses · Extractives · Ultraviolet radiation

4.1 Introduction Changes during photodegradation can be monitored by colour measurement. Extractives are the most sensitive chemical components of wood to light exposure. Even one hour sunlight exposure can cause a measurable colour alteration in the case of wood with a high extractive content. It is difficult to identify individual chemical changes using colour measurement, but the effects of the influencing parameters (irradiation intensity, temperature, leaching effect of rain) can be monitored by colour measurement. Photodegradation of wood is generated by the photons of visible and UV light that have enough energy to split chemical linkages. Bonding energy of different chemical linkages are well defined values (Hon 2001). Individual wavelengths can be assigned to the bonding energies of the chemical bonds existing in wood. Figure 4.1 presents the bonding energies of different chemical linkages together with the wavelength of photons that have equal energy. It is important to know, however, that photons with

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 L. Tolvaj, Optical Properties of Wood, Smart Sensors, Measurement and Instrumentation 45, https://doi.org/10.1007/978-3-031-46906-0_4

157

158

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.1 Bonding energies of different chemical linkages and the wavelengths of photons that have the same energy

a wavelength shorter than indicated in Fig. 4.1 can also split the relevant chemical bond. Photodegradation causes chemical changes in a thin surface layer. Hon and Ifjú (1978) studied the penetration depth of UV and visible light by detecting the free radicals generated by the photodegradation. It was found that UV light penetrates up to a thickness of 75 µm in wood, while visible light penetrates deeper layers, up to 200 µm. Dirckx et al. (1987) also studied the penetration depth of UV light in wood by using thin wooden filters with thickness between 25 and 100 µm. Wood samples behind the filters were monitored by IR measurement. The results showed that thin wooden filters did not transmit UV light if the filter thickness reached 80 µm. These results indicate that reflected and scattered light bring information from the same layer where photodegradation occurs. Consequently, colour measurement can provide information on photodegradation.

4.2 Outdoor Weathering Test of Wood Outdoor weathering test is a practical rather than a scientific procedure. It is usually some kind of durability test. Outdoor weathering modifies many parameters of the examined wooden material. Here only the colour change will be discussed. The outdoor weathering test is completely different from the sun radiation test. These are usually long-lasting tests (minimum one year) presenting the effects of all influencing weather elements including sunlight. Therefore, the result of such investigations will always provide some kind of average values. The real disadvantage

4.2 Outdoor Weathering Test of Wood

159

of this type of test is that it is not exactly repeatable and it is unsuitable to analyse short periods where the effect of only one single influencing factor is dominant (e.g., sunny days, cloudy days, rainy days). The usual measuring period is one month. It is recommended to remove the samples after a sunny day for colour measurement. Plus-minus one day to the desired monthly measurement day does not generate real distortion. It is very important to perform the colour measurement in equal moisture content condition of the samples in any cases. Ignoring this rule, higher moisture content might generate greater colour differences than the weathering did (see Sect. 3.4). (Reading the literature of outdoor weathering, it seems that some authors did not pay attention to this problem.) Ideally, the samples should be stored in normal climate for several days to ensure equal wood moisture content before colour measurement, but this relatively long period interrupts the continuous weathering. Experiments proved that a one-day drying right before the colour measurement can generate roughly equal moisture content on the sample surfaces. However, the drying temperature must be not higher than 50 °C. The literature of weathering of wood includes plenty of papers and books. There are numerous review papers as well. The two most recent review articles listing more than 400 papers were published by Kránitz et al. (2016) and Kropat et al. (2020). The results of an outdoor weathering test are presented here according to Csanady et al. (2015). The purpose of the test was to determine the colour stability of steamed black locust during outdoor weathering. Black locust samples were steamed at 95 °C for 4 days. Unsteamed black locust and oak samples were put out beside steamed black locust at the main campus of the University of Sopron, Hungary. The geographical data of Sopron is 47° 40 min latitude. Samples were placed on a metal stand facing south at an angle of 45° to the horizontal and at a height of 2 m from the ground. The sample size was 1000 × 100 × 20 (mm). Specimens were put out at the beginning of August. The test duration was 2 years. Colour measurement was carried out monthly. The specimens were removed from the metal stand at the end of a sunny day. (Sometimes the monthly period was one day shorter or longer depending on the weather conditions.) Specimens were dried in a chamber at 40 °C for a day right before colour measurement. This treatment did not modify the colour of the samples but generated equal wood moisture content on their surface before the colour measurement. Figure 4.2 presents the lightness alterations of steamed black locust, unsteamed black locust and oak samples during two-year outdoor weathering. The main change was the lightness decrease up to the end of the first year. The lightness value hardly changed during the second year of exposure of the examined species. The hectic change of lightness values was generated by the different rates of sunny and rainy days during the monthly tested exposure periods. The photodegradation reduced the lightness values (Tolvaj and Faix 1995; Pastore et al. 2004; Pandey 2005a; Cogulet et al. 2016; Liu et al. 2019). The leeching effect of rain partly increased the lightness values (Hikita et al. 2001; Kannar et al. 2018). During the one-month periods there was both sunshine and rain. During the first sunny month (August), the lightness value of black locust decreased while that of oak and steamed black locust increased showing the great species dependence of the lightness alteration. In the

160

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.2 Lightness change of specimens during two-year outdoor weathering

rainy September, an increase in lightness was observed for all species. This lightness increase continued for black locust in October, while steamed black locust and oak showed lightness decrease. The steamed black locust presented opposite type of lightness change than oak and black locust in November. These alterations present that outdoor weathering do not give precise information regarding lightness change in case of a relatively short testing period. The reason is that the individual wood species show different leaching and photodegradation properties. These deviations are dominant in the first period of the outdoor weathering which was 8 months in our case. The changes were smaller and more balanced as the treatment time progressed. Figure 4.3 shows the movement of the colour dots on the a*–b* plane during the two-year long outdoor weathering. Starting points are marked by filled symbols. These initial colour dots of the species are far from each-other. The biggest difference was found in the redness of the tested samples. Natural black locust had yellow tint represented by a hue value of 86°. Natural oak and steamed black locust had brown tint with 70° and 64° hue values. During the sunny August, black locust presented great redness increase generated by the photodegradation. Its hue value turned from 86° to 73°. The extractives were degraded by the light irradiation, but lignin was hardly affected. The extractives were partly able to protect lignin against photodegradation. This phenomenon is represented by the small yellowness increase. On the other hand, oak showed mainly yellowness increase generated by the degradation of lignin. Steamed black locust produced only small redness and yellowness increase during the first month of outdoor exposure. Black locust underwent great redness increase during the steam treatment modifying the extractives responsible for its red colour. It is the reason why the photodegradation did not cause considerable redness increase for steamed black locust. The colour hue of the three species got more closer to each other after the first month of the exposure already. The rain leached out partly the degradation products of extractives and lignin reducing both redness and yellowness

4.2 Outdoor Weathering Test of Wood

161

values during September. Steamed black locus showed the most intensive decrease. Both redness and yellowness values decreased continuously during the next months. This tendency slowed down and continued up to the end of the first year of exposure representing that the leaching effect of rain was dominant after the first month of outdoor exposure. The saturation value of the specimens increased during the first month of exposure and decreased dramatically afterward. The results show that all the extractives involved in the formation of wood colour disappeared from the surface layer together with the coloured degradation products of lignin. The colour dots are located close to the origin of the coordinate system representing that the colour turned to gray after one year exposure. No real change in the wood colour occurred during the second year of the outdoor weathering. Similar results were found by Kubovsky et al. (2018) during the 2-year outdoor exposure of oak, black locust, poplar, alder and maple species. Figure 4.4 shows the hue change of oak, black locust and steamed black locust specimens during 2-year outdoor exposure. Colour hue of steamed black locust turned continuously from brown to tawny-brown, less saturated shade until the end of May of the following year. (There was no change during wintertime.) The reason was the photodegradation of lignin together with the leaching effect of rain. Rain leached out both steaming modified and unmodified extractives to some extent, pushing the colour towards a garish shade (Fig. 4.3). Colour hue change of oak followed the tendency of that of steamed black locust, only the lignin degradation was a little more intensive until November and some hue value decrease occurred during the winter. Black locust also followed the hue value change of steamed black locust except for the first two months. This two-month weathering was enough to equalise the modification effect of steaming.

Fig. 4.3 The yellowness and redness change of the specimens during two-year outdoor weathering. Filled symbols represent the starting points of the treatment. Constant hue (h*) values are also indicated

162

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.4 Hue change of specimens during two-year outdoor weathering

During the second sunny summer (June-July) the photodegradation of extractives reduced the hue values turning the hue towards brown tint for all investigated samples. After degrading most part of the extractives and lignin in the surface layer, the leaching effect of rain became dominant. Outdoor weathering test shows the integrated results of all the factors that affect the colour change of wood during outdoor exposure. That is why the outdoor tests durations are long periods. The results presented above show that most changes occurred during the first year of exposure to natural wooden surfaces. Similar results were found by Hikita et al. (2001) The test period can be much longer if the surface of the samples is impregnated or finished. (These types of tests are not discussed here.)

4.3 Colour Change of Wood During Photodegradation The initial colour of wood is mainly determined by the extractive content (Romagnoli et al. 2013). Colour measurement is a highly sensitive method for monitoring the changes during photodegradation. Even one hour sunlight exposure creates measurable colour alteration for black locust. Light exposed wood starts being darker and alters in tone. Extractives and lignin play the main role in colour alteration of wood during light exposure. The degradation of lignin and extractives, followed by oxidation processes, creates new chromophore chemical groups (Feist and Hon 1984). It is expected that the extent of these changes depends on the amount of lignin as well as the type and concentration of extractives. There are several types of extractives in wood such as lipids, phenolic compounds, terpenoids, fatty acids, resin acids, steryl

4.3 Colour Change of Wood During Photodegradation

163

esters, sterols and waxes (Shebani et al. 2008). Phenolic components are the most sensitive to light irradiation (Zahri et al. 2007; Chang et al. 2010). The intensity of colour change is related to the intensity of light exposure and highly determined by the wavelength distribution of incident light. The discoloration is also species dependent. The rate of discoloration is influenced by the environmental conditions such as temperature, presence of oxygen and water. The colour of an object is determined by the conjugated double bond chemical systems in it. The colour alterations are generated by the changes of these chemical systems. In terms of colour change, the effect of sunlight is the most interesting from scientific and practical point of view especially the time dependence of the caused changes. The intensity of the colour change is high from the beginning of the light radiation and it decreases with the passing time (Tolvaj and Faix 1995; Pastore et al. 2004; Pandey 2005a; Wang and Ren 2008; Sharratt et al. 2009; Cogulet et al. 2016). It is important to know that the intensity of the sun radiation itself is not constant either but changes during the day and throughout the year as well. Sun radiation is much more intensive in summer than in winter and the wavelength distribution is also different. Furthermore, in summer, sunlight contains additional UVB radiation. The daily sun radiation energy in June is 7 times higher than in December (in Hungary). The radiation intensity also depends on where the test position is located on the earth’s surface. These circumstances must be carefully considered during the planning of sun radiation tests to minimize the effects of the above-mentioned alterations. Ignoring the alterations, the result will be some kind of average not real time dependence. When planning the tests, it is first necessary to decide whether a summer or a winter test will be performed as the results of these two test types are completely different. The effect of temperature is also different comparing the two seasons. The temperature of the sun radiated wooden surface can reach 70 °C in July if the surface is dark (e.g. steamed black locust, measured in Hungary). This temperature is high enough to multiply the effect of photodegradation (Persze and Tolvaj 2012). Air humidity is a factor which is altering during different seasons at continental climate. The centre of testing periods should be the end of June or end of December to ensure the smallest possible alteration of radiation intensity. It is not recommended to plan test periods longer than 6 months. Samples must be exposed only on sunny days and set aside every day. The treatment time should not be longer than 8 h with midday centre as the intensity and wavelength distribution of sun radiation decreases intensively away from the midday peak. The most effective short wavelengths are missing in the early morning and late afternoon and the sun radiation intensity is 2.3 times higher at noon than at 8 am (in April in Hungary). Examining the colour change of wood materials, the effect of sunlight can be interesting under two conditions, outdoor and indoor (behind window glass). The indoor test can be imitated outdoor by covering the sample surface with a glass sheet, although the temperature and air humidity conditions will not be the same as in a closed room. In the following, the results of an outdoor sun radiation test will be presented. The test was carried out in the city of Takayama, Japan, during the summer. Geographical data for Takayama are 36° 9.3 min latitude and 560 m altitude. The test period

164

4 Monitoring of Wood Photodegradation by Colour Measurement

was between the 5th of May and the 19th of August. The air temperature varied between16 and 41 °C, max. r.h. was 80%, and the daily average of total solar power density was between 436 and 459 W/m2 . Samples were outside between 9 am and 3 pm, but only in sunny days. The investigated hardwood species were beech (Fagus crenata), earlywood part of black locust (Robinia pseudoacacia) heartwood and poplar (Populus cauesceus) earlywood. Softwood samples included earlywood part of Sugi (Japanese cedar) (Cryptomeria japonica D. Don) sapwood and spruce (Picea abies) earlywood. The specimens were of Japanese origin (except for the black locust grown in Hungary). Planed surfaces with a tangential orientation were prepared for colour measurement. The results are presented here according to a previously published paper (Tolvaj and Mitsui 2005). The investigated species produced rapid colour alteration (including L*, a* and b* coordinates) during the first 30-h outdoor exposure and this tendency slowed down with elapsed time (Figs. 4.5, 4.6 and 4.7). Most of the lightness decrease occurred in the first 20 h of treatment (Fig. 4.5). This rapid decrease accounted for 74% and 65% of the total lightness decrease for black locust and beech, respectively. Poplar was an exception to this, given that its lightness slowly decreased during the first 20 h of sun radiation, and this trend remained until the end of the treatment. In the first 30 h of the treatment, conifers produced similar lightness decrease as black locust and beech, however, the change during this period was only 40% of the total change. The reason is that conifers produced continuous lightness decrease until the end of the treatment. In contrast, hardwood species (except poplar) presented slow lightness decrease after 20-h sun radiation. The extraordinary behaviour of poplar (which has low extractive content) confirms the hypothesis that extractives are responsible for the rapid colour alteration at the beginning of photodegradation (Pandey 2005b). Figure 4.6 shows the redness change generated by sunlight exposure. Most of the samples showed unusual redness decrease during the first 5 h of the treatment.

Fig. 4.5 Lightness change of the investigated species generated by outdoor sun radiation

4.3 Colour Change of Wood During Photodegradation

165

Fig. 4.6 Redness change of the investigated species generated by outdoor sun radiation

Fig. 4.7 Yellowness change of the investigated species generated by outdoor sun radiation

Artificial light irradiation never produces similar alteration. (The first part of test was repeated in August, and the repeated sun radiation showed the same results.) This redness decrease shows that sun radiation degraded those chromophore groups that are inherently in wood and are responsible for the initial redness value of the species. The only exception was black locust which produced highly intensive redness value increase at the beginning of the treatment. Probably, some of the initial chromophore groups were also degraded, however, the number of newly formed chromophore groups was much higher than those that lost their chromophore structure at the beginning of sun radiation. These results lead to the conclusion that sun radiation can generate new chromophore chemical groups, but at the same time it can damage

166

4 Monitoring of Wood Photodegradation by Colour Measurement

the initial chromophore groups responsible for the redness intensity. The measured results show that the dominant change was the creation of new chromophore groups after 5 h of irradiation. Beech and black locust presented rapid redness increase during the 30-h exposure followed by moderate increase afterward. The other species showed continuous and close to linear redness increase. Poplar samples produced the smallest redness increase during the 200-h sunlight exposure. Figure 4.7 shows the yellowness change generated by sunlight exposure. Sun radiation caused more uniform yellowness alteration than redness alteration. The yellowing was intensive during the first 30-h exposure, followed by moderate increase between 30 and 120 h of sun radiation. The last 60 h of the exposure generated barely visible changes. The only exception was poplar, whose yellowness did not change during the first 10 h of irradiation, although after that it followed the trend of other species. The yellowness increase during light irradiation is mainly attributed to the degradation products of lignin. The absorbed photons can split the aromatic ring of lignin generating free radicals. These radicals react whit oxygen producing conjugated and unconjugated carbonyl groups (mainly quinonoid and phenolic structures) (Tolvaj and Faix 1995; Müller et al. 2003; Pandey 2005a; Timar et al. 2016). It is interesting to mention that the dots representing the 60-h irradiation do not fit exactly to the trendlines in Figs. 4.5, 4.6 and 4.7. The measured lightness values are somewhat higher, while the redness and yellowness values are slightly lower than expected. The reason is that the period of the second 30 h sunlight radiation was highly humid. The high humidity slightly reduced the a* and b* values and generated some lightening. Figure 4.8 shows the locations of colour dots on the a*–b* plane during the 200h outdoor sun radiation. Filled and enlarged symbols represent the starting points belonging to the untreated samples. These dots are followed by the dots of irradiated samples after 5-, 10-, 20-, 30-, 60-, 120- and 200-h exposure. This figure represents the change of hue and chroma values as well. The chroma values increased considerably during the treatment. The alteration of hue values was highly species dependent. Poplar showed the smallest hue alteration. Its hue values slightly increased during the first 60 h of the treatment and decreased over time. The final colour tone was the same as the initial one. Beech presented small hue value decrease (from 76° to 72°). This change was demonstrated by small colour shift towards brown tint. The lowdensity species are characterised by curved trendlines demonstrating that yellowing was dominant in the first quarter of the treatment and the redness increase became dominant thereafter. Finally, the difference between the starting and ending points was only 8° towards brown tint. The greatest hue value difference was generated by black locust (10°). The trendline of hue for black locust is close to linear presenting the proportional redness and yellowness increase during the outdoor sun radiation. The colour modification properties of indoor sunlight exposure are of both practical and scientific interest. Window glass works as a cut-off filter. It strains of the photochemically most active UV part of sun radiation. The transmission spectrum of double glass layer is presented in Fig. 4.9. The window glass does not transmit the UVB portion of sun radiation and filters also the shorter wavelength region of

4.3 Colour Change of Wood During Photodegradation

167

Fig. 4.8 The yellowness and redness change of the specimens during the 200-h outdoor sun radiation. Filled and enlarged symbols represent the starting points of the treatment. Constant hue (h*) values are also indicated

UVA. It means that the cleavage of C-O linkage occurs only rarely behind window glass. This cleavage needs photons with wavelengths shorter than 344 nm. There are two main approaches to carry out sun radiation test behind window glass. The traditional one is to place the samples indoor just behind the window. Another technic is to settle the samples outdoor with glass cover (Liu et al. 2022). Although precipitation and air humidity change must and could be excluded in the second

Fig. 4.9 Transmittance spectrum of doubled window glass

168

4 Monitoring of Wood Photodegradation by Colour Measurement

case, it is impossible to prevent temperature alterations outdoor. The absorbed sun radiation can generate heat which raises the surface temperature (glasshouse effect). This temperature increase does not occur during indoor testing, as the natural air ventilation in a spacious room carries most of the generated heat away from the surface of the samples. Indoor sun radiation test was carried out to determine the colour modification effects of sunlight on different wood species behind double glassed window. The test was carried out in a building of the University of Sopron, Hungary. The geographical data of Sopron is 47° 40 min latitude. The samples were put behind the window only in sunny periods during May and June. The chosen window faced towards south. The total irradiation time was 200 h. The selected species were black locust heartwood (Robinia pseudoacacia), poplar (Populus x euramericana cv. Pannonia), oak heartwood (Quercus petraea), Scots pine sapwood (Pinus sylvestris) and spruce (Picea abies) The radial surface of the specimens was used for colour measurement. The results of indoor sun radiation are visible in Figs. 4.10, 4.11 and 4.12. Interestingly, the rapid colour change at the beginning of the treatment is partially or completely missing compared to that observed during the outdoor sun radiation (Figs. 4.5, 4.6 and 4.7). Figure 4.10 shows the lightness decrease generated by indoor sunlight exposure. Conifer species presented close to linear lightness decrease during the whole examination period. Deciduous species showed more intensive lightness decrease than conifer species during the first 20 h of the treatment. The lightness decrease was similar for all investigated species between the 20th and 90th hours of the exposure. There was no real lightness decrease for deciduous species during the second half of the treatment. Comparing the darkening effect of sun radiation under outdoor and indoor conditions, two main differences can be observed. The lightness decrease was less rapid

Fig. 4.10 Lightness change of the investigated species generated by indoor sun radiation

4.3 Colour Change of Wood During Photodegradation

169

Fig. 4.11 Redness change of the investigated species generated by indoor sun radiation

Fig. 4.12 Yellowness change of the investigated species generated by indoor sun radiation

in the case of indoor exposure and the total lightness decrease was greater in the case of outdoor exposure. Figure 4.11 shows the redness change caused by sun radiation behind window glass. Most samples showed redness decrease during the first 5 h of the treatment (except black locust). This decrease was small for poplar and oak. Scots pine showed the greatest redness value decrease and this decrease was continuous during the first 20 h of the treatment. The redness values showed continuous increase after the first decreasing period for all investigated species during the whole treatment period. This increase was more intensive for conifers than for deciduous species.

170

4 Monitoring of Wood Photodegradation by Colour Measurement

There is no significant difference between the redness modification effects of inand outdoor sun radiations only the total redness changes were greater during the outdoor exposure These differences are 16% and 23% for black locust and for spruce samples. Figure 4.12 shows the yellowness change induced by the indoor sun radiation. Although, photodegradation usually generates rapid yellowness increase at the beginning of treatment, sun radiation behind window glass produced only moderate yellowness increase during the first 20 h of the irradiation. Moreover, black locust produced yellowness decrease during this period. Spruce showed the greatest yellowness increase (6 units) in this first period. The rapid increase was only visible between the 20th and 40th hours of the sun irradiation. All trendlines showed slow but linear and continuous yellowness increase after the 50th hour of sunlight irradiation. Conifers produced more intensive increase than deciduous species. In the last three-quarters of the treatment, poplar showed an increase in yellowness similar to conifers. The yellowness increase was less rapid in the case of indoor exposure than during the outdoor exposure, and the total yellowness increase was greater during the outdoor testing. These differences are 269% for black locust and 18% for spruce samples. The reason was the lack of the rapid yellowness increase at the beginning of the treatment. The results confirm that extractives are responsible for the rapid yellowness increase during photodegradation (Sharratt et al. 2009). The high robinetin type extractives content of black locust was slightly decreased during the first 20 h of the irradiation reducing the yellowness value. Previous investigations demonstrated that extractives can partly protect the lignin molecules against photodegradation (Nemeth et al. 1992; Pastore et al. 2004; Tolvaj and Varga 2012). As the indoor sun radiation did not modify the high flavonoid content of black locust, these extractives were able to protect the lignin content, reducing the yellowness increase. This does not happen during outdoor sun exposure. That is why a huge difference in yellowness change was observed between indoor and outdoor sun radiation in the case of black locust. Figure 4.13 shows the colour dots of the investigated samples on the a*–b* plane during the 200-h indoor sun radiation. Filled and enlarged symbols represent the starting points belonging to the untreated samples. These dots are followed by the dots of irradiated samples after 8, 20, 40, 90 and 200 h of exposure. This figure also represents the change of hue and saturation values. The saturation values increased considerably during the treatment represented by the increasing distance between the colour dot and the origin of the coordinate system. The alteration of hue values was species dependent. Oak and poplar showed the smallest hue alteration. Their hue values hardly changed during the whole treatment. The final colour tints were the same as the initial ones. Colour dots of the investigated conifers form curved trendlines demonstrating that yellowing was determinant in the first quarter of the treatment, while the redness increase became dominant thereafter. Finally, in terms of hue, the difference between the start and end points was only 5° for spruce and 3° for Scots pine towards brown tint. The greatest hue value difference was generated by black locust (12°). The trendline of hue for black locust was slightly curved because of both yellowness decrease and increase during sun radiation.

4.3 Colour Change of Wood During Photodegradation

171

Fig. 4.13 Yellowness and redness change of the specimens during 200-h indoor sun radiation test. Filled and enlarged symbols represent the starting points of the treatment. Constant hue (h*) values are also indicated

Based on the comparison of the effects of outdoor and indoor sun radiations, it can be concluded that wavelength distribution of the applied irradiation is a key factor during photodegradation. Pure sun radiation tests try to minimize the effects of the additional parameters, such as rainfall or cloudy days. However, the effect of air humidity and temperature change could not be eliminated during outdoor exposure. Moreover, the intensity of sun radiation is continuously changing during a day (increasing and decreasing with midday centre). Accurate and repeatable weathering tests can only be carried out under artificial conditions, where the untested modifying effects can be kept constant. One of the greatest challenges is to choose the proper artificial light source. The light sources are typified based on their emission spectrum. The emission intensity distribution of the light source must fulfil the desired conditions for correct imitation. In order to simulate solar radiation, the emitted intensity distribution must be as close as possible to the emission spectrum of sun radiation perceived on the Earth’s surface. None of the artificial light sources emit exactly the same intensity distribution as the sun, but xenon lamp has similar emission spectrum. The intensity distribution of xenon lamp can be adjusted with filters to simulate correctly the emission spectrum of sun in the visible wavelength region. The only drawback is the missing UV part, as xenon lamp hardly emits radiation below 350 nm. Normally, only the UVA portion of sun radiation, ranging from 315 to 380 nm, used to reach the sea level of the earth, however, the thinning of the ozone layer allows also the UVB radiation to achieve the ground level. Thus, the UVB part of sun radiation, which is between 280 and 315 nm, may also reach the ground level if the atmosphere conditions let it travelling through. The most important difference between sun radiation and xenon lamp radiation is that sun radiation can split the C-O linkage. This bond can be split

172

4 Monitoring of Wood Photodegradation by Colour Measurement

by photons with wavelengths of 343 nm or less. The xenon lamp does not emit such wavelengths. The partially missing UVA emission of xenon light sources prevents the correct simulation of sun radiation. Comparison of the photodegradation effects of different light sources is highly difficult. The correct comparison is hardly possible not only because of the differences in the emitted wavelengths, but because of the differences in the emitted intensity distributions as well. Cirule et al. (2022) studied the photodegradative effect of fluorescent, incandescent, warm and cool LED light sources. The results showed that fluorescent lamp caused the biggest effect followed by the incandescent lamp and LEDs. The wavelength dependence of chemical changes generated by the photodegradation is a highly important and interesting question. Researchers tried to solve this question by using series of filters between the light source and the irradiated sample (Derbyshire and Miller 1981; Andrady et al. 1992; Chang et al. 1999, 2000; Mitsui 2004; Zivkovic et al. 2014). It is important to exclude all other parameters which can modify the effect of the chosen wavelength interval. Only stable artificial light source is recommended to investigate the wavelength dependence of photodegradation. (The intensity of sun radiation is changing during a day and it is impossible to exclude the effects of temperature and humidity changes.) The emission intensity distribution (wavelength dependence) of the light source must be taken into account together with the transmissivity differences among the filters. It is important to know that one photon interacts with one electron during the contact. Consequently, the number of transmitted photons must be equal for all filter combinations (not the light intensity). These requirements show that the application of the cut-off filter method is extremely complicated. Fortunately, the development of laser physics can offer individual lasers covering the whole visible and UV region nowadays. Defocusing the laser beam relatively large surface area can be irradiated and this method allows also to modify the power density of the incident light. Measuring the power density, the number of incident photons can be calculated using the emitted wavelength. There were some preliminary researches published in the first decade of the twenty-first century regarding the study of the photodegradation properties of wood generated by (mainly UV) lasers (Papp et al. 2004, 2005; Barta et al. 2005; Mitsui et al. 2005, Pandey and Vourinen 2008; Preklet et al. 2012). Unfortunately, there were not enough types of lasers available to cover the whole UV and visible wavelength region at that time. Discovering the wavelength dependence of the photodegradation of wood with lasers can be a marvellous challenge for young scientists. There are some cases where the wavelength distribution of the light source is less important than in the case of sunlight imitation. It is not necessary to use sunlight for examining the species and temperature dependence of wood photodegradation. Other light sources emitting visible and UV light are also suitable for such investigation. The emission intensity of the light source must be stabile during long-term exposure. A possible light source can be the mercury vapour arc-discharge lamp. There are several types of mercury vapour lamps. The emission spectrum of mercury vapour contains narrow emission bands at 184, 254, 365, 405, 436, 546 and 578 nm. The

4.3 Colour Change of Wood During Photodegradation

173

emitted wavelengths and their intensities strongly depend on the pressure within the lamp. The low-pressure mercury lamp emits only at 184 and 254 nm, the other emission intensities are negligible. The medium-pressure mercury lamps emit in the 200–600 nm interval, however, the emission at 254, 405 and 436 nm are dominant. Increasing the pressure within the lamp, the emission in the UVC region will be less dominant while the emission in the visible region is increasing. High-pressure mercury lamps were the main sources for street lighting in the last decades of the twentieth century. However, the emission of these lamps in the visible region was below 20% of the total light emission (visible and UV). This disadvantage was altered by using an outer-bulb. The inner surface of the glass bulb was covered by phosphor which absorbed the UV radiation and emitted visible light. The highpressure mercury lamp with outer tube emits acceptable cool-white light and it is suitable for street-lighting. Further increasing the pressure in the lamp results in a broadening of the emission bands. The extra high-pressure (5–10 Bar) mercury lamp emits continuous emission spectrum with high peaks on top. The elevated pressure increases the visible light portion of the emission. Adding metal halides to mercury vapour produces even more emission in the visible light region. Discharge takes place in both mercury vapor and vaporized metals from the metal-halide components. This combination elevates the visible light portion above 50% and the UVC part mainly disappears. These types of lamps are mainly used in projectors. The high-pressure mercury lamp has two disadvantages: It takes about 5–6 min to give full output. The cooling time is also 5–6 min after switching off the lamp. The lamp cannot be switched on again within this period. All types of high-pressure and extra high-pressure mercury lamps are suitable to study the photodegradation properties of wood. Although, these lamps cannot imitate the sun radiation. Mercury lamps produce rapid changes which is an advantage in accelerated tests. Mercury lamps are stable light sources during long term exposures. (The extra high-pressure mercury lamps have considerably shorter life time than high-pressure mercury lamps.) The long lifetime of high-pressure mercury lamp is also an advantage when using for artificial light exposures. It is important to apply the same type and brand of lamps made by the same producer. The pressure in the lamps made by different producer might be different modifying the emitted intensity distribution. The experiments presented in this book applied the same (old) type mercury lamps since the beginning of the twenty-first century. (The individual lamps were replaced by a new one regularly.) This decision gives the possibility of comparison among experiments carried out during the last two decades. The irradiation conditions were the same in all cases. The total electric power of the applied double mercury lamps was 800 W and the samples were located 64 cm from the lamps. The UV radiation was 80% of the total (UV and visible) light emission (31% UVA, 24% UVB and 25% UVC).

174

4 Monitoring of Wood Photodegradation by Colour Measurement

4.4 Species Dependence of Colour Change Caused by Photodegradation The main chemical components (cellulose, hemicelluloses and lignin) of wood are colourless macromolecules only lignin has conjugated double bonded chemical groups. The degradation products of hemicelluloses can be coloured by different chemical processes. The concentration and types of extractives determine the natural colour of different wood species. There are extractives being naturally coloured and others become coloured after outer effects such as light and heat. The light absorption followed by oxidation and dehydration generates chromophore structures mainly quinoidal structures. Lignin as a main chemical component of wood is intensive light absorber. The UV light can split the aromatic ring of lignin. The produced free radicals react with oxygen to generate carbonyl and carboxyl groups. These double bonded groups can create chromophore conjugated double bond systems. The degradation of extractives and lignin determines the colour alteration of wood during photodegradation. Species used in carpenter and joinery industry were chosen to clarify the species dependence of colour change during photodegradation (Persze 2011). The investigated conifers were: larch (Larix decidua L.) heartwood, Scots pine (Pinus sylvestris L.) sapwood and spruce (Picea abies Mill.). The chosen deciduous species were: alder (Alnus glutinosa L.), ash (Fraxinus excelsior L.), beech (Fagus sylvatica L.), birch (Betula pendula Roth), black locust (Robinia pseudoacacia L.) heartwood, cherry (Prunus serotina Ehrh.), linden (Tilia cordata Mill.), maple (Acer pseudoplatanus L.), oak (Quercus petraea) heartwood, poplar (Populus x euramericana cv. Pannonia) and walnut (Juglans regia L.) heartwood. Additionally, steamed beech was also involved to the test. (Steaming temperature was 90 °C and the treatment time was one day.) Beech is often used in steamed version because a light steaming turns its greyish-white tone to a more attractive reddish-white tone. The radial surface of the samples was used for colour measurement. The irradiation chamber was equipped with a doubled mercury vapour lamp described at the end of the previous section. Constant 30 °C was kept in the chamber with 32–38% relative air humidity. All other important parameters were constant during the test except the treatment time which was increased up to 200 h. The wood colour was measured before treatment and after 8-, 20-, 40-, 90- and 200-h light irradiation. Three diagrams were used to present the changes of the colour coordinates of the investigated 15 species (Figs. 4.14, 4.15 and 4.16). All three diagrams of a figure have the same vertical scale for better comparison. Based on the results, the greatest differences among the species were found in the case of the a* redness coordinate. Figure 4.14 shows the redness alterations for all species as a function of irradiation time. The redness value increased continuously during the whole treatment period for all investigated species. There are differences in the alteration tendencies at the beginning and in the last three-quarter of the irradiation period. The species were classified into three groups according to the similarities of these changes. The first group covers the species that have high extractive content

4.4 Species Dependence of Colour Change Caused by Photodegradation

175

Fig. 4.14 Redness change of the investigated species generated by mercury vapour lamp

(black locust, cherry, oak and walnut) and steamed beech. These species produced rapid redness increase during the first 8 h of the irradiation. Black locust, which has the highest extractive content, presented 67% of total redness change during the first 8 h. These values were between 40 and 50% for the other species belonging to the first group. Species in this group showed slow redness increase after the first 40 h of irradiation. Steamed beech showed only a minor redness alteration because the steaming as pre-treatment generated great redness increase already. Species belonging to the second group (alder, ash, beech, linden, maple and poplar) showed moderate redness increase during the first 8 h of the artificial irradiation, which counted 20–34% of total redness change. These species showed continuous redness increase after the first 40 h of irradiation. All investigated conifer species (larch, Scots pine and spruce) and birch can be found in the third group. Redness increase of these species was not as rapid at the beginning of the light irradiation as that of the species in the other two groups. Their redness increase in the first 8 h was between 13 and 16% of the total redness change. Species in the third group showed close to linear redness increase after 20 h of irradiation. In general, the slope of the trendlines is different comparing the three groups. The trendlines in the first group are close to horizontal after 50-h treatment while conifers in the third group presented the steepest curves. The results confirm the hypothesis that the redness value increases rapidly at the beginning of light irradiation if the extractive content is high enough. Moreover, the type of the extractives is particularly important. Initially, larch and steamed beech

176

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.15 Yellowness change of the investigated species generated by mercury vapour lamp

had the highest redness value. Extractives responsible for the initial redness of larch seems to be significantly more stable against short-term light irradiation, but not stable against long-term light irradiation. In contrast, chemical compounds in steamed beech are rather stable against long-term light irradiation. The artificial irradiation induced yellowness change of the investigated species did not show that kind of diversity as the redness change showed. All species presented rapid yellowness increase during the first 8 h of the treatment followed by moderate and later on slow yellowness increase (Fig. 4.15). The yellowness increase was generated mainly by the oxidation products of the photodegraded lignin. Species in the second and third group showed similar yellowness change. Species in the first group had highly diverse initial yellowness values (between 16 and 28). Consequently, the trendlines were located relatively fare from each other. Larch showed the smallest yellowness alteration (13 units) during the 200-h light irradiation followed by that of black locust with 15 units. The yellowness alteration of the other species was around 20 units. Spruce and poplar produced the greatest yellowness increase (25 and 24 units). These findings demonstrate that the high extractive content (larch, black locust) partly protected the lignin against photodegradation. Spruce and poplar, which have the lowest extractive content among the investigated wood species, suffered the greatest yellowness alteration confirming the protective effect of extractives.

4.4 Species Dependence of Colour Change Caused by Photodegradation

177

Fig. 4.16 Lightness change of the investigated species generated by mercury vapour lamp

The lightness change of the species was similar to their redness change (Fig. 4.16). All species showed rapid darkening at the beginning of the treatment followed by moderate and later on slow lightness decrease. Steamed beech and walnut (the darkest species originally) presented slight lightening at the end of treatment. There are differences among the groups regarding the lightness change intensities. Species in the first group showed the greatest lightness decrease during the first 8 h of irradiation. These changes were between 63 and 69% of the total lightness change. Similar data for the second group are between 43 and 51%. Species in the third group presented the smallest percentages of change (37–39%) during the first 8 h of irradiation. The results of artificial light irradiation by mercury vapour lamp showed the importance of extractives (both quality and quantity) during the photodegradation of wood. The degradation of lignin was similar for all investigated species, however, the extractives in some of the selected species were able to protect partly the lignin molecules against the degradation effect of light irradiation.

178

4 Monitoring of Wood Photodegradation by Colour Measurement

4.5 Temperature Dependence of Colour Change During Photodegradation Chemical changes are usually temperature dependent. This temperature dependence can be described by the Arrhenius law (see Sect. 3.2). Although, temperature dependence of the colour change during the photodegradation of wood has mainly theoretical importance. However, the surface of a dark wood species can reach 70–80 °C outdoor and summertime. This temperature is high enough to modify the colour change during photodegradation (Persze and Tolvaj 2012). The effect of low temperature was also investigated. Mitsui and Tsuchikawa (2005) irradiated wood samples that were kept in a conditioning chamber set to − 40 °C. It was concluded that the degradation measured at − 40 °C was much less than under normal laboratory conditions. A comprehensive investigation was designed to study the colour modification of photo-irradiation at elevated temperatures. To determine the effect of temperature during photodegradation, samples were irradiated with a mercury lamp at 30, 80, 120 and 160 °C. Our experiments differed from Mitsui’s previous work (Mitsui et al. 2001, 2004a, b), because he applied separate UV irradiation and thermal treatment on after the other. In our case, these two types of treatments were simultaneous, as it happens in nature. The elevated temperature generates higher vibration energy so that the photons can split chemical bonds more easily than at ambient temperature. The investigated hardwood samples were: ash (Fraxinus excelsior L.) and poplar (Populus x euramericana cv. Pannonia), the softwood samples were: Scots pine (Pinus sylvestris L.) and spruce (Picea abies Karst.). The samples of different series were prepared from the same board to minimize the effect of wood inhomogeneity. Radial surface of the samples was used for colour measurement. Strong UV light emitter, mercury vapour lamp was used to irradiate specimens at four different air temperatures 30, 80, 120 and 160 °C. A series of samples were treated in the same chamber set for 30, 80, 120 and 160 °C but without light irradiation. The effect of pure thermal degradation was determined by this experiment. The results are presented here according to a previous paper (Tolvaj et al. 2015) with written permission from Springer Nature. The total treatment time was 16 h in all cases. This period was short for the slow changes but proved to be long for the fast changes. This second statement was confirmed by the data of UV irradiation at 160 °C. The exposure was interrupted after 6, 11 and 16 h to measure the colour. At the beginning of our study, the aim was to design experiments for separating the effect of UV radiation and the effect of thermal treatment. Pure thermal treatment can be carried out in total darkness, but light sources generate lots of heat beside light irradiation. The irradiation chamber can keep the desired temperature if equipped with proper air ventilation and heating. The results of pure thermal treatment at 30 °C did not show any measurable colour change during a 16-h thermal treatment. (This thermal treatment was extended for one week, but no significant changes were recorded after that.) This finding suggests that the effect of light irradiation at 30 °C is only photodegradation (without thermal degradation). This assumption allows the

4.5 Temperature Dependence of Colour Change During Photodegradation

179

Fig. 4.17 Relative lightness generated by pure UV radiation (UV30), pure thermal treatment at 160 °C (T160) and simultaneous UV and thermal treatment at 160 °C (UV160)

separation of the effect of photodegradation and thermal degradation. The relative lightness values of two wood species (ash and spruce) generated by pure UV radiation (UV30), pure thermal treatment at 160 °C (T160) and simultaneous UV and thermal treatment at 160 °C (UV160) are presented in Fig. 4.17. The 16-h pure UV radiation produced 9–10% lightness decrease, whereas the 16-h thermal treatment at 160 °C generated 16–18% lightness decrease. The lightness decrease caused by simultaneous UV irradiation and thermal treatment at 160 °C was 38–46%, that is much higher than the sum of the previous two treatments (sum of UV30 and T160). The discolouration effect of discrete and simultaneous UV and thermal treatments can be seen in Figs. 4.18, 4.21 and 4.22 represented by the alteration of the colour coordinates generated by 16-h treatment. The change of lightness coordinate (darkening) increased by elevating the temperature in the case of simultaneous treatments (Fig. 4.18). There was a moderate increase between 30 °C and 120 °C. The only exception is the Scots pine showing a more intensive increase (more than doubled) when comparing the treatment at 80 °C and at 120 °C. The lightness change of the other 3 species was also doubled but between 120 and 160 °C. There were no consequent differences in the lightness change among the species. Only ash was a little more sensitive to the treatment than the others. The samples in the totally dark chamber suffered considerably less darkening at 160 °C than the light irradiated samples at 160 °C. As the UV irradiation at 30 °C was identified as pure UV irradiation (without thermal effect), the first group of columns represents the effect of photodegradation in Figs. 4.18, 4.21 and 4.22. The group of columns marked T160 °C represents the effect of thermal degradation at 160 °C (without UV light irradiation). The last group of columns (UV30 + T160) represents the sum of respective values of these two groups of columns. Note that this group of columns does not represent a real treatment but the simple addition of the colour coordinate changes generated

180

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.18 Lightness decrease (as positive values) of the investigated wood species caused by UV irradiation at different temperatures and by dark thermal treatment (T) at 160 °C. The duration of all treatments was 16 h

by both photodegradation and thermal degradation separately. The darkening effect of thermal treatment at 160 °C was slightly greater than the effect of light irradiation (UV30). Interestingly, the sum of the lightness changes caused by these two separate treatments (UV30 + T160) is considerably smaller than the lightness change caused by the simultaneous light irradiation and thermal treatment (UV160). It means that the thermal effect during photodegradation is not only the simple addition of two effects but the elevated temperature multiplies the effect of photodegradation. The multiplying factor was 1.8 (average of the 4 investigated species) for lightness change at 160 °C (comparing thr walues of UV160 and UV30+T160). The representation by Fig. 4.18 does not allow deeper insight into the lightness and colour hue variations as a function of influencing factors. Due to the simultaneous action of light irradiation and heat, it is interesting to examine first their separate effect. Figure 4.19 shows the alteration of hue values together with the lightness change. The lightness values showed continuous decrease in all cases as it is represented in Fig. 4.17 as well. The hue value increased at the beginning of UV irradiation at 30 °C (UV30). This increase demonstrated the yellowing of the specimens. Slight shift towards brown tint (hue value decrease) was observed in the second half of the treatment period. Probably, both changes were present during the whole treatment but the yellowing was dominant at the beginning while the colour shift towards brown tint was dominant afterward. The UV irradiation at 120 °C (UV120) presented the dominance of brown shift. However, this shift was slowed down by the yellow shift at the beginning of the treatment. The UV irradiation at 160 °C (UV160) showed rapid hue value decrease during the whole treatment period. The main part of hue shift towards brown tint happened during the first 6 h of the treatment. This hue decrease was 68% of the total hue decrease. The temperature

4.5 Temperature Dependence of Colour Change During Photodegradation

181

of 160 °C was high enough to cause thermal degradation of hemicelluloses. This additional process increased the reduction of hue values. The thermal treatment in total darkness (T160) showed colour shift towards brown only. The thermal treatment generated only small hue decrease and lightness decrease during the 16-h treatment at 120 °C (T120). The hue decrease was only 1.5 unit for spruce and 1 unit for poplar. The thermal treatment at 160 °C produced much greater hue value decrease than at 120 °C. It is well-known that lignin is fairly stable at 160 °C. It means that the hue value decrease was generated by the degradation of extractives and hemicelluloses during thermal treatment. The slope of trendlines can be an indicator to differentiate among the treatments. The slope of a trendline can be determined by the quotient of hue change and lightness change (Δh*/ΔL*) generated by the 16-h treatment. The values of this quotient are presented in Table 4.1. The slope values of the pure UV treatment are one-tenths of those of the pure thermal treatment. This big deviation demonstrates the different nature of these two treatment types. The quotient values increased with increasing temperature during the UV irradiation.

Fig. 4.19 Lightness and hue changes caused by simultaneous UV irradiation and heat treatment (UV160 and UV120) as well as due to pure thermal treatment (T160 and T120) and pure UV irradiation (UV30). Filled symbols represent the starting points of the treatment followed by results of 6-, 11- and 16-h treatments

Table 4.1 The slope values of trendlines for the applied treatment types UV30

UV120

UV160

T160

Δh*/ΔL* Spruce

0.04

0.32

0.43

0.55

Poplar

0.03

0.26

0.35

0.49

182

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.20 Redness change generated by pure UV radiation (UV30), pure thermal treatment at 160 °C (T160) and simultaneous UV and thermal treatment at 160 °C (UV160)

Figure 4.20 presents the redness change as a function of treatment time for the different treatments. The redness value of the investigated species increased continuously during all three treatments. However, there are great differences regarding the intensity of the change. The initial redness value was multiplied by 1.6 for UV 30 treatment during the 16-h irradiation. These values are between 1.5 and 2 for thermal treatment of poplar and spruce at 160 °C. The simultaneous UV irradiation and thermal treatment at 160 °C generated much greater redness increase than the other two treatments together. The initial redness values were multiplied by 6.8 and 5.6 for spruce and poplar during the 16-h treatment. The redness change (Fig. 4.21) showed similar temperature dependence during UV light irradiation as the lightness change (Fig. 4.18). The red colour coordinate increased rapidly with elevated temperature. However, this increase was more balanced in the investigated temperature range than the lightness decrease. There are no obvious differences among the species regarding the redness change. The pure thermal treatment at 160 °C resulted in much smaller redness change compared to the UV irradiation at the same temperature. The effect of thermal treatment at 160 °C on the red coordinate change was slightly greater than the effect of light irradiation at 30 °C. The sum of the red coordinate changes caused by these two separate treatments (UV30 + T160) is much smaller than the red coordinate change caused by the simultaneous light irradiation and thermal treatment (UV160). Similarly, to what for lightness change was found, this figure also proves that the elevated temperature multiplies the effect of photodegradation. The average multiplication factor was 3 for redness change at 160 °C (comparing thr walues of UV160 and UV30+T160). The change of the yellow colour coordinate was different compared to the change of lightness and redness coordinate. Figure 4.22 shows that UV irradiation caused a significant yellow coordinate increase even at 30 °C. This increase was 91% of

4.5 Temperature Dependence of Colour Change During Photodegradation

183

Fig. 4.21 Redness change of the investigated wood species caused by UV irradiation at different temperatures and by pure thermal treatment (T) at 160 °C. The duration of all treatments was 16 h

Fig. 4.22 Yellowness change of the investigated wood species caused by UV irradiation at different temperatures and by pure thermal treatment (T) at 160 °C. The duration of all treatments was 16 h

the initial value for poplar. (For comparison, the same treatment caused only 9% lightness decrease and 55% redness increase for poplar.) The light irradiation at 80 °C induced similar yellowing as at 30 °C. The yellow coordinate increase is created mostly by the degradation products of lignin. It means that the lignin degradation is hardly temperature dependent in the 30–80 °C temperature range. The only exception was spruce, producing considerably greater yellowing at 80 °C than at 30 °C. This species was more sensitive regarding the yellowing at all applied treatments than

184

4 Monitoring of Wood Photodegradation by Colour Measurement

the other species. The figure also suggests that the UV degradation products of lignin are not stable enough at 160 °C, as most of the investigated species showed a smaller increase in yellowness at 160 °C than at 120 °C during UV irradiation. This high temperature can degrade newly formed chromophore compounds, resulting in yellowness decrease. The pure thermal treatment at 160 °C produced much smaller yellowness change than the UV irradiation at the same temperature. Furthermore, this yellowness increase caused by the pure thermal treatment at 160 °C was much smaller than the yellowness increase created by the photodegradation at 30 and 80 °C. In contrast, both the photodegradation at 80 °C and the pure thermal treatment at 160 °C induced similar redness change (Fig. 4.21) These findings strengthen the fact that the alteration of red coordinate is created mostly by the degradation products of the extractives while the yellowing is due to the degradation products of lignin. The sum of the changes of yellow coordinate caused by pure UV radiation and pure thermal treatment (UV30 + T160) is similar to the yellowness change caused by the simultaneous light irradiation and thermal treatment (UV160). The small difference was positive for spruce and poplar and negative for ash. This similarity does not mean that there is no multiplication effect. Figure 4.22 shows clearly that there is a third effect at 160 °C during UV irradiation as well. Namely, this temperature is high enough to degrade the new chemical products of photodegradation. This secondary degradation causes the decrease of yellowness at 160 °C. Among the applied treatment temperatures, 120 °C seems to be the one where this secondary degradation is not obvious, or has a small effect. This finding presents that the photodegradation products of lignin are thermally stable if the temperature does not exceed the 120 °C limit. The effects of 120 °C are visible in Fig. 4.23 showing the yellow coordinate change of each investigated wood species. The pure thermal treatment at 120 °C caused only a minor yellow coordinate increase, much less than at 160 °C (see Fig. 4.22). The sum of the yellow coordinate increases caused by the two separate treatments (UV30 + T120) is considerably smaller than the yellow coordinate increase caused by the simultaneous light irradiation and thermal treatment (UV120). It shows that the thermal effect during photodegradation is not only the simple addition of two effects, but the elevated temperature multiplies the effect of photodegradation. The multiplying factor is 1.6 (average of the 4 investigated species) for the yellowness change at 120 °C. Figure 4.24 shows the relative yellowness values as a function of treatment time. It presents that the yellowness change has a maximum after around 8 h of UV irradiation at 160 °C. The 120 °C treatment does not induce maximum yellowness change in the examined treatment period. The Arrhenius equation relates the rate of a chemical reaction to temperature and it includes the activation energy (see Sect. 3.2). For a single rate-limited thermally activated process, an Arrhenius plot gives a straight line, representing that the temperature dependence of the investigated chemical change is exponential. Arrhenius plots are often used to analyse the effect of temperature on the rates of chemical reactions. The Arrhenius plots for lightness, redness and yellowness are presented in Figs. 4.25, 4.26 and 4.27, respectively. The data of one selected species are presented

4.5 Temperature Dependence of Colour Change During Photodegradation

185

Fig. 4.23 Yellowness change of the investigated wood species caused by UV irradiation at different temperatures and by pure thermal treatment (T) at 120 °C. The duration of all treatments was 16 h

Fig. 4.24 Relative yellowness generated by pure UV radiation (UV30), pure thermal treatment at 160 °C (T160) and simultaneous UV and thermal treatment at 120 °C (UV120) and at 160 °C (UV160)

in all cases, because the other examined species showed similar behaviours. All Arrhenius plots have a breaking point close to 100 °C. The logarithm of lightness decreased with increasing temperature (Fig. 4.25) presented here by the data of ash. The tendency of decrease was similar for both dry thermal treatment and simultaneous UV and thermal treatment, although this decrease was more intense in the latter case. The existence of a breaking point means that the lightness change is

186

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.25 Arrhenius plots of the lightness change for ash samples caused by UV irradiation (Therm + UV) and pure thermal treatment (Therm) at the indicated temperatures

determined by more than one temperature dependent chemical change above 100 °C. Hemicelluloses are the most temperature sensitive molecules among the main wood substances. The thermal degradation of hemicelluloses can be one of the additional chemical changes above 100 °C. The other effect can be the creation of quinones during photodegradation at elevated temperatures. Popescu et al. (2013) reported the softening and the condensation of lignin during thermal treatment at 140 °C. Probably, the UV light irradiation can create much more quinones during the redistribution and condensation of lignin at 160 °C than at ambient temperatures (Kubovsky and Kacík 2014). These quinone structures can partly be responsible for the darkening and for the redness increase. The Arrhenius plots of red coordinates are shown in Fig. 4.26 presented here by the data of spruce. The logarithmic value of redness increased with rising temperature. The increase was more rapid for simultaneous UV and thermal treatment than for pure dry thermal treatment. The Arrhenius plots are close to linear in both cases. (The R2 value of linear regression line is 0.92 for spruce in case of simultaneous UV and thermal treatment.) This is because the redness change is determined mostly by the degradation products of the extractives. The trend lines of Fig. 4.26 have a breaking point at around 100 °C. Below 100 °C, the red coordinate increase was slower for both treatment types than above 100 °C indicating the complexity of chemical changes above 100 °C (similar to the lightness). The Arrhenius plots of yellowness data are visible in Fig. 4.27 presented by the data of Scots pine. The logarithm of yellow coordinates increased by rising temperature. The tendency of this increase was completely different for simultaneous UV and thermal treatment and for pure dry thermal treatment. The rising temperature of the pure thermal treatment resulted in continuous yellow coordinate increase represented by a straight line. The coefficient of determination is rather high (R2 = 0.96). It represents that the thermal degradation of extractives follows the Arrhenius law.

4.5 Temperature Dependence of Colour Change During Photodegradation

187

Fig. 4.26 Arrhenius plots of the redness change for spruce samples caused by UV irradiation (Therm + UV) and pure thermal treatment (Therm) at the indicated temperatures

Fig. 4.27 Arrhenius plots of the yellowness change for Scots pine samples caused by UV irradiation (Therm + UV) and pure thermal treatment (Therm) at the indicated temperatures

The photodegradation of wood at elevated temperatures (above 100 °C) is a highly complex process. That is why the Arrhenius plot is not a straight line. The yellowness change caused by the photodegradation is created mainly by the degradation products of lignin. The degradation of lignin below 80 °C is hardly dependent on temperature, as shown in Fig. 4.27. Only a small yellow coordinate decrease was detected for poplar and ash and a little increase for the conifers between 30 °C and 80 °C. Above 80 °C the yellow coordinate increase was rapid, but it was visible up to 120 °C only. Above 120 °C the degradation products of lignin underwent further

188

4 Monitoring of Wood Photodegradation by Colour Measurement

thermal degradation reducing the yellowness. These three different tendencies are visible on the Arrhenius plot. The Arrhenius equation indicates that there will always be some reaction at any temperatures, even if the temperature is low. Slow changes become observable if the experimental period is long enough. This phenomenon is visible in case of indoor claddings made of spruce. The originally yellowish surface becomes more and more reddish during years even at 20–25 °C. Matsuo et al. (2011) found that colour alteration of a historical hinoki sample during 921 years at ambient temperature was almost equivalent to that of heating at 180 °C for approximately 6.7 h. The experimental results showed that UV light irradiation produces much greater discolouration at elevated temperatures above 80 °C than at ambient temperature. All three colour parameters (L*, a*, b*) reflected this magnifying effect of the elevated temperature. For comparison, the discolouration data caused by UV light irradiation at 30 °C (effect of UV light) were added to the discolouration data caused by pure thermal treatment at 160 °C. Simultaneous light irradiation and thermal treatment at 160 °C resulted in much greater discoloration than expected based on the sum of the discolorations caused by UV light irradiation at 30 °C and by pure thermal treatment at 160 °C. Photodegradation at elevated temperature is not only the simple addition of two effects, but higher temperatures multiply the effect of photodegradation. The biggest difference was revealed in the change of redness. The Arrhenius plots of all three colour coordinates have a breaking point close to 100 °C representing that above this temperature the chemical changes become more complex.

4.6 Air Humidity Dependence of Colour Change During Photodegradation The sun radiation tests presented in Sect. 4.3. showed that high air humidity modifies the colour change during photodegradation (see the text after Fig. 4.7). To test the air humidity dependence of photodegradation beech and spruce samples were put into a closed quartz tube. A water bath under the samples ensured 100% relative air humidity within the closed tube (wet condition). The temperature inside the tube was measured by thermocouple. For comparison, samples were placed into a similar tube but without water bath (dry condition). Mercury vapour lamp was used for UV irradiation. The total electric power of the applied double mercury lamps was 800 W and the samples were placed 64 cm from the lamps. It is important to mention that the surface of quartz tube reflected approximately 10% of the incident light. This portion of the incident light did not reach the surface of the samples. The temperature was 32 °C for the first series and 53 °C for the second series within the wet tube during the irradiation. (The temperature within the irradiation chamber was set to 30 and 50 °C.) The irradiation was interrupted after 6, 16, 36 and 72 h for colour and IR measurements. Five replicates were cut from the same board for both species

4.6 Air Humidity Dependence of Colour Change During Photodegradation

189

Fig. 4.28 Lightness change of spruce (S) samples due to UV light irradiation under wet (W) and dry (D) conditions at the indicated temperatures

with a dimension of 30 × 10x5 mm. The initial colour data of samples were slightly different at 32 °C and 53 °C. The initial moisture content of the samples was 12 ± 1% and it increased up to 16 ± 2% during the treatment in wet condition at 32 °C. The results are presented here according to a previously published paper (Tolvaj et al. 2016). The usual decrease in lightness was observed in all cases and for both examined species (Fig. 4.28). The rapid decrease in lightness at the beginning was followed by a moderate decrease. The wet condition caused considerably greater darkening than the dry condition at both temperatures. The deviation started after 6-h irradiation and the difference increased by prolonged treatment time. The treatment at higher temperature resulted in greater differences in lightness between wet and dry condition during the 10–60 h irradiation time. This difference was close to equal after 72 h irradiation. Figure 4.29 shows the redness change of beech samples generated by mercury lamp irradiation under wet and dry conditions at 32 and 53 °C. Irradiation under wet condition resulted in greater red colour change than under dry condition for all examined species at both temperatures. The red colour coordinate increased rapidly during the first 16 h of exposure, followed by a moderate continuous increase. The elevated temperature (53 °C) produced a more rapid redness increase at the beginning of the treatment than the lower one (32 °C). After 36 h the trend lines were close to parallel. Redness change of spruce samples are similar to that of beech. Figure 4.30 shows the yellowness change of beech samples generated by mercury lamp irradiation under wet and dry conditions at 32 and 53 °C temperatures. The yellow colour change of the samples was similar for both examined species at both temperatures (Fig. 4.30). Wet condition resulted in somewhat greater change than dry condition. The treatment at 53 °C induced greater differences between the dry and wet trendlines than the treatment at 32 °C.

190

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.29 Change in the redness of beech (B) samples due to UV light irradiation under wet (W) and dry (D) conditions at the indicated temperatures

Fig. 4.30 Yellowness change of spruce (S) samples due to UV light irradiation under wet (W) and dry (D) conditions at the indicated temperatures

Summarizing the UV light induced colour changes under wet and dry conditions, it can be stated that the irradiation under wet condition produced considerably greater colour change than under dry condition. This finding is particularly important because the air humidity in tropical countries is much higher than in continental climates. Consequently, the results of different outdoor weathering tests are not necessarily comparable unless the air humidity data are carefully monitored.

4.7 Effect of Water Leaching for Photodegraded Wood

191

4.7 Effect of Water Leaching for Photodegraded Wood The colour of wood is mainly determined by the extractive content. Many types of the extractives are water soluble. Consequently, the colour of both photodegraded and natural wood can be modified by water leaching. This phenomenon can be recognised in case of outdoor wooden constructions. As rain leaches out natural and photodegraded extractives over the years, the colour of outdoor wooden constructions will become less saturated and turn towards grey. Photodegraded surfaces become usually grey after two-year outdoor exposure. The grey cellulose remains on the surface only. The leaching effect of water was studied under artificial conditions in the case of photodegraded spruce wood. Samples were irradiated by a mercury lamp, and then plunged into distilled water (wet treatment). A double mercury vapour lamp, as a strong UV light emitter, provided the light irradiation. The total electric power of the applied double mercury lamps was 800 W. The UV radiation was 80% of the total emission of the lamps. Specimens were placed 64 cm from the lamp. The temperature in the chamber was 50 °C during the irradiation. The irradiation time was 24 h, followed by 6-h water leaching. These two treatments together formed one cycle, and this cycle was repeated 50 times. The other series of specimens got light irradiation only (dry treatment). The colour of the wood specimens was measured after each (UV and leaching) treatments on the radial surfaces, as the average colour of earlywood and latewood. The colour measurement was carried out after both light irradiation and water leaching during the first 10 cycles. During the next 40 cycles the colour data were measured after UV radiation only. Wet samples were dried at 30 °C up to the initial weight after each leaching. This process guaranteed that the colour was measured at the same moisture content each time. The results are presented here according to an earlier published paper (Kannar et al. 2018 with written permission of Wood Research). The main goal of the study was to determine the colour modification of photodegraded wooden surface by water leaching. These alterations are visible in Figs. 4.31, 4.32 and 4.33 for the 3 colour coordinates (L*, a* and b*). Figure 4.31 shows the change of lightness. The first bar represents the initial lightness of the samples. The second bar shows the effect of UV radiation (24 h), followed by the bar generated by water leaching (6 h). These two treatments were repeated 10 times. The UV radiation produced continuous lightness decrease, as reported in earlier papers (Tolvaj and Faix 1995; Oltean et al. 2010; Wang and Ren 2008; Sharratt et al. 2009; Cogulet et al. 2016). The water leaching partly washed out the degradation products of the light irradiation generating lightness increase in all 10 cycles of treatments. Similar results were found by Furuno (2001) treating hinoki samples. The lightness intensity showed a very slight decrease after the seventh cycle, which indicates that the majority of possible degradation has already taken place on the surface of the tested samples. Figure 4.32 shows the redness change during the cyclic UV radiation and water leaching. The UV radiation generated continuous redness value increase during the

192

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.31 Lightness change of spruce samples caused by UV radiation and water leaching (w)

Fig. 4.32 Redness change of spruce samples caused by UV radiation and water leaching (w)

treatments. This increase diminished after the seventh cycle. The water leaching partly washed out the water-soluble chromophore degradation products. The greatest leaching effect can be observed during the third cycle. After the seventh cycle, the leaching effect was negligible. Figure 4.33 shows the yellowness change during the cyclic UV irradiation and water leaching. The UV radiation produced substantial yellowness value increase during the first day of the treatment, as it was reported in earlier papers (Tolvaj and

4.7 Effect of Water Leaching for Photodegraded Wood

193

Fig. 4.33 Yellowness change of spruce samples caused by UV radiation and water leaching (w)

Faix 1995; Wang and Ren 2008; Sharratt et al. 2009; Persze and Tolvaj 2012; Cogulet et al. 2016). This rapid change was followed by moderate increase up to the fifth cycle and the yellowness value decreased slightly afterward. The water leaching reduced the yellowness intensity in all cases. Figures 4.31, 4.32 and 4.33 show the detailed effects of UV radiation and water leaching during the first ten cycles. The experiments were continued for 50 cycles, but the colour of the sample surfaces was measured only after 12, 16, 21, 25, 30, 42 and 50 cycles. Figures 4.31, 4.32 and 4.33 show that the effect of leaching on the colour was small right before the tenth cycle. This is why the cumulative effect of leaching was determined compared to the effect of pure UV treatment (dry treatment) in the second part of the investigation. The changes are presented based on UV irradiation time as independent variable. The greatest difference between the effects of dry and wet treatments was found for the yellow colour coordinate (Fig. 4.34). The irradiation time in this figure include the duration of UV irradiation only, for the correct comparison of the two types of treatments (leaching time is not included). The yellowness value increased rapidly up to the fourth cycle. The curve of leached samples runs slightly above the curve of dry UV treatment during the first four cycles. The reason could be that the leaching opened new surface for the photodegradation. After this point, the two curves separate and the curve of the leached samples runs under the curve of the dry UV treatment. The distance between the two trendlines increased with prolonged irradiation time. This fact shows that yellow chromophores were leached out continuously by water. The exact determination of the types of leached molecules requires further chemical investigation. The yellowness intensity of dry samples increased up to 12 days of UV radiation, and remained almost constant during further treatment. (The measured colour data after the 25th irradiation day show some kind of distortion

194

4 Monitoring of Wood Photodegradation by Colour Measurement

in both Figs. 4.34 and 4.35. The reason might be the incorrect installation of the colorimeter.) The yellowness value of leached samples increased during the first four days of treatment only, and slightly decreased afterward. The 50-day UV irradiation multiplied the initial yellowness value 2.13 times, and the leaching reduced this factor up to 1.71. Three phenomena contributed to the yellowness intensity change. The UV light degraded the chromophore compounds of wood reducing the yellowness value. The chromophore products of lignin degradation intensified the yellowness while the leaching mitigated it by washing out the newly formed molecules. During the first

Fig. 4.34 Yellowness change of spruce samples caused by UV radiation (Dry) and by UV radiation and water leaching (Leached)

Fig. 4.35 Redness change of spruce samples caused by UV radiation (Dry) and by UV radiation and water leaching (Leached)

4.7 Effect of Water Leaching for Photodegraded Wood

195

four cycles the effect of lignin degradation was dominant. Between the fourth and the twelfth cycles, the colour reducing effect of leaching was compensated by the yellow chromophore production and the intensity of yellowness remained constant. After the twelfth cycle the leaching was slightly more dominant than the photodegradation and the yellowness value decreased. Figure 4.35 shows the redness change during the 50 cycles UV radiation and water leaching. Pure UV radiation (dry treatment) generated intensive redness value increase during the first twelve days of irradiation. After this period, the redness value had a plateau up to the 25th day of exposure. Further exposure generated slight redness value increase. The curve of leached samples runs slightly above the curve of dry UV treatment in the first six cycles. The reason, could be that the leaching opened new surfaces for the photodegradation. The redness value of leached samples increased during the first twelve days of treatments, then slightly decreased and later remained unchanged. The distance between the two trendlines (dry and leached) increased slightly with irradiation time. This fact shows that red colorants were leached out continuously by water. The 50 days UV radiation multiplied the initial redness value 4.48 times, and the leaching reduced this factor up to 3.28. The increase of redness value was generated partly by the degradation products of extractives (Timar et al. 2016). Similar procedures can be observed during the steaming of wood (Tolvaj et al. 2010). Spruce wood has low extractive content, that is why the degradation products of lignin play important role in redness change. During light irradiation phenolic groups react with photons and form phenolic radicals that (e.g., phenoxyl- and galvinoxyl radicals) transform into o- and p-quinonoid structures (Leary 1968; Pandey 2005a; Cogulet et al. 2016). These newly generated quinones increase the value of redness. The majority of red dyestuffs produced by plants are quinones (Melo 2009; Mills and White 2011). The tendency of lightness decrease (Fig. 4.36) was similar to the tendency of yellowness and redness increase. The two curves (dry and leached) diverge after four days of UV irradiation and run almost parallel. The distance between the two curves is small, approximately three units, which is only 4% of the initial lightness value. Previous studies of 17 wood species showed good linear correlation between their lightness (L*) and hue angle (h*) (Tolvaj et al. 2013a). Experiments demonstrated that this correlation holds throughout the colour change caused by steaming (Tolvaj and Nemeth 2008), and photodegradation (Tolvaj and Mitsui 2010) as well. The colour data gained in this study offers an opportunity to investigate the validity of this correlation also for combined UV radiation and water leaching. Figure 4.37 presents the correlation between lightness and hue values. The dots representing the first cycle are in the top-right corner followed by the colour dots of the irradiated samples from right to left with growing irradiation time. The two trend lines are close to each another. The high values of the coefficients of determination show good correlation between the hue angle and the lightness in both cases. The slope of the two trend lines differs slightly, showing that the hue of leached samples decreased somewhat less intensively than the hue of dry samples, as compared to the lightness change.

196

4 Monitoring of Wood Photodegradation by Colour Measurement 90

Dry Leached

L* Lightness

85 80 75 70 65

0

10

20

30

40

50

Light irradiation time (day) Fig. 4.36 Lightness change of spruce samples caused by UV radiation (Dry) and by UV radiation and water leaching (Leached)

Fig. 4.37 Correlation between lightness and hue angle values of spruce wood during UV irradiation (Dry) and combined UV irradiation and water leaching (Leached)

The experimental results showed that the leaching partly removed the yellow and red chromophore molecules generated by the UV radiation. The specimens became slightly lighter after water leaching. The leached specimens were slightly yellower and redder during the first 4–6 days of UV irradiation than those in the dry series. The 50 days UV radiation multiplied the initial redness value 4.48 times, and the leaching reduced this factor up to 3.28. These values for yellowness were 2.13 and 1.71, respectively. Good correlation was found between the hue angle and the lightness in both types of treatments.

4.8 Photodegradation Properties of Thermally Modified Wood

197

4.8 Photodegradation Properties of Thermally Modified Wood Thermal treatment is applied in wood processing industry to improve some properties of natural wood. The colour modification generated by thermal treatments belongs to the topic of this study. The result of thermal treatment is highly temperature dependent (see Sect. 3.2). The effect of pure thermal treatment is slow below 120 °C but rapid above 160 °C. Pure thermal treatment is used to modify the properties of natural wood between 160 and 240 °C. The colour change is not the main purpose of pure thermal treatment, it is only a subsidiary result. The presence of water amplifies the effect of thermal treatment. Consequently, thermal treatment can be effective in wet condition (steaming) below 120 °C. The main purpose of steaming is usually the colour modification. The colour stability of thermally modified wood is an important practical and theoretical issue. The photodegradation properties of thermally modified wood are discussed in this study. Artificial light irradiation was applied for studying the colour stability of steamed wood species and species treated by dry thermal processes. Steaming as a pre-treatment was done at 100, 110 and 120 °C with duration of two days. The involved species were black locust heartwood, beech, poplar and spruce. Radial surfaces were prepared for colour measurement. The steamed samples were irradiated by mercury vapour lamp (introduced at the end of Sect. 4.2). As the 80% of light emission was in the UV region, the treatment is called as UV irradiation. The total UV irradiation time was 90 h. It was interrupted after 7-, 16-, 36- and 60-h treatment for colour measurement. The results are presented here according to a previous paper (Varga et al. 2021 with written permission of Wood Research). The two-hour steaming as pre-treatment generated highly different colour parameters for the investigated species. These parameters were the starting points of UV irradiation. Figures presented in this section show the results of steaming and UV irradiation as two continuous treatments. Figure 4.38 shows the lightness decrease for poplar. Steaming produced decreasing lightness values with increasing steaming temperature for all investigated species. The distances among the lightness values after steaming were greater for black locust than for poplar. Spruce and beech showed smaller distances among the steaming generated lightness values than poplar. The UV irradiation caused rapid lightness decrease for the natural samples during the first 16 h followed by slow lightness decrease. In the case of steamed samples, the rapid lightness decrease occurred during the first 7 h of UV irradiation, for all investigated species. Samples steamed at 100 and 110 °C showed slow lightness decrease up to 16 h of irradiation and the lightness value remained constant for all species thereafter. Black locust presented slow lightening after 16-h irradiation. Samples steamed at 120 °C showed darkening during the first 7-h irradiation only. This change was followed by slow lightness value increase up to the end of the irradiation. Spruce did not show lightness change in this condition. Black locust presented the greatest lightness increase, which counted 5 units for samples steamed at 120 °C. The results show that steaming reduces the UV

198

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.38 Lightness change of natural and steamed (St) poplar specimens caused by steaming (48 h) and by UV irradiation (90 h)

light induced lightness alterations compared to the unsteamed (natural) samples. The differences can be confirmed by the values of lightness change during the 90-h UV irradiation for spruce. Natural spruce presented 10.7 units lightness decrease while the same data for samples steamed at 100, 110 and 120 °C were 5.9, 3.1 and 3 units, respectively. It is important to mention that spruce was the only species which did not show lightness increase due to the UV irradiation. The redness change among the species was not as uniform as the lightness change. Steaming generated redness increase for all investigated species. This increase was similar and the greatest for spruce and poplar. Beech produced the smallest redness increase due to the steaming. This increase was around two units and the change was hardly temperature dependent for beech. The photodegradation behaviour of black locust is different compared to the behaviour of other species (Fig. 4.39). Not only the natural but also the steamed samples also showed much greater redness increase than the other species during the first 7 h of irradiation. Natural samples presented similar redness increase as the other species later on. The redness value of steamed black locust samples did not change after 16-h UV irradiation. Only the samples steamed at 120 °C showed slow redness decrease in this irradiation period. The results show that extractives in unsteamed black locust are highly sensitive to UV irradiation. In contrast, steamed black locust was fairly stable during UV irradiation. The reason could be that steaming modified those extractives during the pre-treatment that are sensitive to UV irradiation. Spruce, beech and poplar showed similar tendencies in redness change during both steaming and UV irradiation (Fig. 4.40). Natural samples produced greater redness increase during the first 16-h of irradiation than the steamed ones. The trendlines run parallel after 36-h UV irradiation representing the equal changes. This redness increase was continuous during the investigated period. Only the samples steamed at 120 °C presented small redness decrease during the first 36 h of irradiation, but then

4.8 Photodegradation Properties of Thermally Modified Wood

199

Fig. 4.39 Redness change of natural and steamed (St) black locust specimens caused by steaming (48 h) and by UV irradiation (90 h)

joined to the other samples. The applied different steaming temperatures resulted in small deviations in redness values for beech (not presented). Consequently, the trendlines of beech run even closer to each other than those of other species. The four trendlines merged into one single line after 60 h of UV irradiation. The results show that steaming in general is unable to protect the wood against the redding effect of photodegradation. The only exception was the black locust, which was quite stable against the damaging effects of UV light irradiation. Steaming resulted in very different alterations in yellowness for the investigated species. Black locust suffered yellowness decrease (Fig. 4.41). The greatest decrease

Fig. 4.40 Redness change of natural and steamed (St) spruce specimens caused steaming (48 h) and UV irradiation (90 h)

200

4 Monitoring of Wood Photodegradation by Colour Measurement

occurred at 120 °C (16 units). The greatest yellowness increase for spruce and poplar were 15 and 14 units. Beech showed small yellowness decrease (about 2 units) due to the steaming. Black locust was an exception also in terms of the UV irradiation induced yellowness change. Natural black locust showed rapid yellowness increase during the first 7 h of irradiation and the increase was moderate afterward. Steamed black locust samples presented yellowness increase only during the first 7 h of irradiation and there were no changes during the irradiation thereafter. Only steamed samples at 120 °C showed a slight decrease in yellowness. In contrast, all other investigated species showed a continuous increase in yellowness during the entire UV irradiation period. While at the beginning of UV irradiation, the modification of the extractives is responsible for the rapid increase in yellowness, the further yellowness change is primarily determined by the degradation products of lignin. Steamed black locust samples did not show yellowness increase in the main part of the applied UV irradiation. The reason could be that the steam-modified extractives were able to protect the lignin in black locust against photodegradation. Figure 4.42 shows the yellowness increase of spruce samples generated by steaming and UV irradiation. The effect of UV irradiation was similar for all samples, regardless of the temperature of the previous steaming. The trendlines run parallel continuously presenting a monotone increase of yellowness values. Poplar and beech presented similar alterations as spruce with the only deviation that the “parallel” trendlines of beech are much closer to each other than those of spruce. The results of yellowness change demonstrated that steaming improved the photostability only for black locust. The investigated species gave different colour change responses to the UV irradiation and the preliminary steaming. Steaming reduced the hue value of beech samples (Fig. 4.43A). The greatest hue value decrease was 7° generated by steaming at 120 °C. The direction of hue value decrease was not temperature dependent but the intensity

Fig. 4.41 Yellowness change of natural and steamed (St) black locust specimens caused by steaming (48 h) and by UV irradiation (90 h)

4.8 Photodegradation Properties of Thermally Modified Wood

201

Fig. 4.42 Yellowness change of natural and steamed (St) spruce specimens caused by steaming (48 h) and by UV irradiation (90 h)

change was slightly temperature dependent. The UV irradiation increased the hue values up to a constant value. This value was 70° for the steamed samples and 73° for the natural beech samples. Reaching this value, the hue remained constant during further irradiation. It is interesting that all steamed sample reached the 70° hue value sooner or later, regardless of the hue decrease caused by the preliminary steaming. The straight trendlines show slight convergence towards a common dot far away. The b*/a* quotient as function of UV irradiation time compares the intensity change of redness and yellowness. Redness change is usually generated by the degradation of extractives, while the yellowness change is produced by the degradation of lignin during photodegradation. Black locust is an exception. Its naturally yellowish colour values can change according to the degradation of extractives as well. Figure 4.43/B shows the irradiation time dependence of the b*/a* quotient for beech. Dots on the vertical axis represent the b*/a* value of untreated and steamed samples. Steaming reduced the b*/a* values considerably but the temperature dependence was small. Yellowing was dominant during the first 7 h of UV irradiation for the natural samples. This trend changed and the b*/a* values showed slow but continuous decrease after 16-h irradiation. Yellowing was dominant also for the steamed specimens during the first 7 h irradiation but the intensity of the change was slower than for natural samples. This tendency was slower but continued up to the 36th hour of the UV irradiation. The b*/a* quotient remained constant (around 2.76 units) afterward. The linear and horizontal trendlines represent that the redness and the yellowness increased proportionally after 36-h UV irradiation. Steaming generated much greater chromaticity alterations for black locust (Fig. 4.44/A) than for beech. The greatest hue value decrease was 29° generated by the steaming at 120 °C. The direction and intensity of hue value decrease was strongly temperature dependent. The hue value of natural black locust continuously decreased during UV irradiation. The decrease was 8° and its tendency was different

202

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.43 (A) Colour change of untreated (natural) and steamed (48-h steaming at different temperatures) beech caused by 90-h UV irradiation, represented on the a*–b* plane. Filled symbol represents the starting point of the steaming. Constant hue (h*) values are also indicated. (B) b*/a* quotients of beech generated by 48-h steaming followed by 90-h UV irradiation

compared to that of beech. The linear trendline represents that redness and yellowness of natural black locust increased proportionally. The first 7 h of UV irradiation induced intensive redness and yellowness increase in the steamed samples, but these changes were not accompanied by hue alterations. Further UV irradiation did not modify neither the redness nor the yellowness. The visible small deviations were generated by the colour inhomogeneity of the samples. Figure 4.44/B shows the irradiation time dependence of the b*/a* quotient for black locust. Steaming reduced the b*/a* values considerably represented by the dots on the vertical axis. This reduction was strongly temperature dependent. Natural samples presented rapid b*/a* value decrease during the first 7 h of UV irradiation

4.8 Photodegradation Properties of Thermally Modified Wood

203

Fig. 4.44 (A) Colour change of untreated (natural) and steamed (48-h steaming at different temperatures) black locust caused by 90-h UV irradiation, represented on the a*–b* plane. Filled symbol represents the starting point of the steaming. Constant hue (h*) values are also indicated. (B) b*/a* quotients of black locust generated by 48-h steaming followed by 90-h UV irradiation

followed by slow and continuous decrease. In most wood species, the yellowness value increase is caused mainly by the degradation of lignin at the beginning of photodegradation. Black locust, however, is an exception. Its original yellow colour value changed due to two different chemical alterations. While lignin decomposition increased, the degradation of (robinetin type) extractives reduced the yellowness value resulting in a moderate yellowness value increase at the beginning of the photodegradation. In contrast, the redness change was intensive at the beginning of the photodegradation (see Fig. 4.39). These two effects resulted in a decrease in the b*/a* value of unsteamed black locust. Steamed specimens showed uniform b*/a* value alterations. The b*/a* values hardly changed by elapsed irradiation time. The

204

4 Monitoring of Wood Photodegradation by Colour Measurement

b*/a* value may remain constant in two ways; both parameters change proportionally or both remain constant. The first scenario happened in the first 7 h of irradiation and the second occurred afterward. The average value of the b*/a* quotients after 90-h UV irradiation was 1.6 units (h* = 58 deg.) for steamed samples. Steaming generated remarkable hue value decrease for poplar samples (Fig. 4.45/ A). The greatest hue value decrease was 11° generated by steaming at 120 °C. The tendency of hue value decrease was independent of the temperature but the value of change was strongly temperature dependent. The hue value of natural poplar increased during the first 7 h of UV irradiation and linearly decreased afterward. Steamed samples presented similar hue changes as the natural poplar. All trendlines have a linear section represented by straight lines. These straight lines run nearly parallel towards a distant common dot. Spruce showed similar hue value alterations as poplar. Figure 4.45/B shows the irradiation time dependence of b*/a* quotient for poplar. Steaming produced great reduction in b*/a* values represented by the dots on the vertical axes. This reduction was slightly temperature dependent. Yellowing due to lignin degradation was dominant during the first 7 h of UV irradiation for the natural samples represented by the rapid increase of b*/a*. This trend changed and the b*/a* values showed continuous decrease up to the end of the 90-h UV irradiation. Steamed specimens also showed small yellowing dominance during the first 16-h irradiation and the b*/a* values remained constant afterward. The average value of the b*/a* quotients after the 90-h UV irradiation was 3.62 units (h* = 74.6 deg.) for steamed samples. The trendlines of the steamed samples moved towards this hue value during the photodegradation, as shown in Fig. 4.45/A. Spruce samples presented similar b*/a* diagram as poplar. Only the intensity increase was smaller during the first 7 h of UV irradiation for natural samples. The average value of the b*/a* quotient at the beginning of the UV irradiation was 3.89 units (h* = 75.6 deg.) and it hardly changed during the 90-h UV irradiation and the finale average value was 3.81 units (h* = 75.3 deg.). It is an interesting finding that the b*/a* trendlines aim for a common value during UV irradiation independently of the different effects of steam pre-treatment and the b/a* values remained constant for all investigated species after 36-h UV irradiation independently of the applied steaming temperature. This constant value represented that the value of b* and a* coordinates changed proportionally or both remained constant. The average of these constant values was 1.6, 2.76, 3.62 and 3.81 for black locust, beech, poplar and spruce, respectively. This order is the opposite of the order of extractive content. These results show that although high extractive content gives dominance to redness and low extractive content to yellowness change, the ratio remains constant for each steamed sample during UV irradiation. Understanding the chemical changes occurring during the superposition of the two effects (steaming and UV irradiation) requires further chemical investigations. The results of the physical examinations demonstrated that steaming can improve the photostability only for black locust during short term UV irradiation. Similar positive protecting effect of steaming was found by Dudiak and his co-workers for beech and birch species (Dudiak et al. 2022; Dudiak and Dzurenda 2023).

4.8 Photodegradation Properties of Thermally Modified Wood

205

Fig. 4.45 (A) Colour change of untreated (natural) and steamed (48-h steaming at different temperatures) poplar caused by 90-h UV irradiation, represented on the a*–b* plane. Filled symbol represents the starting point of the steaming. Constant hue (h*) values are also indicated. (B) b*/a* quotients of poplar generated by 48-h steaming followed by 90-h UV irradiation

Another, reverse possibility for combined thermal and light irradiation treatment is where light irradiation is followed by heat treatment. Mitsui and his co-workers studied this phenomenon (Mitsui et al. 2001, 2004a, b; Mitsui 2004). Samples got first 60-h light irradiation followed by 150-h steaming (in 90% air humidity) at 50, 70 and 90 °C. Control samples got only steam treatment. The results of spruce specimens are presented here. The initial colour parameters of the different series of samples were slightly different presented by the filled marks in the left down corner of the diagram (Fig. 4.46/A). Photodegradation produced intensive yellowness and redness increase. The yellowness increase was proportionally more intensive than the redness increase presented by the small hue increase. Steaming after light irradiation generated even

206

4 Monitoring of Wood Photodegradation by Colour Measurement

greater chromaticity increase during the first 5 h of steaming but here the increase of redness value was proportionally more intensive than that of the yellowness, and this resulted in hue decrease. After this period, the yellowness increase slowed down at 50 °C. The other two steaming temperature caused yellowness decrease after the first 5-h steaming. Moreover, the redness value also decreased after 20h steaming at 90 °C. For comparison, the effect of pure steaming (without light irradiation pre-treatment) is also presented in Fig. 4.46/A. Steaming alone generated much smaller chromaticity alteration than the steaming after light irradiation. The pure 150-h steaming created 0.8, 1.1 and 1.4° hue decrease at 50, 70 and 90 °C, respectively. Similar data for the previously light irradiated samples are 4.3°, 6.9° and 10°, respectively. It means that light irradiation pre-treatment multiplied the effect of steaming 5, 6 and 7 times at 50, 70 and 90 °C, respectively. This great difference demonstrates that light irradiation pre-treatment greatly accelerates the colour modification effect of steaming. Figure 4.46/B shows the irradiation time dependence of the b*/a* quotient for spruce samples. Pure steaming generated slow b*/a* value decrease demonstrating that the redness increase was proportionally more intensive than the yellowness increase. In contrast, photodegradation induced proportionally more intensive yellowness increase than redness increase. The effect of steaming on the UV irradiated samples was completely different compared to the samples without preliminary UV irradiation. The decrease of b*/a* value was extremely intensive during the first 5 h of steaming. The trendlines became divergent after this period. Comparing the effects of two types of combined treatments (steaming + light irradiation Figs. 4.43/B and light irradiation + steaming Fig. 4.46/B), it can be concluded that the mechanism of chemical changes for these combined treatments are different. The first type of combined treatment generated convergent while the second one divergent b*/a* values. Thermal modification of timbers in dry condition needs relatively high treatment temperatures. The 160–260 °C interval is used in industrial practice. The main advantages of thermal treatment are: reduced hygroscopicity, improved dimensional stability, and better resistance to degradation due to insects and micro-organisms. The improvement of these properties gives the possibility to use thermally modified wood for outdoor applications. The colour of wood is also modified by thermal treatment. This effect of dry thermal treatment fits to the topics of this study. Thermal treatment of wood creates dark and attractive brown colour, which is strongly determined by the applied treatment temperature and time. The treatment conditions are also important. Chemical changes are different if oxygen is included or excluded. The colour stability of thermally modified wood under outdoor conditions is also an important aspect for costumers. Colour change of thermally treated wood caused by photodegradation is not well investigated, however, and the published results are sometimes contradictory. The real comparison is difficult because of the different conditions applied by the authors. Ayady et al. (2003) tested the colour stability of heat-treated ash, beech, maritime pine, and poplar wood samples. The heat treatment was done at 240 °C for 2 h, under nitrogen atmosphere. The heat-treated samples were exposed to UV light for

4.8 Photodegradation Properties of Thermally Modified Wood

207

Fig. 4.46 (A) Colour change of natural and previously UV irradiated spruce during 150-h steaming at different temperatures, represented on the a*–b* plane. Duration of the UV pre-treatment was 60 h. Filled symbols represent the starting point of the UV irradiation. Constant hue (h*) values are also indicated. (B) b*/a* quotients of spruce generated by 150-h steaming with and without preliminary 60-h UV irradiation

835 h. The total colour change was determined representing the changes. The results showed that the colour stability of the heat-treated wood was better than that of the untreated control samples. Similar result was found by Yu et al. (2021) during artificial UV light exposure of thermally modified bamboo. Yildiz et al. (2011) exposed outside heat-treated alder wood in Turkey for 3 years. The treatment parameters were: 150 °C, 180 and 200 °C, for periods of 2, 6, and 10 h. The results showed that heat treatments delayed and decreased the rate of colour change caused by the weathering but did not completely prevent it. The most advantageous treatment parameters were 200 °C for 10 h. Similar results were found

208

4 Monitoring of Wood Photodegradation by Colour Measurement

during artificial weathering (UV irradiation and water spray) of Scots pine, spruce, iroko and ash as well (Yildiz et al. 2013). The artificial photodegradation properties of heat-treated jack pine was studied using xenon lamp (Huang et al. 2012a, b). The total irradiation time was 1500 h. There was no difference in the a* and b* colour coordinate values between thermally treated and the untreated wood after 400 h light irradiation. There were slight differences at shorter irradiation time, but the authors did not measure the colour parameters during the first 72 h of the treatment. The same authors stated in other paper: “The heat treatment increases the lignin and crystallised cellulose contents, which to some extent protects heat-treated birch against degradation due to weathering” (Huang et al. 2013). Cirule et al. (2021) studied the indoor photodegradation properties of thermally (170 °C for 1 h in a water vapour medium under elevated pressure 0.8 MPa) modified ash, aspen and Scots pine samples. Slower rate and less variations in colour change among species were detected for thermally modified wood compared to unmodified ones. Nuopponen et al. (2004) reported that heat treated wood was more resistant to natural weathering mainly because some of its lignin degradation products are less leachable than those of untreated wood. Mikleˇci´c et al. (2011) studied the discoloration properties of natural and coated beech, ash and hornbeam species. The duration of UV treatment was 32 days. It was found that in the first half of exposure to UV light, the surface of uncoated thermally modified (at 190 and 212 °C) ash, beech and hornbeam wood samples discolored slowly compared to uncoated unmodified wood samples. FTIR spectra of thermally modified ash, beech and hornbeam wood samples exposed to UV light showed similar chemical changes as unmodified wood samples exposed to UV light, but less pronounced. Srinivas and Pandey (2012) investigated the photodegradation behaviour of thermally treated rubber wood. The thermal treatment was carried out in vacuum atmosphere at 225 °C for 2, 4, and 6 h. The IR spectra showed significant lignin degradation in thermally modified wood within few hours of exposure. Results of colour changes and FTIR spectroscopy revealed that thermal modification of wood does not induce resistance against UV radiation. The light wavelength dependence of photodegradation for thermally modified and unmodified aspen samples were also investigated (Cirule et al. 2016a, b). The total irradiation time was 100 h. It was found, that wavelength longer than 600 nm did not generate degradation of wood. Greater changes in IR spectra were observed for thermally modified wood compared to unmodified wood for all analysed wavebands and irradiation systems, which suggests that thermally modified wood was more chemically transformed by irradiation. Black locust (Robinia pseudoacacia L.) and poplar (Populus × euramericana cv. Pannónia) wood were oil-heat treated (OHT) in sunflower oil and other series of samples underwent thermal treatment at the same temperatures in atmospheric condition (dry thermal treatment). These species were chosen because black locust has high- while poplar has low extractive content. The dimensions of the samples were relatively small 100 × 20 × 10 (mm) to help the heating up process. The

4.8 Photodegradation Properties of Thermally Modified Wood

209

applied heating temperatures were 160 and 200 °C. The treatment durations were 2, 4 and 6 h. The samples were immersed directly into the hot oil bath without preheating, and at the end of the oil-heat treatment were taken out immediately. 30min preheating was applied for the samples treated in atmospheric condition. The thermally treated samples together with the untreated control samples were subject to artificial photodegradation. A strong UV light emitter, mercury vapour lamp (introduced at the end of Sect. 4.2) provided the light irradiation. An irradiation chamber set for 70 °C ensured ambient temperature conditions. The total irradiation time was 36 h. The irradiation was interrupted after 3, 7 and 16 h for measuring the colour change. The 2-h steaming time at 160 °C in atmospheric condition produced only very little colour change, therefore these samples were not used for further investigation. The results of samples treated in air condition are presented here according to Tolvaj et al. (2014) while the results of oil-heated samples according to Nemeth et al. (2016) with written permission from Elsevier. For proper comparison, all similar graphs have the same vertical scale magnitude. The natural colour of wood is mostly determined by the extractives. Thermally modified wood has additional chromophores produced by the degradation of the hemicelluloses. The investigated samples suffered intensive colour change during thermal treatment. The left side of the figures present the colour modification effects of the applied thermal treatments, followed by the colour alterations generated by UV irradiation. The lightness of black locust (Fig. 4.47) decreased from 70 down to 24, and varied between 24 and 34 depending on the heating temperature and time generated by OHT. There were only small differences between the lightness decreasing effects of the applied two (160 and 200 °C) temperatures. The dry thermal treatment generated greater diversity in lightness than OHT (Fig. 4.48), applying the same temperature and time conditions. The dry thermal treatment reduced the lightness of black locust up to 11 units because of oxidation of the heat generated chemical radicals. The variation was large between 50 and 12 units. It is almost 4 times larger interval than in case of oil-heat treatment. The lightness alteration behaviour of oxidation process was more time and temperature dependent for dry thermal treatment than for the oil-heat treatment. The lightness modification effect of dry thermal treatment was more time and temperature dependent for poplar (Fig. 4.49) than for black locust. The 6-h dry thermal treatment of poplar at 160 °C resulted in 13 units while at 200 °C resulted in 52 units lightness decrease. Similar data for OHT poplar were 6 and 47 units. The treatment in oil reduced the lightness change intensity at all temperatures more uniform compared to dry thermal treatment. Comparing the effects of dry treatment at 160 °C in Figs. 4.48 and 4.49 shows the dominance of extractives in lightness change for black locust. This treatment produced 2.7 times greater lightness decrease for black locust than for poplar. This multiplication factor at 200 °C was only 1.1, representing the dominance of hemicelluloses in lightness decrease effect at 200 °C. The UV radiation produced intensive lightness decrease during the first 7 h of irradiation for both types of natural (not treated) species (Figs. 4.47 and 4.48). This change was followed by slow lightness decrease up to the end of the 36-h irradiation. All of the OHT black locust samples showed small lightness decrease during the

210

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.47 Lightness change of black locust (BL) by thermal (T) treatment in heated oil and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

Fig. 4.48 Lightness change of black locust (BL) by dry thermal (T) treatment and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

first 3 h of UV irradiation followed by very slow increase and all trendlines run parallel (Fig. 4.47). In contrast, the lightness values of the samples after dry thermal treatment slightly increased during the first 3 h of UV irradiation (Fig. 4.48). All samples treated at 200 °C presented slow but continuous lightening.

4.8 Photodegradation Properties of Thermally Modified Wood

211

Fig. 4.49 Lightness change of poplar (P) by dry thermal (T) treatment and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

The results showed that both types of thermal treatments (in oil and in air) produced highly diverse lightness values. The diversity of lightness values was more time and temperature dependent for the dry thermal treatment than for the oil-heat treatment. All of the thermally modified samples were highly stable against UV irradiation. Only the dark samples (treated at 200 °C) showed slight lightening. Changes in redness showed greater differences between species and types of thermal treatments as well compared to the lightness change. The two applied temperature generated more uniform redness change in case of the OHT than in case of dry treatment. This phenomenon is vell visible in Figs. 4.50 and 4.51. The value of red colour coordinate of black locust more than doubled during the OHT. The heating time and temperature of OHT hardly affected the redness alteration of black locust. The dry thermal treatment generated more diverse redness change than OHT (Fig. 4.51). The most diverse differences were caused by photodegradation in the redness change among the colour coordinates. Natural black locust showed rapid redness increase during the first 7 h of UV irradiation. This increase was 75% of the total redness increase during the 36-h irradiation. The red colour value of untreated poplar changed almost uniformly during the whole irradiation period (Fig. 4.52). The rapid increase at the beginning of the UV irradiation is completely missing due to the low extractive content of poplar. The OHT black locust samples showed similar photodegradation properties independently of the temperature of thermal pretreatment. The samples become a little more reddish during the first 3 h of UV exposure and the redness values slowly returned to the starting values during the further exposure (Fig. 4.50). These results demonstrated that black locust samples pre-treated in hot oil were fairly stable in terms of red hue change caused by photodegradation.

212

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.50 Redness change of black locust (BL) by thermal (T) treatment in heated oil and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

Fig. 4.51 Redness change of black locust (BL) by dry thermal (T) treatment and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

Black locust samples pre-treated at 200 °C under atmospheric conditions showed similar photostability as the OHT samples. The samples treated at 200 °C for 2 h showed the most stable redness values during photodegradation, while samples pretreated at 160 °C followed the redness alteration of the natural samples during UV

4.8 Photodegradation Properties of Thermally Modified Wood

213

Fig. 4.52 Redness change of poplar (P) by thermal (T) treatment in heated oil and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

irradiation. This result shows that dry thermal treatment at 160 °C does not improve the photostability of black locust wood. The starting points of poplar samples for UV irradiation were highly diverse generated by the pre-treatments in hot-oil (Fig. 4.52). There were great differences in photostability between the samples pre-treated at 160 and at 200 °C. Samples pre-treated at 160 °C showed moderate but continuous redness increase during UV irradiation. This change was similar as the redness increase of natural poplar samples. The redness coordinate of poplar samples pre-treated at 200 °C decreased slightly during UV irradiation. Samples treated for the shortest period (2-h) did not show real redness decrease. The redness values remained nearly constant during the entire UV irradiation period. The poplar samples pre-treated under atmospheric conditions (not presented here) showed similar changes as OHT samples, only the starting points were slightly different. The trendlines of red colour changes generated by UV irradiation are convergent. Both type of pre-treatments increased the photostability of black locust. The only exception was the dry treatment at 160 °C. Poplar samples thermally treated at 200 °C also showed photostability. The originally high yellowness value of black locust decreased due to both oilheat and dry thermal treatments (Figs. 4.53 and 4.54). In contrast, the low yellow colour coordinate of poplar increased following the tendency of red colour change (Fig. 4.55). The dry thermal treatment produced more diverse yellowness changes than the OHT, for both species. The yellowing of wood during photodegradation is mainly generated by the degradation products of lignin (Tolvaj and Faix 1995; Müller et al. 2003; Pandey 2005a; Timar et al. 2016). The yellow colour of natural black locust changed only

214

4 Monitoring of Wood Photodegradation by Colour Measurement

Fig. 4.53 Yellowness change of black locust (BL) by thermal (T) treatment in heated oil and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

Fig. 4.54 Yellowness change of black locust (BL) by dry thermal (T) treatment and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

a little during UV irradiation. Previous studies demonstrated that the high extractive content of black locust partly protects its lignin content during light irradiation (Nemeth et al. 1992; Pastore et al. 2004; Tolvaj and Varga 2012). In contrast, natural poplar samples, which have low extractive content, revealed a more intensive yellowness value increase compared to black locust. All of oil-heat treated black locust samples showed similar photodegradation behaviour as natural samples

4.8 Photodegradation Properties of Thermally Modified Wood

215

Fig. 4.55 Yellowness change of poplar (P) by dry thermal (T) treatment and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

(Fig. 4.53) representing that the heat treatment in oil does not increase the protection of lignin against photodegradation. Moreover, all thermally pre-treated samples showed greater yellowness increase during the first 7 h of UV irradiation than natural samples. Black locust samples after dry thermal treatment at 160 °C showed much greater and continuous yellowing than the natural black locust presented (Fig. 4.54). This finding shows that the oxidation processes after the thermal treatment at 160 °C demolished almost all extractives that protected the lignin. This yellowness increase was similar as the yellowness increase of poplar (with low extractive content) samples dry treated at 160 °C (Fig. 4.55). Black locust samples pre-treated at 200 °C for 4 and 6 h showed similar photodegradation behaviour as the natural samples. The yellow colour change of the samples treated at 200 °C for 2 h seemed to be the most stable, as discussed in the case of red colour change. Figure 4.55 shows the yellowness data of poplar subjected to dry thermal treatment. The pre-treatment caused 1.8 times greater yellowness change at 200 °C than at 160 °C. This data was 1.5 in case of OHT. The UV irradiation resulted in continuous yellowness increase for the samples pre-treated at 160 °C. The tendency of increase was close to linear during the entire irradiation period. Test results of both poplar and black locust samples confirmed that the extractives did not protect the lignin if the pre-treatment at 160 °C occurred in the presence of air. The same was found in a previous study as well (Srinivas and Pandey 2012). Yellowness value of samples pre-treated at 200 °C did not change after 3-h of irradiation, only the samples after 2-h thermal treatment showed slow yellowness increase. Poplar samples pre-treated in oil showed the same yellowness change as those subjected to dry thermal treatment (therefore this figure is not presented here).

216

4 Monitoring of Wood Photodegradation by Colour Measurement

Results showed that the extractive content of the wood plays an important role in the colour change not only during thermal treatment but also during light irradiation. It was found that thermal treatment at 200 °C reduces the red colour change caused by the photodegradation. Oil-heat treatment was found to be more effective than dry thermal treatment in terms of protection of samples against redness change. The yellow colour change during photodegradation was hardly influenced by the applied thermal treatments, showing that thermal treatments are not able to reduce the UV light-induced degradation of lignin. The applied treatments reduced the lightness change effect of UV irradiation. Figure 4.56 shows the hue change of black locust samples subjected to dry thermal treatment. The thermal treatments at different temperatures resulted in very diverse hue values. The greatest hue difference generated by the two types of treatments (at 160 and 200 °C) was 15°. The subsequent UV irradiation reduced this deviation continuously. The difference was only three degrees at the end of the 36-h UV irradiation. The OHT created less diverse hue values (not shown here). The greatest hue difference generated by the two types of treatments was only two degrees. This small difference remained constant also during the entire UV irradiation procedure. The hue values of the black locust samples were 65 (±2)° after both thermal treatment types and the subsequent 36-h UV irradiation. Figure 4.57 shows the hue change of poplar samples thermally treated in hot oil. The thermal treatments at different temperatures resulted in diverse hue values. The greatest hue difference generated by the two types of treatments (at 160 and 200 °C) were 4° and 10° . The UV irradiation raised the hue difference during the first 16-h irradiation and reduced afterward. The greatest difference was 13° and it was reduced to 8° up to the end of the UV irradiation with centre at 75°. Poplar samples after dry thermal pre-treatment showed similar hue values as OHT samples. The greatest

Fig. 4.56 Hue angle change of black locust (BL) by dry thermal (T) treatment and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

4.8 Photodegradation Properties of Thermally Modified Wood

217

Fig. 4.57 Hue angle change of poplar (P) by thermal (T) treatment in heated oil and by UV irradiation. (The first number indicates the temperature of thermal treatment, and the second shows the treatment time in hours.)

hue difference was also 8° at the end of the UV irradiation with centre at 74°. It is interesting to mention that steaming of poplar at low temperatures (100–120 °C) followed by UV irradiation resulted in similar changes as dry thermal pre-treatment at much higher temperatures and subsequent UV irradiation. The same final hue values were produced in both cases (see Fig. 4.45). The trendlines of UV irradiations after thermal treatments showed similar alteration tendencies for poplar as for black locust. The main difference was that the trendlines of UV irradiation were more convergent for black locust than for poplar. Experimental results showed that the extractive content of the wood plays an important role in the colour change not only during thermal treatment but also during UV light irradiation. It was found that thermal treatment at 200 °C reduces the red colour change caused by the photodegradation. The oil-heat treated black locust samples showed similar photodegradation properties regardless of the thermal pretreatment time and temperature. The redness of OHT black locust samples hardly changed during UV radiation showing the photostability of the thermally modified extractives. The redness of poplar samples pre-treated at 200 °C was fairly stable to photodegradation, while the samples pre-treated at 160 °C showed the same photodegradation properties as natural poplar samples. The yellow colour change during photodegradation was hardly influenced by the applied preliminary thermal treatments, showing that thermal treatments cannot reduce the UV light-induced degradation of lignin. The applied thermal treatments slightly reduced the lightness change effect of the subsequent UV light irradiation.

218

4 Monitoring of Wood Photodegradation by Colour Measurement

References Andrady, A.L., Searle, N.D., Crewdson, L.F.E.: Wavelength sensitivity of unstabilized and UV stabilized polycarbonate to solar simulated radiation. Polym. Degrad. Stab. 35, 235–247 (1992). https://doi.org/10.1016/0141-3910(92)90031-Y Ayady, N., Lejeune, F., Charrier, F., Merlin, A.: Color stability of heat treated wood during artificial weathering. Holz Roh Werkstoff 61, 221–226 (2003). https://doi.org/10.1007/s00107-0030389-2 Barta, E., Papp, G., Preklet, E.J., Tolvaj, L., Berkesi, O., Nagy, T., Szatmári, S.: Changes in the DRIFT spectra of softwood materials irradiated by UV-laser as a function of energy. Acta Silv. Lign. Hung. 1(1), 83–91 (2005) Chang, ST., Wang, SY., Cheng, SS.: Red color enhancement of sugi (Cryptomeria japonica D.Don) heartwood by light irradiation. J. Wood Sci. 45, 271–273 (1999) Chang, ST., Wang, S.Y., Cheng, S.S.: Environmental effects on the color of sugi (Cryptomeria japonica D.Don) heartwood. J. Wood Sci. 46, 390–394 (2000) Chang, T.C., Chang, H.T., Wu, C.L., Lin, H.Y., Chang, S.T.: Stabilizing effect of extractives on the photooxidation of Acacia confusa wood. Polym. Degrad. Stab. 95, 1518–1522 (2010). https:// doi.org/10.1016/j.polymdegradstab.2010.06.012 Cirule, D., Meija-Feldmane, A., Kuka, E., Andersons, B., Kurnosova, N., Antons, A., Tuherm, H.: Spectral sensitivity of thermally modified and unmodified wood. BioResources 11, 324–335 (2016a) Cirule, D., Kuka, E., Antons, A.: Disparity in discolouration of thermally modified wood exposed to solar and artificial ultraviolet irradiation. Rural Sust. Res. 34, 12–20 (2016b). https://doi.org/ 10.1515/plua-2015-0008 Cirule, D., Kuka, E., Kevers, M., Andersone, I., Andersons, B.: Photodegradation of unmodified and thermally modified wood due to indoor lighting. Forests 12(8), ID1060 (2021). https://doi. org/10.3390/f12081060 Cirule, D., Kuka, E., Sansonetti, E., Hivrica, M., Andersone, I., Andersons, B.: Wood photosensitivity to different artificial light sources. Eur. J. Wood Prod. 80, 1507–1519 (2022). https://doi. org/10.1007/s00107-022-01842-2 Cogulet, A., Blanchet, P., Landry, V.: Wood degradation under UV irradiation: a lignin characterization. J. Photochem. Photobiol. B: Biol. 158, 184–191 (2016). https://doi.org/10.1016/j.jphoto biol.2016.02.030 Csanady, E., Magoss, E., Tolvaj, L.: Quality of Machined Wood Surfaces. Springer, Heidlberg New York London (2015). https://doi.org/10.1007/978-3-319-22419-0 Derbyshire, H., Miller, E.R.: The photodegradation of wood using solar irradiation. Part 1. Effects on the structural integrity of thin wood strips. Holz Roh Werkstoff 39, 341–350 (1981) Dirckx, O.G., Masson, D.M., Deglise, X.M.: Photooxydation du bois. Wood Pulp. Chem. Simp. (1987) Dudiak„ M., Dzurenda, L., Kucerova, V.: Effect of Sunlight on the change in color of unsteamed and steamed Beech Wood with water steam. Polymers 14, ID 1697 (2022). https://doi.org/10. 3390/polym14091697 Dudiak„ M., Dzurenda, L.: The effect of sunlight on the color change of steamed birch wood. Acta Facultatis Xylologiae Zvolen 65(1), 35−43 (2023). https://doi.org/10.17423/afx.2023.65.1.03 Feist, W.C., Hon, D.N.S.: Chemistry of weathering and protection. In: Rowell, R. (ed.) The Chemistry of Solid Wood. American Chemical Society, Washington (1984) Furuno, T.: Histochemical study on wood deterioration by UV irradiation and prevention of deterioration using poly PEGMA. In: Imamura, Y (ed.) High-Performance Utilization of Wood for Outdoor Uses. Press-Net, Kyoto (2001). ISBN: 4-9900870-0-3 Hikita, Y., Toyoda, T., Azuma, M.: Weather testing of timber-discoloration. In: Imamura, Y (ed.) High-Performance Utilization of Wood for Outdoor Uses. Press-Net, Kyoto (2001). ISBN: 4-9900870-0-3

References

219

Hon, D.N.S., Ifjú, G.: Measuring penetration of light into wood by detection of photo-induced free radicals. Wood Sci. 11(2), 118–127 (1978) Hon, D.N.S.: Weathering and photochemistry of wood. In: Hon, D.N.S., Shiraishi, N. (eds.) Wood and Cellulose Chemistry. Marcel Dekker, New York (2001) Huang, X., Kocaefea, D., Kocaefea, Y., Boluk, Y., Pichette, A.: Study of the degradation behavior of heat-treated jack pine (Pinus banksiana) under artificial sunlight irradiation. Polym. Degrad. Stab. 97, 1197–1214 (2012a). https://doi.org/10.1016/j.polymdegradstab.2012.03.022 Huang, X., Kocaefe, D., Kocaefe, Y., Boluk, Y., Pichette, A.: A spectrocolorimetric and chemical study on color modification of heat-treated wood during artificial weathering. Appl. Surf. Sci. 258, 5360–5369 (2012b). https://doi.org/10.1016/j.apsusc.2012.02.005 Huang, X., Kocaefe, D., Kocaefe, Y., Boluk, Y., Pichette, A.: Structural analysis of heat-treated birch (Betule papyrifera) surface during artificial weathering. Appl. Surf. Sci. 264, 117–127 (2013). https://doi.org/10.1016/j.apsusc.2012.09.137 Kannar, A., Tolvaj, L., Magoss, E.: Colour change of photodegraded spruce wood by water leaching. Wood Res. 63(6), 935–946 (2018) Kránitz, K., Sonderegger, W., Bues, T., Niemz, P.: Effects of aging on wood: a literature review. Wood Sci. Technol. 50(1), 7–22 (2016). https://doi.org/10.1007/s00226-015-0766-0 Kropat, M., Hubbe, M.A., Laleicke, F.: Natural, accelerated, and simulated weathering of wood: a review. BioResources 15(4), 9998–10062 (2020) Kubovsky, I., Kacik, F.: Colour and chemical changes of the lime wood surface due to CO2 laser thermal modification. Appl. Surf. Sci. 321, 261–267 (2014). https://doi.org/10.1016/j.apsusc. 2014.09.124 Kubovsky, I., Oberhofnerová, E., Kacik, F., Pánek, M.: Surface changes of selected hardwoods due to weather conditions. Foreasts 9(9), ID 557 (2018). https://doi.org/10.3390/f9090557 Leary, G.: Photochemical production of quinoid structures in wood. Nature 217, 672–673 (1968) Liu, X.Y., Liu, M., Lv, M.Q., Lv, J.F.: Photodegradation of three hardwood species by sunlight and xenon light sources. BioResources 14(3), 6909–6922 (2019) Liu, X., Timar, M.C., Varodi, A.M., Nedelcu, R., Torcatoru, M.J.: Colour and surface chemistry changes of wood surfaces coated with two types of waxes after seven years exposure to natural light in indoor conditions. Coatings 12, ID 1689 (2022). https://doi.org/10.3390/coatings1211 1689 Matsuo, M., Yokoyama, M., Umemura, K., Sugiyama, J., Kawai, S., Gril, J., Kubodera, S., Mitsutani, T., Ozaki, H., Sakamoto, M., Imamura, M.: Aging of wood: Analysis of color changes during natural aging and heat treatment. Holzforschung 65, 361–368 (2011). https://doi.org/10.1515/ HF.2011.040 Melo, MJ.: History of natural dyes in the Ancient Mediterranean World. In: Bechtold, T., Mussak, R. (eds.) Handbook of Natural Colorants, pp. 6–10. Willey, London (2009) Mikleˇci´c, J., Jirous-Rajkovi´c, V., Antonovi´c, A., Špani´c, N.: Discoloration of thermally modified wood during simulated indoor sunlight exposure. BioResources 6, 434–446 (2011) Mills, J., White, R.: The Organic Chemistry of Museum Objects. Routledge, London (2011) Mitsui, K., Takada, H., Sugiyama, M., Hasegawa, R.: Changes in the properties of light-irradiated wood with heat treatment. Part 1. Effect of treatment conditions on the change in color. Holzforschung 55, 601–605 (2001). https://doi.org/10.1515/HF.2001.098 Mitsui, K.: Changes in the properties of light-irradiated wood with heat treatment. Part 2. Effect of light-irradiation time and wavelength. Holz Roh Werkstoff 62, 23–30 (2004). https://doi.org/ 10.1007/s00107-003-0436-z Mitsui, K., Murata, A., Tsuchikawa, S., Kohara, M.: Wood photography using light irradiation and heat treatment. Color. Res. Appl. 29, 312–316 (2004a). https://doi.org/10.1002/col.20027 Mitsui, K., Murata, A., Tolvaj, L.: Changes in the properties of light-irradiated wood with heat treatment: Part 3. Monitoring by DRIFT spectroscopy. Holz Roh Werkstoff 62, 164–168 (2004b). https://doi.org/10.1007/s00107-004-0476-z Mitsui, K., Tolvaj, L., Papp, G., Bohus, J., Szatmári, S., Berkesi, O.: Changes in the properties of light-irradiated wood with heat treatment—application of laser. Wood Res. 50(1), 1–8 (2005)

220

4 Monitoring of Wood Photodegradation by Colour Measurement

Mitsui, K., Tsuchikawa, S.: Low atmospheric temperature dependence on photodegradation of wood. J. Photochem. Photobiol. B: Biol. 81, 84–88 (2005). https://doi.org/10.1016/j.jphotobiol. 2005.05.011 Müller, U., Rätzsch, M., Schwanninger, M., Steiner, M., Zöbl, H.: Yellowing and IR changes of spruce wood as result of UV-irradiation. J. Photochem. Photobiol. B: Biol. 69, 97–105 (2003). https://doi.org/10.1016/S1011-1344(02)00412-8 Nemeth, K., Vano, V., Faix, O.: The effect of wood extractives on the photodegradation of wood. In: EWLP conference, Grenoble, France, pp. 191–192 (1992) Nemeth, R., Tolvaj, L., Bak, M., Alpar, T.: Colour stability of oil-heat treated black locust and poplar wood during short-term UV radiation. J. Photochem. Photobiol. A: Chem. 329, 287–292 (2016). https://doi.org/10.1016/j.jphotochem.2016.07.017 Nuopponen, M., Wikberg, H., Vuorinen, T., Jamsa, S., Viitaniemi, P.: Heat treated softwood exposed to weathering. J. Appl. Polym. Sci. 91(4), 2128–2134 (2004). https://doi.org/10.1002/app.13351 Oltean, L., Hansmann, C., Nemeth, R., Teischinger, A.: Wood surface discolouration of three Hungarian hardwood species due to simulated indoor sunlight exposure. Wood Res. 55, 49–58 (2010) Pandey, K.K.: Study of the effect of photo-irradiation on the surface chemistry of wood. Polym. Degrad. Stab. 90, 9–20 (2005a). https://doi.org/10.1016/j.polymdegradstab.2005.02.009 Pandey, K.K.: A note on the influence of extractives on the photo-discoloration and photodegradation of wood. Polym. Degrad. Stab. 87, 375–379 (2005b). https://doi.org/10.1016/j. polymdegradstab.2004.09.007 Pandey, K.K., Vuorinen, T.: Comparative study of photodegradation of wood by a UV laser and a xenon light source. Polym. Degrad. Stab. 93, 2138–2146 (2008). https://doi.org/10.1016/j.pol ymdegradstab.2008.08.013 Papp, G., Preklet, E.J., Kosiokva, B., Barta, E., Tolvaj, L., Bohus, J., Szatmári, S., Berkesi, O.: Effect of UV laser radiation with different wavelengths on the spectrum of lignin extracted from hard wood materials. J. Photochem. Photobiol. A: Chem. 163(1–2), 187–192 (2004). https:// doi.org/10.1016/S1010-6030(03)00446-5 Papp, G., Barta, E., Preklet, E.J., Tolvaj, L., Berkesi, O., Nagy, T., Szatmári, S.: Changes in DRIFT spectra of wood irradiated by UV laser as a function of energy. J. Photochem. Photobiol. A: Chem. 173(2), 137–142 (2005). https://doi.org/10.1016/j.jphotochem.2005.01.012 Pastore, T.C.M., Santos, K.O., Rubim, J.C.: A spectrocolorimetric study on the effect of ultraviolet irradiation of four tropical hardwoods. Biores. Technol. 93, 37–42 (2004). https://doi.org/10. 1016/j.biortech.2003.10.035 Persze, L.: Magyarországi fafajok fotodegradációjának összehasonlítása: Színváltozás. [Photodegradation of wood species grown in Hungary: colour change.] Faipar 59(2–3), 35–46 (2011) Persze, L., Tolvaj, L.: Photodegradation of wood at elevated temperature: colour change. J. Photochem. Photobiol. B: Biol. 108, 44–47 (2012). https://doi.org/10.1016/j.jphotobiol.2011. 12.008 Preklet, E., Papp, G., Barta, E., Tolvaj, L., Berkesi, O., Bohus, J., Szatmári, S.: Changes in DRIFT spectra of wood irradiated by lasers of different wavelength. J. Photochem. Photobiol. B: Biol. 112, 43–47 (2012). https://doi.org/10.1016/j.jphotobiol.2012.04.006 Romagnoli, M., Segoloni, E., Luna, M., Margaritelli, A., Gatti, M., Santamaria, U., Vinciguerra, V.: Wood colour in Lapacho (Tabebuia serratifolia): chemical composition and industrial implications. Wood Sci. Technol. 47(4), 701–716 (2013). https://doi.org/10.1007/s00226-0130534-y Sharratt, V., Hill, CAS., Kint, DPR.: A study of early colour change due to simulated accelerated sunlight exposure in Scots pine (Pinus sylvestrris). Polym. Degrad. Stab. 94, 1589–1594 (2009). https://doi.org/10.1016/j.polymdegradstab.2009.04.010 Shebani, A.N., van Reenen, A.J., Meincken, M.: The effect of wood extractives on the thermal stability of different wood species. Thermochim. Acta 471, 43–50 (2008). https://doi.org/10. 1016/j.tca.2008.02.020

References

221

Srinivas, K., Pandey, K.: Photodegradation of thermally modified wood. J. Photochem. Photobiol. B: Biol. 117, 140–145 (2012). https://doi.org/10.1016/j.jphotobiol.2012.09.013 Timar, M.C., Varodi, A.M., Gurau, L.: Comparative study of photodegradation of six wood species after short-time UV exposure. Wood Sci. Technol. 50, 135–163 (2016). https://doi.org/10.1007/ s00226-015-0771-3 Tolvaj, L., Faix, O.: Artificial ageing of wood monitored by DRIFT spectroscopy and CIE L*a*b* color measurements. I. Effect of UV light. Holzforschung 49, 397–404 (1995). https://doi.org/ 10.1515/hfsg.1995.49.5.397 Tolvaj, L., Nemeth, K.: Correlation between hue-angle and colour lightness of steamed black locust wood. Acta Silv. Lign. Hung. 4, 55–59 (2008) Tolvaj, L., Mitsui, K.: Correlation between hue angle and lightness of light irradiated wood. Polym. Degrad. Stab. 95, 638–642 (2010). https://doi.org/10.1016/j.polymdegradstab.2009.12.00 Tolvaj, L., Molnar, S., Nemeth, R., Varga, D.: Color modification of black locust depending on the steaming parameters. Wood Res. 55(2), 81–88 (2010) Tolvaj, L., Varga, D.: Photodegradation of timber of three hardwood species caused by different light sources. Acta Silv. Lign. Hung. 8, 145–155 (2012) Tolvaj, L., Nemeth, R., Pasztory, Z., Bejo, L., Takats, P.: Colour stability of thermally modified wood during short-term photodegradation. BioResources 9(4), 6644–6651 (2014) Tolvaj, L., Tsuchikawa, S., Inagaki, T., Varga, D.: Combined effects of UV light and elevated temperatures on wood discoloration. Wood Sci. Technol. 49, 1225–1237 (2015). https://doi.org/ 10.1007/s00226-015-0749-1 Tolvaj, L., Popescu, CM., Molnar, Zs., Preklet, E.: Effects of air relative humidity and temperature on photodegradation processes in beech and spruce wood. BioResources 11(1), 296–305 (2016) Varga, D., Tolvaj, L., Preklet. E.: Colour stability of steamed black locust, beech and spruce timbers during short-term photodegradation. Wood Res. 66(4), 544–555 (2021) Wang, X., Ren, H.: Comparative study of the photo-discoloration of moso bamboo (Phillostachys pubescens Mazel) and two wood species. Appl. Surf. Sci. 254, 7029–7034 (2008). https://doi. org/10.1016/j.apsusc.2008.05.121 Yildiz, S., Yildiz, U.C., Tomak, E.D.: The effects of natural weathering on the properties of heattreated alder wood. BioResources 6, 2504–2521 (2011) Yildiz, S., Tomak, E.D., Yildiz, U.C., Ustaomer, D.: Effect of artificial weathering on the properties of heat treated wood. Polym. Degrad. Stab. 98, 1419–1427 (2013). https://doi.org/10.1016/j.pol ymdegradstab.2013.05.004 Yu, H., He, S., Zhang, W., Zhan, M., Zhuang, X., Wang, J., Yu, W.: Discoloration and degradation of bamboo under ultraviolet radiation. Int. J. Polym. Sci. ID 6803100 (2021). https://doi.org/10. 1155/2021/6803100 Zahri, S., Belloncle, C., Charrier, F., Pardon, P., Quideau, S., Charrier, B.: UV light impact on ellagitannins and wood surface colour of European oak (Quercus petraea and Quercus robur). Appl. Surf. Sci. 253, 4985–4989 (2007). https://doi.org/10.1016/j.apsusc.2006.11.005 Zivkovic, V., Arnold, M., Radmanovic, K., Richter, K., Turkulin, H.: Spectral sensitivity in the photodegradation of fir wood (Abies alba Mill.) surfaces: colour changes in natural weathering. Wood Sci. Technol. 48, 239–252 (2014). https://doi.org/10.1007/s00226-013-0601-4

Chapter 5

Applications of IR Spectrum Measurement in Wood Research

Abstract The chapter presents the possible usages of infrared spectrum measurement in wood research. These spectra provide information about the chemical changes generated by different treatments. A detailed discussion of the chemical changes generated by steaming and by photodegradation can be found in this chapter. Experimental results demonstrate that the difference spectrum can show the positions of absorption maxima in the fingerprint region more precisely than the initial absorbance spectrum. The change of IR spectrum demonstrates that deacetylation is the main chemical change during steaming. The photodegradation effect of different artificial light sources are compared to the effect of sun radiation. The effect of such influencing parameters as temperature, air relative humidity and leaching effect of rain during photodegradation are also followed by IR spectrum measurement. Detailed discussion of these influencing processes can be found in this chapter. Behaviour of steamed wood samples during photodegradation can be determined by IR absorbance measurement. The results show that steaming reduces the susceptibility to photodegradation only for black locust wood and the protective effect increases with rising steaming temperature. Keywords Wood · Infrared spectrum · Photodegradation · Difference spectrum · Extractives · Lignin · Ultraviolet radiation

5.1 Introduction Microscopes have limited magnification. Organic chemical groups (carbonyl, carboxyl, methyl, acetyl, hydroxyl) cannot be seen by microscopes. Photons can “communicate” with these groups by absorption. Spectroscopy helps us to determine the structures of the organic compounds and it is a good analytical tool to trace the chemical changes of organic molecules. Wood is a chemically complex material. It is a well organised composite of the structured polyoses and the lignin matrix molecules. The advantage of the absorption spectroscopy is that an individual band provides information about one chemical group. On the other hand, the place

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 L. Tolvaj, Optical Properties of Wood, Smart Sensors, Measurement and Instrumentation 45, https://doi.org/10.1007/978-3-031-46906-0_5

223

224

5 Applications of IR Spectrum Measurement in Wood Research

of maximum can be modified by the surrounding chemical groups. Due to the many types of groups within the wooden structure, which have different absorption positions, the IR spectrum of wood is highly compound. Nevertheless, that complexity makes an IR spectrum a molecular “fingerprint” since no two organic compounds have exactly the same IR spectrum. To identify the absorbing chemical groups of the measured sample, the chemists compiled lists of the absorption peaks of the IR spectrum. There are plenty of lists in the literature giving different positions for the same peak. The reason of the deviations lies in the determining method. The peak position can be determined using the original absorption spectrum, the second derivative spectrum and the difference spectrum created by one kind of treatment. The IR spectrum of wood consists of several overlapping individual bands in the fingerprint region. Unfortunately, the visible peak positions are not the real ones if the bands are overlapped (see Figs. 1.26 and 1.27). It means that the original absorption spectrum does not give exact peak positions for wood constituents. The mathematical process generating the second derivative of a spectrum creates invalid peak positions as well. It is difficult to separate the real peaks. Calculating the difference spectrum reduces the number of visible bands since only those bands will appear that were modified during the treatment. This manipulation reduces the number of overlapped bands considerably and the peak positions are usually well visible. The advantage of this method is a disadvantage as well. The peak position of bands unaffected by the applied treatment will be invisible. It is therefore highly important to clarify and present the method used for determining the peak position of the bands in publications. Table 5.1 presents the list of absorption bands in wood with the band assignments. The list represents the maxima of the peaks. The place of the maxima slightly depends on the wood species. That is why intervals are given in Table 5.1. The intervals were determined by experimental data using the difference spectra. Data of Table 5.1 slightly differ from the data of Table 1.2 because the peak positions in Table 1.2 were determined by the original absorbance spectrum, while the peak positions in Table 5.1 were specified using the difference spectra generated by various treatments. Absorption spectrum of an opaque material (such as wood) cannot be determined directly. Measurement of the reflectance spectrum, however, gives the possibility for calculating the absorption spectrum. The IR reflectance measurement protocol used in this study is as follows. The diffuse reflectance infrared Fourier transformed (DRIFT) spectrum of the samples was measured before and after treatment. Before all IR measurements, samples were stored for 3 days in the same laboratory in total darkness, to ensure equal moisture content for all measurements. Measurements were carried out using an IR spectrophotometer (JASCO FT/IR 6300). The resolution was 4 cm−1 and 64 scans were obtained and averaged. The background spectrum was obtained against an aluminium plate. The spectral intensities were calculated in Kubelka–Munk (K-M) units. Two-point baseline correction at 3800 and 1900 cm−1 (2632 and 5263 nm) was carried out. The intensity of spectra was normalised to the band maximum at around 1375 cm−1 (7273 nm). The intensity of spectra was adjusted to 1.0 by this normalisation at maximum around 1375 cm−1 . This C–H band of cellulose is often used as internal standard because of its high

5.1 Introduction

225

Table 5.1 Characteristic IR bands of wood and the band assignments (Csanady et al. 2015; Huang et al. 2008; Huang 2012) Wavenumber Wavelength (cm−1 ) (nm)

Assignment

3600–3550

Intramolecular hydrogen bond in phenolic group (in lignin) and weakly bounded absorbed water

2778–2817

3360–3310

2976–3021

Intramolecular hydrogen bonds in cellulose and hemicellulose

2957–2928

3382–3415

Asymmetric CH stretching in methyl and methylene groups

2908–2860

3439–3496

Symmetric CH stretching in methyl and methylene groups

1770–1757

5650–5692

C=O stretching vibration of non-conjugated ketones and γ lactones

1758–1743

5688–5737

acetyl groups in hemicelluloses

1736–1705

5760–5865

C=O stretching vibration in unconjugated ketone carbonyl and carboxyl groups

1660–1653

6024–6050

Conjugated C–O in quinines coupled with C=O stretching of various groups (flavones)

1628–1618

6143–6180

C=O stretching in flavones

1604–1593

6234–6277

Aromatic skeletal breathing with CO stretching (syringyl lignin)

1512–1505

6614–6645

Aromatic skeletal breathing (guaiacyl lignin)

1470–1460

6802–66,849

C–H deformation in lignin and C–H deformation in xylan

1435–1425

6969–7017

C–H deformation in lignin and in carbohydrates

1390–1380

7194–7246

C–H deformation in cellulose and hemicellulose

1380–1370

7246–7299

Aliphatic C–H deformation (symm.) in cellulose (Internal standard for normalisation)

1369–1366

7305–7321

Aliphatic C–H stretching in methyl and phenol OH

1333–1342

7502–7452

C–H deformation, C–OH stretching, syringyl ring

1319

7582

C–H2 wagging, C–H deformation (conifers)

1285

7782

C–H bending mode in cellulose

1277–1265

7831–7905

Caryl -O, guaiacyl ring breathing with CO stretching

1240–1230

8065–8130

C–O linkage in guaiacyl aromatic methoxyl groups and acetyl groups in xyloglucan

1180–1169

8474–8554

C–O–C stretching (asymm.) in cellulose and hemicelluloses

1160–1156

8620–8726

C–O–C stretching in pyranose rings, C=O stretching in aliphatic groups

1139–1130

8780–8849

C–O–C stretching (symm.), aromatic C–H i.p. deformation, glucose ring vibration

1108–1096

9025–9124

C–O–C stretching

1078–1066

9276–9381

C−O stretching mainly from C(3)−O(3)H in cellulose I

1058–1045

9451–9569

C−O and C−C stretching in cellulose and hemicelluloses

898–896

11,136–11,160 C–H deformation in cellulose and hemicellulose

226

5 Applications of IR Spectrum Measurement in Wood Research

intensity, central position and strong stability. The difference spectrum was calculated by subtracting the initial IR data from the data of irradiated sample. In this case the absorption increase is represented by a positive band while a negative band represents the absorption decrease.

5.2 Monitoring the Chemical Changes Caused by Steaming Wood has one of the most beautiful colour harmonies created by nature, because its colour range and colour pattern provide emotional comfort to the observer (Masuda 2001). However, not all wood species have attractive colour. There are a few species with a white-greyish tone (poplar, hornbeam, linden) without any visible texture and some of them have highly inhomogeneous colour (black locust, Turkey oak). Both disadvantages can be modified by steaming. The result of a heat treatment depends on the treatment atmosphere. The presence of water strongly amplifies the changes. Heating of wood in the presence of water or water steam resulted in the decrease of the carbonyl peak associated with the hydrolysis of acetyl and uronic groups from hemicelluloses to form acidic compounds (Mitchell 1988). The continuous presence of water as steam during the treatment process can prevent the occurrence of oxidative processes (Hill 2006). There are only a few papers dealing with the chemical changes induced by low temperature hydrothermal treatment. Popescu et al. (2013) investigated lime wood treated at 140 °C under semidry conditions. The formation of acetic acid, which catalyses the hydrolysis reactions of hemicelluloses and amorphous cellulose, was reported. The cleavage of the β-O-4 linkages and splitting of the aliphatic methoxyl chains from the aromatic lignin ring were found. For the first treatment interval, a higher extent of carbohydrate degradation was observed, then an increase of the extent of the lignin degradation also took place. The total duration of the treatment was 21 days. Li et al. (2015) studied the chemical changes of teak (Tectona grandis LF.) wood generated by steam-heat treatment at temperatures 120–220 °C. The treatment time was 2 h. The FTIR spectra proved the degradation of acetyl groups from hemicelluloses and the releasing of acetic acid through a deacetylation reaction during steamheat treatment. The cleavage of the β-O-4 linkages and the splitting of the aliphatic methoxyl chains from the aromatic ring in lignin were observed with increasing treatment temperature. Nemeth et al. (2016) studied the chemical changes of back locust, poplar and spruce species generated by 2-day steaming at 80 and 120 °C. Guaiacyl lignin in hardwoods showed slight degradation, but the syringyl lignin did not changed. The absorption decreases at 1175 cm−1 indicated the cleavage of ether linkage in cellulose and hemicelluloses at both steaming temperatures. The diffuse reflectance IR spectroscopy is a good analytical tool to follow the chemical modifications of wood during different treatments. This measurement technique was used to determine the chemical changes of wood generated by steaming

5.2 Monitoring the Chemical Changes Caused by Steaming

227

process. Beech (Fagus sylvatica L.), poplar (Populus x euramericana cv Pannonia) and spruce (Picea abies Mill.) species were involved in the test. The applied steaming temperatures were 80, 100 and 120 °C and the treatment time was 2 days. The steaming was carried out in a special steaming vessel at 100% relative humidity. The vessel was able to maintain the pressure up to 3 Bar. The initial moisture content of the samples was 10–11%. The samples were kept under normal laboratory conditions (i.e., 65% RH and T = 21 ˚C) before and after treatment to guaranty equal moisture content for IR measurement. The resolution of IR measurement was 4 cm−1 and 64 scans were obtained and averaged. The same area of the samples was measured before and after treatment. The background spectrum was determined against an aluminium plate. Two-point baseline correction at 3800 cm−1 and 1900 cm−1 (2632 and 5263 nm) was carried out. The absorbance intensities were calculated in KM units. The spectra were normalised to the band maximum around 1377 cm−1 (7262 nm). All presented spectrum was calculated as the average of 4 spectra. The difference spectrum was determined by subtracting the initial IR data from the data of steamed sample. The difference spectrum shows only the bands where alteration happened. The positive peaks represent the absorption increases and the negative peaks indicate the absorption decreases. The main chemical change during steaming is the deacetylation caused by the cleavage of acetyl groups linked as an ester group to the hemicelluloses (Dietrichs et al. 1978; Bourgois and Guyonnet 1988; Carrasco and Roy 1992). The acetyl group has detectable absorption in the carbonyl region. The acetyl content of wood depends on the wood species. The highly acetylated xylan is the main polyose component in hardwoods; however, glucose units in spruce are rarely acetylated. Figure 5.1/a shows the carbonyl absorption band of the investigated species. The carbonyl band is composed of minimum three individual overlapping bands. These bands are close to each other, consequently, the measured spectrum does not accurately show the real location of the maxima (see Fig. 1.26). The spectra of beech and poplar present two visible maxima at 1746 and 1741 cm−1 (5727 and 5744 nm). One of them is dominant for beech and the other one is dominant for poplar. The maximum at 1746 cm−1 is missing in case of spruce, thus the maximum shifted to 1737 cm−1 (5757 nm). All three spectra have a shoulder at 1724 cm−1 (5800 nm). Hydrothermal treatments reduce the intensity of the carbonyl band because of deacetylation of the hemicelluloses. Many papers reported this phenomenon as absorption decrease at the maximum absorption intensity between 1730 and 1740 cm−1 (5780 and 5747 nm), (Tjeerdsma and Militz 2005; Mikleˇci´c et al. 2011, Srinivas and Pandey 2012; Huang 2012; Esteves et al. 2013; Yildiz et al. 2013; Timar et al. 2016a; Okon et al. 2017). Figure 5.1/b shows the carbonyl bands of beech sample before and after steaming. Deacetylation reduced the absorption intensity in the whole carbonyl region and the place of maximum was shifted towards lower wavenumbers. Having a closer look at the spectra, it is obvious that the greatest difference between the spectra is not located at the maximum but at 1758 cm−1 (5688 nm). The difference at 1746 cm−1 (5727 nm) is 0.21 units while at 1758 cm−1 it is 0.29 units. This latter is the greatest difference between the spectra generated by steaming. This result represents that the absorption maximum of carbonyl unit in acetyl group is located at 1758 cm−1 (not at the visible

228

5 Applications of IR Spectrum Measurement in Wood Research

maximum). The reduced absorption intensity of the acetyl groups generated the shift of the place of maximum towards shorter wavenumbers. The difference spectrum clearly shows the actual situation (Fig. 5.2). Figure 5.2 shows the difference spectra of beech generated by steaming at 80, 100 and 120 °C. (Wavelength was chosen as independent variable for representing the whole IR spectrum in a single diagram.) This figure shows clearly that the absorption decrease in the carbonyl region is located at 1758 cm−1 (5688 nm). All absorption value changes increased with rising temperature for beech. The greatest absorption decrease was found at 3593 cm−1 (2783 nm). This decrease represents the reduction in the number of hydroxyl groups as a result of hemicellulose degradation (Tjeerdsma

Fig. 5.1 Relative absorbance spectra of the carbonyl region for the investigated species before steaming (a) and relative absorbance spectra of the carbonyl region for beech before and after steaming at 120 °C (b)

Fig. 5.2 Difference IR spectra of beech generated by steaming at 80, 100 and 120 °C for 2 days

5.2 Monitoring the Chemical Changes Caused by Steaming

229

and Militz 2005; Fabiyi and Ogunleye 2015; Timar et al 2016a, b). There is an absorption increase of hydroxyl groups around 3344 cm−1 (2990 nm). This band is assigned as intramolecular hydrogen bond (Popescu et al. 2013). This absorption increase of hydroxyl groups is generated by further chemical changes after deacetylation. The interpretation of the changes in the C–H region is questionable because of the baseline shift. Comparing the spectra of the species studied, a small absorption increase is observed in this range at wavenumbers of 2857 and 2907 cm−1 (3500 and 3440 nm). The absorption of the carbonyl group in the acetyl unit decreased at 1758 cm−1 (5688 nm). This change was accompanied by the absorption decreases at 1478 cm−1 (6766 nm) representing the C–H deformation in xylan and the absorption decrease of C–H deformation in hemicelluloses at 1394 cm−1 (7174 nm). All these absorption decreases clearly present the deacetylation effect of steaming. Deacetylation generates acetic acids in wet condition. Presence of acetic acids result in accelerated degradation of the polysaccharide components of the cell wall (Stamm 1956). The second largest absorption decrease occurred at 1178 cm−1 (8489 nm), indicating ester bridge splitting. There is a wide negative absorption band around 1659 cm−1 (6028 nm) representing the degradation of extractives containing conjugated carbonyl groups. The guaiacyl lignin in beech underwent degradation at 120 °C. This change is demonstrated by the absorption decrease at 1520 cm−1 (6579 nm). This peak for beech is normally found at 1508 cm−1 (6631 nm) but usually shifted towards higher wavenumbers during thermal treatment (Windeisen et al. 2007; Kocaefe et al. 2008; Chen et al. 2012; Esteves et al. 2013). The absorption decrease at 1287 cm−1 (7770 nm) belongs also to the guaiacyl lignin (guaiacyl ring breathing). A broad positive band is visible between 950 and 1300 cm−1 (10,500 and 7692 nm), however, this peak does not represent a real absorption increase. The K–M equation does not provide the absorption spectrum properly for intensive absorption bands if the surface roughness changes. This phenomenon elevates the difference spectrum intensities in this region. The absorption decreases of the ester bridge (asymm.) at 1178 cm−1 (8489 mm) is visible as a valley within two hills at 80 and 100 °C. The absorption decreases of C–O–C stretching (symm.) at 1138 cm−1 (8787 nm) also generates a small valley. The roughness increase modifies the light scattering adjusting the K–M values. A detailed discussion of this phenomenon can be found in a previous work (Tolvaj et al. 2011). Two phenomena cause the increase in surface roughness during steaming. Swelling lifts up the edges of the wood fibres, and these fibres never return to the original position during drying. Steaming results in residual swelling of 3–5%, which changes the surface roughness (see Sect. 3.4). There are two positive bands at 1367 and 1324 cm−1 (7315 and 7553 nm). The real meaning of these peaks is questionable. There are five absorption increases observable at 1119, 1092, 1070, 1045 and 1000 cm−1 (8936, 9157, 9346, 9569 and 10,000 nm). These peaks are visible on the top of a broad and high anomalous band. The intensities and the positions of these five bands are highly questionable because of the superposition. That is why the correct interpretation of these increases is not possible.

230

5 Applications of IR Spectrum Measurement in Wood Research

Figure 5.3 shows the difference spectra of poplar generated by steaming at 80, 100 and 120 °C. Poplar samples presented similar absorption changes to beech, but these alterations were smaller and less temperature dependent. Figure 5.4 shows the difference spectra of spruce generated by two-day steaming at 80, 100 and 120 °C. The changes are smaller compared to deciduous species and the places of maxima are different in some cases. The differences are clearly visible when the spectra of beech and spruce are plotted together (Fig. 5.5). Figure 5.5 represents the difference IR spectra of beech and spruce generated by two-day steaming at 120 °C. The IR spectra clearly show that the deacetylation of hemicelluloses is the main chemical change during steaming. The difference IR

Fig. 5.3 Difference IR spectra of poplar generated by steaming at 80, 100 and 120 °C for 2 days

Fig. 5.4 Difference IR spectra of spruce generated by steaming at 80, 100 and 120 °C for 2 days

5.3 Examination of Chemical Changes Generated by Photodegradation

231

Fig. 5.5 Difference IR spectra of beech (B) and spruce (S) generated by steaming at 120 °C for 2 days

absorption spectra of beech and spruce show great differences. The reason lies in the different hemicellulose content of these species. In hardwoods, highly acetylated xylan is the major polyose component; in softwoods, however, glucose units are rarely acetylated. Absorption changes of beech due to steaming is discussed above (Fig. 5.2). Spruce specimens presented only small absorption decrease at those wavenumbers where beech demonstrated the deacetylation process, indicating a low deacetylation rate of hemicelluloses and the low presence of acetic acids. The experimental results showed that deacetylation of the glucose rings of hemicelluloses was the main change caused by steaming. The hydrothermal treatment resulted in the cleavage of ether bonds in deciduous species.

5.3 Examination of Chemical Changes Generated by Photodegradation Photodegradation is usually the first step in the decomposition of outdoor wooden applications. The UV part of the sun radiation splits of some chemical bonds creating free radicals due to the dehydrogenation, dehydroxylation, demethoxylation, dehyroxymethylation and generate chain scission in cellulose, hemicellulose, and lignin (Hon 2001). The free radicals react with oxygen composing hydroperoxides. Lignin and extractives are good light absorbers. These components suffer the most degradation during UV radiation. Chemical analyses show that the main deterioration of light irradiated wood is associated with the decomposition of lignin (Colom et al. 2003; George et al. 2005; Pandey and Vuorinen 2008; Cogulet et al 2016). The consequence of this photodegradation embodied in drastic changes in the appearance of the

232

5 Applications of IR Spectrum Measurement in Wood Research

wood showing discoloration, lightness decrease, and loss of gloss, roughening and checking of the surface. Rain can then leach out the degradation products, leaving room for further degradation. These processes cause the destruction of the mechanical and physical properties of the surface layer. Wood is opaque to light because of the variety of chromophore groups or systems distributed in its surface components. The penetration depth of light is wavelengthdependent. Longer light waves penetrate deeper. The change in absorption of infrared light can be used as an indicator because its penetration depth is greater than that of UV or visible light. The decrease in absorption directly reflects the cleavage of chemical bounds caused by the absorption of light. This happens only at the place reached by the light. An increase in absorption, on the other hand, indicates newly formed chemical groups. This is a secondary process. In most cases, the degradation process is followed by oxidation. This may affect deeper layer than the one reached by the light. Kataoka et al. (2005) found that there is an inversely proportional relationship between the maximum depth of photodegradation and wood density, which explains why low-density earlywood erodes faster than denser latewood. The wavelength dependence of light penetration was tested by the intensity of the IR band at 1730 cm−1 (Kataoka et al. 2007). Significant changes were detected in the carbonyl band at 1730 cm−1 at depths of up to 70, 100, 150, 250, 300 and 500 μm for specimens irradiated with light of wavelengths of 278, 310, 341, 372, 403 and 434 nm, respectively. Light with wavelengths longer than 434 nm did not affect the intensity of this carbonyl band.

5.3.1 Basic Chemical Changes During Photodegradation The most obvious way to study the effect of photodegradation on wood is to use the sun as a light source. Sun irradiation is, however, not well repeatable because of the continuous change of the irradiation intensity on the ground level. It is therefore advisable to apply artificial light source. In this study, mercury vapour lamp is used as a light source, and its description can be found at the end of Sect. 4.3. Details of the IR spectrum measurement are given at the end of Sect. 5.1. The chemical changes during photodegradation are presented here by the data of beech wood irradiated by mercury lamp at 70 °C. The irradiation period was one day. Figure 5.6 shows the normalised absorption spectra of beech before and after UV irradiation. The normalisation was performed at 1378 cm−1 (7257 nm), meaning that spectra intensities were adjusted to 1.0 at this wavenumber. This C–H band of cellulose is often used as an internal standard because of its high intensity, central position and strong stability. The changes generated by UV irradiation are visible if the spectra of beech measured before and after irradiation are compared. The band intensity of aromatic ring vibrations arising from lignin at 1505 cm−1 (6644 nm) decreased while the intensity of unconjugated carbonyl band between 1680 and 1820 cm−1 (5952 and 5494 nm) increased during the irradiation. The comparison of spectrum lines shows

5.3 Examination of Chemical Changes Generated by Photodegradation

233

Fig. 5.6 Absorbance spectra of beech before and after 24-h UV irradiation

that there are absorption decreases at 1172 and 1134 cm−1 (8532 and 8818 nm) as well representing the ether bond splitting. The band intensity for the aromatic ring vibration at 1596 cm−1 (6266 nm) also decreased but this can be observed in the case of hardwoods only. This peak does not exist in the case of softwoods because it belongs to the so-called syringyl lignin found mostly in hardwood. Examining the width of the bands it was found that the bands at 1596, 1505, 1455 and 1430 cm−1 (6266, 6644, 6873 and 6993 nm) have similar width but the carbonyl band at 1740 cm−1 (5747 nm) is much wider than the others. This suggests that this carbonyl band is a superposition of some individual bands. Nevertheless, most publications treat the band of unconjugated carbonyls as a single absorption band with a maximum around 1740 cm−1 . If we have a closer look at the absorption change of the unconjugated carbonyl band, one can see that the band is shifted towards higher wavenumbers. The maximum moves to 1748 cm−1 . Moreover, the greatest difference between treated and initial spectra appears not at the maximum but at the left side of the band. The difference at the maximum (1740 cm−1 , 5747 nm) is 0.14 and at 1763 cm−1 (5672 nm) it is 0.37 units, which is more than double. Similar band shift is visible at 1134 cm−1 (8818 nm) but towards shorter wavenumbers. These results show that in many cases the biggest changes did not occur at the peak position of the band. These findings highlight the disadvantage of the simple comparison method where the initial and the treated spectra are presented on top of each other. The difference spectrum method solves this problem. Creating the difference spectrum (irradiated minus initial), only those absorption bands appear where a change has occurred. The positive band indicates an increase in absorption, while the negative band indicates a decrease in absorption. Figure 5.7 shows the absorbance difference spectrum of beech generated by 24-h UV irradiation. This spectrum correctly shows the changed bands. The difference spectrum of beech consists of two positive bands at 1763 and 1709 cm−1 (5672 and

234

5 Applications of IR Spectrum Measurement in Wood Research

5851 nm). This result was strengthened by 2D IR spectroscopy (Popescu et al. 2011). The band at 1763 cm−1 represents the absorption of CO stretching for unconjugated ketones and γ lactones generated by the oxidation after the splitting of the aromatic ring. The band at 1709 cm−1 represents the absorption of aliphatic carboxyl groups. There is a broad band around 1654 cm−1 (6046 nm). The conjugated carbonyls in phenolic molecules, the carbonyl groups of quinones as well as the OH groups in water absorb here. The absorption decrease at 1596 cm−1 (6266 nm) indicates the degradation of aromatic ring in syringyl lignin. The negative peak at 1506 cm−1 (6640 nm) belongs to the aromatic skeletal vibration of guaiacyl lignin. This negative peak is detectable together with the absorption decrease of the aromatic C–H deformation at 1469 and 1428 cm−1 (6807 and 7003 nm) and with the absorption decrease of the guaiacyl ring breathing at 1268 cm−1 (7886 nm). The absorption decrease at 1337 cm−1 (7479 nm) represents the degradation of syringyl lignin. The positive peaks between 1200 and 1400 cm−1 (8333 and 7143 nm) might be the result of the baseline shift. The greatest absorption decrease is visible at 1174 and 1139 cm−1 (8517 and 8780 nm). The first decrease belongs to the asymmetric stretching of ether bond in cellulose and in hemicellulose. The second decrease belongs to the symmetric stretching of the ether bond, the aromatic C–H deformation and to the glucose ring vibration. There is an absorption decrease at 1094 cm−1 (9141 nm) which belongs to the C–O–C stretching in cellulose and hemicelluloses. These absorption decreases indicate the ether splitting and the depolymerisation of cellulose. It is important to compare the places of peaks in Fig. 5.7 with the absorption peaks in Fig. 5.6. The comparison shows that in many cases the maximum absorption change does not occur at the place of the original band maximum. Figure 5.8 shows the absorbance difference spectrum of spruce generated by 24-h UV irradiation. (Figs. 5.7–5.9 have the same vertical scaling for correct comparison.)

Fig. 5.7 Absorbance difference spectrum of beech generated by 24-h UV irradiation

5.3 Examination of Chemical Changes Generated by Photodegradation

235

Although the main changes in the spectrum of beech can be observed in the spectrum of spruce, there are also important differences. The absorption decreases belonging to the syringyl lignin are missing. The absorption decrease of guaiacyl lignin is shifted to higher wavenumbers (1510 cm−1 , 6623 nm). This shift is usual for conifers. The increase of the two carbonyl bands produced a broad band. It is the superposition of two bands. The visible peak positions are 1746 and 1720 cm−1 (5727 and 5814 nm), but these are not the real peak positions. The real distance between the two individual peaks is more than 26 cm−1 (see Fig. 1.26). All three bands exist for both species in the C–O–C stretching region, but there are large differences in intensity. The band at 1174 cm−1 (8517 nm) is a shoulder

Fig. 5.8 Absorbance difference spectrum of spruce generated by 24-h UV irradiation Fig. 5.9 Absorbance difference spectra of beech and Scots pine in the OH region generated by 24-h UV irradiation

236

5 Applications of IR Spectrum Measurement in Wood Research

and the band at 1094 cm−1 (9141 nm) is negligible for beech. In contrast, all three bands are visible as individual band with close to equal intensity for spruce. Figure 5.9 shows the absorption changes of beech and Scots pine samples in the hydroxyl and methyl region. The spectra present both absorption decrease and increase in the OH region. The maxima of the absorption decreases are located around 3606 cm−1 (2773 nm). The peaks of absorption increases are at 3326 and 3341 cm−1 (3006 and 2993 nm) for beech and Scots pine, respectively. The negative band presents the absorption decrease of the intramolecular hydrogen bond in the phenolic group in lignin. This intensity loss can also be interpreted as the rupture of intramolecular OH bonds of cellulose, comparable to that of the ether bridge. The absorption decreases in the OH region can also be interpreted by the loss of OH groups followed by dehydration upon UV irradiation. Dehydration belongs to the plausible reactions, in which double bonds—responsible for the coloured compounds—are formed. The formation of esters and lactone rings, also common features of photodegradation, occurs via water loss and OH group consumption (Tolvaj and Faix 1995). The absorption increase around 3341 cm−1 (2993 nm) represents the increase of intermolecular and intramolecular hydrogen bonds (Popescu et al. 2013). There are two small absorption decreases in both side of 2900 cm−1 (3449 nm) belonging to the symmetric and asymmetric methoxyl stretching. Presumably these small decreases are triggered by the demethylation, elimination of CH2 OH groups both from lignin and polysaccharides. The time dependence of photodegradation can be monitored by the difference spectrum method. Figure 5.10 shows the difference spectra of beech wood after 1, 2, 4, 8 and 16 h of UV irradiation by mercury lamp. The lignin degradation was continuous demonstrated by the absorption decrease at 1506 and 1596 cm−1 (6640 and 6266 nm). The negative peak at 1654 cm−1 (6046

Fig. 5.10 Difference spectra of beech after 1, 2, 4, 8 and 16 h of UV irradiation by mercury lamp

5.3 Examination of Chemical Changes Generated by Photodegradation

237

nm) decreased during the first 2 h of irradiation, then remained almost constant. The two carbonyl peaks around 1709 and 1763 cm−1 (5851 and 5672 nm) increased parallel. The growing of the peak at 1763 cm−1 (5672 nm) was a little faster than the growing of the nearby peak. The most remarkable change occurred at 1174 and 1139 cm−1 (8517 and 8780 nm). The decrease of the peak at 1174 cm−1 was faster at the beginning but it slowed down after 4 h. The other peak at 1139 cm−1 decreased continuously and has overtaken its neighbour at the 16th hour of the irradiation. The difference between the two peaks continued to increase after 16 h of irradiation (see Fig. 5.7). There are differences in the effects of photodegradation depending on the wood species. Figure 5.11 shows the absorption changes of beech, poplar, ash and Scots pine samples caused by mercury lamp irradiation for 16 h at 70 °C. (The spectrum of Scots pine is presented here only for comparison.) Scots pine specimens contained earlywood in heartwood tissue on the measured surface (tangential surface). The absorption decrease of guaiacyl lignin around 1508 cm−1 (6631 nm) is similar for all species. The absorption decrease of syringyl lignin at1596 cm−1 (6222 nm), however, is missing in case of conifer. The reason is that conifers hardly have syringyl lignin. It is interesting to see that the absorption decrease of aromatic C–H deformation at 1469 (6807 nm) is much smaller for Scots pine than for deciduous species. This finding suggests that the decrease of these bands is mainly determined by the degradation of syringyl lignin rather than that of the guaiacyl lignin. The absorption increase in the region of carbonyl groups (1670–1810 cm−1 5988– 5525 nm) is mainly originated from the degradation products of lignin. Therefore, there are differences between conifers and deciduous species in the carbonyl region as well. Beech and ash have two separate maxima at 1709 and 1763 cm−1 (5851 and 5672 nm). Poplar, like beech and ash, also has two maxima, but one is visible

Fig. 5.11 Difference spectra of Scots pine (earlywood) beech, poplar and ash generated by 16-h UV irradiation

238

5 Applications of IR Spectrum Measurement in Wood Research

as a shoulder. Scots pine earlywood has two positive bands, but only one around 1709 cm−1 is in similar position comparing to the band of deciduous species. The other peak is located closer, creating a visually integrated band with a maximum at 1747 cm−1 (5724 nm) and with a shoulder at 1720 cm−1 (5814 nm). There are differences among the investigated species regarding the ether bond region as well. The absorption decrease around 1097 cm−1 (9116 nm) is similar for all species, but the decrease around 1174 cm−1 (8518 nm) is much smaller for Scots pine than for deciduous species. Here, ash showed twice the absorption loss of the other deciduous species. The positive band at 1119 cm−1 (8936 nm) is visible only for beech, and it seems that this positive band is not a real peak. Probably the baseline shift lifted up the spectrum. All other spectra show a valley between two negative bands here. Figure 5.12 shows the absorption changes of Scots pine, spruce, larch and beech samples caused by mercury lamp irradiation for 16 h at 70 °C. (The spectrum of beech is presented here only for comparison.) Scots pine and larch specimens contained earlywood in heartwood tissue on the measured surface (tangential surface). The measured surface of spruce samples contained earlywood. Syringyl band is missing for all conifer species. Larch samples presented considerably smaller changes in the whole fingerprint region compared to the other species. This result proves the high resistance of larch wood to photodegradation. Three positive bands rose for conifers in the unconjugated carbonyl region. These bands are highly overlapped. Consequently, the real peak positions are not observable. The real peak positions are farther from each other than the visible positions (see Fig. 1.26). The visible peak positions are at 1747 and 1720 cm−1 (5724 and 5814 nm) for both spruce and Scots pine samples. The spectra of conifers have a shoulder at 1680 cm−1 (5952 nm). It is an important result that the absorption decrease of aromatic C–H deformation at 1469 (6807 nm) is much smaller for conifers than for deciduous species. In terms of photodegradation, there are differences in the sensibility of earlywood and latewood. Latewood usually has greater protecting ability than the earlywood (Cirule et al. 2022). Similar results were found by Forsthuber et al. (2014) measuring the colour change differences between earlywood and latewood generated by xenon lamp exposure. Kanbayashi et al. (2021) studied the degradation behaviour of chemical components in Japanese cedar sapwood and heartwood during natural weathering using confocal Raman microscopy. Spectral and imaging analyses revealed that Japanese cedar heartwood is more weather-resistant than sapwood due to its high extractives content. Figure 5.13 shows the degraded surface of an outdoor temple column exposed for hundreds of years. Deep valleys are visible presenting the different degradability of the individual tissues. Apparently, earlywood is much more eroded than the latewood. The transition from earlywood to latewood shows the continuous degradability decrease. In contrast, the transition from latewood to earlywood presents a sharp border. This degradability profile corresponds to the density profile of a year-ring. The knot shows excellent stability against outdoor weathering as well. Figure 5.14 demonstrates the absorption change difference between poplar earlywood and latewood. The lignin in earlywood is more degradable than that in the

5.3 Examination of Chemical Changes Generated by Photodegradation

239

Fig. 5.12 Difference spectra of Scots pine, spruce, larch earlywood and beech generated by 16-h UV irradiation

Fig. 5.13 Degraded surface of a column of an old Japanese temple

latewood. The absorption decreases of both syringyl (at 1595 cm−1 , 6270 nm) and guaiacyl (at 1507 cm−1 , 6636 nm) lignin are smaller in latewood than in earlywood. As a consequence, the absorption increase of carbonyl groups is smaller in latewood than in earlywood. Also, the reduction of ether bridges is highly greater for earlywood than for latewood. The change of the IR spectra confirms that the UV degradability of latewood is considerably smaller than that of the earlywood. Conifers have somewhat different photodegradation behaviour than deciduous species. The photodegradation properties of spruce are discussed here according to a previous paper (Preklet et al. 2021a, b) with written permission of Acta Silvatica

240

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.14 Absorption change of earlywood (E) and latewood (L) for poplar (P) caused by 16 h UV irradiation

et Lignaria Hungarica. Figure 5.15 shows the changes in absorption differences of spruce earlywood and latewood generated by one-day and three-day UV irradiation. The conjugated carbonyls (around 1660 cm−1 ) in latewood suffered greater degradation than in earlywood. This phenomenon represents that latewood contains a higher quantity of extractives than earlywood. This broad negative peak pulled down the partly overlapped positive peak at 1706 cm−1 (5862 nm) in case of latewood. The negative peak at 1510 cm−1 (6623 nm) belongs to the aromatic skeletal vibration of guaiacyl lignin. This negative peak is detectable together with the absorption decrease of the aromatic C–H deformation at 1470 and 1430 cm−1 (5747 and 6993 nm) and with the absorption decrease of the guaiacyl ring breathing at 1270 cm−1 (7874 nm). The greatest absorption decrease was visible at 1174 and 1133 cm−1 (8518 and 8826 nm). The first decrease belongs to the asymmetric stretching of ether bond in cellulose. The second decrease belongs to the symmetric stretching of ether bond, the aromatic C–H deformation, and to the glucose ring vibration. These absorption decreases indicate the ether splitting and the depolymerisation of cellulose. Free radicals were generated during lignin degradation. These free radicals react with oxygen to produce carbonyl groups. The absorption increase of unconjugated carbonyls is visible in the 1680–1820 cm−1 wavenumber (5952–5495 nm) interval. Two bands emerged in this region for both earlywood and latewood around 1705 and 1764 cm−1 wavenumbers (5865 and 5669 nm). The band at 1764 cm−1 represents the absorption of CO stretching for unconjugated ketones and γ lactones generated by the oxidation after the splitting of the aromatic ring. The band at 1705 cm−1 represents the absorption of aliphatic carboxyl groups. Although the place of these peaks should be the same, the intensities and the apparent places of the peaks are different for earlywood and latewood. Latewood produced smaller absorption increase than earlywood. Great intensity difference was found between the two peaks. The peak

5.3 Examination of Chemical Changes Generated by Photodegradation

241

Fig. 5.15 Absorption change of earlywood (E) and latewood (L) for spruce (S) caused by one and three days of UV irradiation

intensity at 1705 cm−1 was small compared to the neighbouring peak. The two bands are well separated because of the low intensities. In contrast, the peak intensity at 1705 cm−1 for earlywood is almost equal to that of the neighbouring peak at 1764 cm−1 . The superposition of the two bands is visible in Fig. 5.15. The real positions of the peaks are not visible since the superposition pulled the locations of the peaks toward each other. These two bands finally joined into one single band after 9-day UV irradiation (Fig. 5.16). The negative intensity changes of the peak at 1510 cm−1 (6623 nm) increased during UV irradiation. (Time dependence of this intensity will be discussed later.) The intensity of the two types of ether bond at 1174 and 1133 cm−1 (8518 and 8826 nm) decreased, but in different ways for earlywood and latewood during UV irradiation. The two negative peak intensities were almost equal for earlywood. The peak at 1133 cm−1 was a little higher than the peak at 1174 cm−1 . Latewood showed opposite peak intensities. The peak at 1174 cm−1 presented the greatest negative change and the peak at 1133 cm−1 was visible as a shoulder. There was a visible positive peak at 1068 cm−1 . This positive peak is associated with the C–O–C stretching in cellulose and hemicelluloses. The prolonged treatment time intensified the absorption changes. Figure 5.16 shows the absorption change of earlywood and latewood for spruce caused by 13- and 20-day UV irradiations. The absorption intensities of the two types of unconjugated carbonyl groups at 1705 and 1764 cm−1 wavenumbers (5865 and 5669 nm) increased with prolonged irradiation time. The two bands composed a single band after nine days of irradiation for earlywood, which remained up to the end of treatment. The peak intensity at 1705 cm−1 (5865 nm) was growing faster than the peak intensity at 1764 cm−1 (5669 nm) for latewood.

242

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.16 Absorption change of earlywood (E) and latewood (L) for spruce (S) caused by 13- and 20-day UV irradiation

The evaluation of the changes in the ether bond region is difficult because the Kubelka–Munk equation does not provide the absorption spectrum properly if the absorption is high and the surface roughness changes. A previous study showed that photodegradation increases the surface roughness of wood (Tolvaj et al. 2014). The roughness increase lifts up the intensities due to the increasing scattering. A detailed discussion of this phenomenon can be found in a previous study (Tolvaj et al. 2011). The lifting effect overlaps the real absorption changes, interfering with the evaluation of IR spectrum in the 1000–1200 cm−1 wavenumber interval (10,000–8333 nm). Although Figs. 5.15 and 5.16 present the absorption decrease of guaiacyl lignin, these figures do not clearly show the intensity changes of this negative peak due to overlapping. Figure 5.17 reveals the absorption change at 1510 cm−1 (6623 nm) in all investigated situations. The results show that earlywood suffered greater lignin degradation than latewood. The higher extractive content in latewood provided greater protection for lignin than the lower extractive content in earlywood. The protecting effect of the extractives was demonstrated in previous studies (Nemeth et al. 1992; Pastore et al. 2004; Tolvaj and Varga 2012; Varga et al. 2020). Lignin degradation was fast at the beginning of the treatment and stopped after 11 days of UV irradiation for both earlywood and latewood. The reason could be that most of the lignin molecules of the examined surface layer degraded during the first 11 days of UV irradiation. Figure 5.18 presents the intensity changes of the unconjugated carbonyl groups absorbing at 1764 cm−1 (5669 nm). Absorption intensity increased continuously for both earlywood and latewood up to the thirteenth day of the applied UV irradiation, and decreased slightly afterward. The decrease after the thirteenth day of the treatment might not be a real decrease. The values were read exactly at 1764 cm−1 wavenumber; however, changes in the overlapping band at the 1705 cm−1 wavenumber (5865 nm) probably modified the real values. UV irradiation generated greater absorption increase for earlywood than for latewood. The tendency of

5.3 Examination of Chemical Changes Generated by Photodegradation

243

Fig. 5.17 Absorption band intensity change of guaiacyl lignin in spruce (S) earlywood (E) and latewood (L) at 1510 cm−1 wavenumber (6623 nm) generated by UV irradiation (in days)

the absorption changes at 1764 cm−1 was a mirror image of lignin degradation at 1510 cm−1 (6623 nm). This fact demonstrates the correlation between lignin degradation and the generation of unconjugated carbonyl groups absorbing at 1764 cm−1 . Many studies deal with the correlation between lignin degradation and the generation of new unconjugated carbonyl groups (Pandey 2005; Agresti et al. 2013; Timar et al. 2016b; Bonifazi et al. 2016; Reinprecht et al. 2018). All of these publications, however, use the complete integrated unconjugated carbonyl band as the participant in the correlation. Our results demonstrate that only one component of the unconjugated carbonyl band (at 1764 cm−1 wavenumber) is decisive in this correlation. Figure 5.19 shows the irradiation time dependence of the unconjugated carbonyls absorbing at 1705 cm−1 (5865 nm). Earlywood presented rapid absorption increase at the beginning of the UV irradiation, followed by a moderate increase up to the end of the treatment. Latewood produced only a small absorption increase during the first day of irradiation. This small value was followed by continuous absorption increase throughout the whole investigated treatment period. In the second part of the treatment, latewood produced more intensive absorption increase than the earlywood. The absorption increase at 1705 cm−1 does not show similar time dependence than the lignin degradation. This finding raises the question of whether the unconjugated carbonyls absorbing at 1705 cm−1 are derived from the degradation of lignin. This phenomenon requires further chemical investigations. For a more detailed comparison, Fig. 5.20 represents the differences between the absorption properties of the two types of unconjugated carbonyl groups absorbing at 1705 and 1764 cm−1 (5865 and 5669 nm) for latewood. The time dependence of these two absorption increases is completely different. The band at 1764 cm−1 showed a rapid increase at the beginning of the UV irradiation. The change of the absorption

244

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.18 Absorption band intensity change of the unconjugated carbonyls at 1764 cm−1 (5669 nm) for spruce (S) earlywood (E) and latewood (L) generated by UV irradiation (in days)

Fig. 5.19 Absorption band intensity change of the unconjugated carbonyls at 1705 cm−1 (5865 nm) for spruce (S) earlywood (E) and latewood (L) generated by UV irradiation (in days)

increase produced a maximum on the thirteenth day of the treatment. In contrast, the band at 1705 cm−1 presented continuous increase during the whole investigated period. This difference in the growing tendency shows that the generation of these two types of unconjugated carbonyl groups (at 1705 and 1764 cm−1 ) has two different pass ways or different origins.

5.3 Examination of Chemical Changes Generated by Photodegradation

245

Fig. 5.20 Absorption band intensity change of the unconjugated carbonyls at 1705 and 1764 cm−1 wavenumbers (5865 and 5669 nm) for spruce (S) latewood (L) generated by UV irradiation (in days)

The time dependence of the two absorption increases in the unconjugated carbonyl region was similar up to the 13th day of irradiation in case of earlywood. The intensity of the band at 1764 cm−1 (5669 nm) decreased while the intensity of the band at 1705 cm−1 (5865 nm) increased after this period. The artificial simulation of sun radiation is a big scientific challenge. Imitating the sun radiation, the emitted intensity distribution must be as close as possible to the emission spectrum of sun radiation. None of the artificial light sources emit exactly the same intensity distribution as the Sun. The emission spectrum of newly designed light sources is getting closer and closer to that of the Sun. Xenon lamp has similar emission spectrum as sun radiation. The intensity distribution of xenon lamp can be adjusted by filters to properly simulate the emission spectrum of the Sun in the visible wavelength region. The only shortcoming is the missing UV part. The Xenon lamp hardly emits radiation below 350 nm. Scientists must keep in mind that there are several types of xenon lamps having different emission intensity distribution. Although, it is possible to calculate an irradiation time equivalent between the effect of sun radiation and the effect of xenon lamp radiation, this is only valid for the same lamps. It is interesting to study the short-term, medium long-term and long-term equivalences. The long-term equivalence is meaningless. If sunlight can destroy all the lignin molecules on a wooden surface in 500 h and a mercury lamp can do so in 50 h, it does not mean that the effect of the 10-h mercury lamp exposure is the same as that of the 100-h sun exposure. The degradation of a molecule follows an exponential decrease, slowing down the rate of degradation with elapsed time. The change becomes extremely slow before all lignin molecules are degraded. In the example above, this means that the 400-h sunlight irradiation generates almost the same lignin degradation as the 500-h irradiation.

246

5 Applications of IR Spectrum Measurement in Wood Research

The next example confirms that the general validity of a calculated accelerating factor is questionable. Black locust samples were irradiated by sunlight and xenon lamp light (Tolvaj and Mitsui 2005). The sun-test was carried out in the summer in Takayama city, Japan. Geographical data of Takayama are: 36 degrees 9.3 min latitude and 560 m altitude. The test period was between 5 May and 19 August. The air temperature varied between16 and 41 °C, maximum RH was 80% and the daily average of total solar power density was between 436–459 W/m2 . Samples were outside between 9 am and 3 pm, but only in sunny days. The xenon lamp test was carried out in a SX-75 chamber (Suga Test Instruments Co. Ltd., Tokyo). The power density of the xenon lamp was 180 W/m2 , in the range of 300–400 nm. There was a quartz glass filter around the lamp and the temperature in the chamber was 63 °C. Figure 5.21 and 5.22 show the degradation effects of sun and xenon lamp irradiations for black locust. The usual changes are visible generated by photodegradation. The 60-h xenon light irradiation produced greater lignin degradation than the 60-h sunlight irradiation and there were deviations in the unconjugated carbonyl region as well. The xenon light irradiation produced equal increase for both absorption bands in the unconjugated carbonyl region. In contrast, the sunlight generated absorption increase was twice as hight at 1769 cm−1 than at 1705 cm−1 (5653 and 5865 nm). The difference between the effect of sunlight and xenon light decreased during further irradiation. The effect of xenon light proved to be equal to the effect of sunlight after 200 h of irradiation (Fig. 5.22). There is only one visible difference in the absorption region of guaiacyl lignin around 1507 cm−1 (6638 nm), however it is not a real absorbance difference. The spectrum of sun irradiation shows baseline shift in the 1600–1400 cm−1 (6250–7134 nm) region. These results show that there is no general accelerating factor to exactly match two different types of irradiations.

Fig. 5.21 Difference IR spectra of black locust wood caused by 60-h light irradiation

5.3 Examination of Chemical Changes Generated by Photodegradation

247

Fig. 5.22 Difference IR spectra of black locust wood caused by 200-h light irradiation

Figure 5.23 shows the difference absorbance spectra of beech generated by 200-h sun irradiation and 16-h mercury lamp irradiation at 80 °C. The figure represents that it is possible to find special treatment parameters to equate the effects of different treatments. Figure 5.23 shows two deviations. The mercury lamp irradiation (data of mercury lamp are described at the end of Sect. 4.3) generated greater degradation at 1034 cm−1 (9671 nm) and in the conjugated carbonyl region at 1654 cm−1 (6046 nm) compared to sun radiation.

Fig. 5.23 Difference IR spectra of beech wood caused by 200-h sun irradiation and 16-h mercury lamp irradiation at 80 °C

248

5 Applications of IR Spectrum Measurement in Wood Research

These results confirm that the mercury lamp is suitable to study the accelerated photodegradation properties of wood if the treatment time is short enough.

5.3.2 Effect of Temperature The superposition effects of two or more influential parameters were hardly examined in terms of the degradation of wood. Matsuo et al. (2010, 2014) studied the time– temperature superposition during the thermal degradation of wood in the temperature range 90–180 °C. The effect of elevated temperature during photo-irradiation (the superposition of light irradiation and thermal treatment) is also a rarely investigated phenomenon, even though the surface temperature of wood rises considerably during light irradiation. Persze and Tolvaj (2012) irradiated wood samples using mercury vapour lamp at 80 and 30 °C to determine the effect of thermal decomposition during photodegradation. Their results demonstrated that the same light irradiation resulted in considerably greater redness increase at 80 °C than at 30 °C. It was found that the extractive content has an important role in thermal decomposition during photodegradation. Tolvaj et al. (2013) studied the photodegradation properties of wood species at 30 and 80 °C by measuring the IR reflectance spectra. It was found that the degree of degradation was considerably larger at 80 °C than at 30 °C. Remarkable differences were found between the photodegradation behaviours of softwoods and hardwoods and the deviations were temperature dependent. The temperature dependence of photodegradation for wood is hardly investigated above 100 °C. Only one paper deals with the colour modification effect of photodegradation above 100 °C (Tolvaj et al. 2015) and another paper investigated the chemical changes generated by the photodegradation at 30, 80, 120 and 160 °C (Varga et al. 2017). Research was carried out to study the chemical changes in wood due to light irradiation at elevated temperatures using diffuse reflectance IR spectroscopy. To assess the influence of heat during photodegradation, samples were irradiated by UV light at 30, 80, 120, and 160 °C. In the study, heat treatment and light irradiation were applied simultaneously, just as in nature during sun radiation. The investigated hardwood species were beech (Fagus sylvatica L.) and poplar (Populus x euramericana cv. Pannonia). Softwoods included Scots pine (Pinus sylvestris L.) and spruce (Picea abies Karst.). The samples of different series were prepared from the same board to minimalize the effect of inhomogeneity. The light irradiation was provided by mercury vapour lamp (described at the end of Sect. 4.3). The irradiation chamber was able to cool and heat the inner space. The effectiveness of irradiation was tested on spruce samples to determine the upper limit of the treatment time. The irradiation was interrupted after 1; 2; 7 and 16 h to measure the IR spectra. The results demonstrated that 16 h of light exposure was sufficient to generate measurable changes. Another series of samples were treated in the same chamber set to 30, 80, 120 and 160 °C but without light irradiation. The effect of pure thermal degradation

5.3 Examination of Chemical Changes Generated by Photodegradation

249

Fig. 5.24 Difference IR spectra of spruce caused by mercury lamp irradiation at 160 and 30 °C

was determined based on this latter experiment. The total light irradiation and heat treatment time was 16 h in all cases. All IR measurements were performed on the tangential surface of the specimens. The surface of spruce and Scots pine samples contained only earlywood. Details of the IR measurement can be found at the end of Sect. 5.1. The results of the investigations are presented here according to an earlier published paper (Varga et al. 2017) with written permission from Elsevier. Spruce samples were used to determine the optimum irradiation time at the applied highest temperature (160 °C). Figure 5.24 shows the difference IR spectra of spruce caused by mercury lamp irradiation at 160 °C as function of irradiation time. (Wavelength values are not indicated in the figures because too many numbers together would be confusing.) It was found that 16-h irradiation generates enough changes at 160 °C and this time interval is enough also for the lower temperatures. The figure presents the spectrum of spruce generated by the same light source at the smallest temperature (30 °C) during 16-h treatment for comparison. Most of the absorption changes increased intensively by prolonged irradiation time, at 160 °C. The absorption at 1510 cm−1 (6623 nm) decreased representing the degradation of guaiacyl lignin. This change was measurable together with the absorption decrease of the guaiacyl ring breathing at 1267 cm−1 (7837 nm). As a consequence, the unconjugated carbonyl band increased between 1680 and 1820 cm−1 (5952 and 5495 nm). This band consists of two visible bands and a shoulder at 1740, 1724 and 1690 cm−1 (5747, 5800 and 5917 nm), respectively. These bands represent the absorption of CO stretching for unconjugated ketones and aliphatic carboxyl groups. The real existence of two bands is visible after 7-h irradiation at 160 °C. The unconjugated carbonyl band seems to be one band at shorter irradiation periods and at 30 °C. However, the width of the band is much greater than the width of the lignin band at 1510 cm−1 (6623 nm). This fact assumes the existence of sub bands. There are two absorption decreases visible at 1174 and 1139 cm−1 (8518 and 8780 nm). The first decrease belongs to the asymmetric stretching of the ether bond while the second one belongs to the symmetric stretching of the ether bond,

250

5 Applications of IR Spectrum Measurement in Wood Research

the aromatic C–H deformation and to the glucose ring vibration. These absorption decreases indicate the ether splitting and the degradation of cellulose. There is an intensive positive peak at 1066 cm−1 (9381 nm) and a small peak at 1035 cm−1 (9662 nm). These bands are visible at relatively long irradiation times (above 7 h) and not visible below 80 °C. This instance of absorption increase, as a result of the combined effect of photodegradation and thermal treatment, was not described before. These positive peaks are associated with the C–O and C–C stretching in cellulose and hemicelluloses. Popescu et al. (2013) found that the intensity increases of the band at 1058 cm−1 indicates the formation of aliphatic alcohols during thermal treatment at 140 °C. The deviations within the wavenumbers are originated from the different determination techniques. The difference spectrum method makes visible the individual changing bands. In contrast, the measured absorbance spectrum presents the integrated spectrum with maximum between the maxima of the individual bands. The number of conjugated carbonyl groups occurring in phenolic molecules and quinones increased during photodegradation. This increase is represented by the peak at 1596 cm−1 (6266 nm). These groups are partly responsible for the colour change. This intensity increase proved to be relevant only for softwood species. There was a small negative band visible at 1642 cm−1 (6090 nm) but only at 30 °C and at 160 °C in case of short irradiation periods (1- and 2-h). Probably, this negative band exists at longer irradiation periods too, but the growing two neighbouring positive bands overlap it. There are some further small bands not discussed here. It is important to mention that the validity of small bands is questionable. In some cases, the baseline may slightly differ from zero line generating unreal bands. The absorption bands analysed above show clear time dependence. All visible band intensities increased or decreased with elapsed time. On the other hand, the same light irradiation generated much greater absorption change at 160 °C than at 30 °C. The absorption change caused by 16-h UV irradiation at 30 °C was equal to (or smaller than) the absorption change caused by one hour UV irradiation at 160 °C. This shows that the elevated temperature accelerates the degradation effects of light irradiation. Figures 5.25, 5.26 and 5.27 show the temperature dependences of the absorption changes for Scots pine, poplar and beech, respectively. The irradiation time was 16-h in all cases. The spectra of spruce are similar to the spectra of Scots pine (only partly presented in Fig. 5.24). Significant differences were found between hardwoods and softwoods in terms of the difference IR spectra. The most noticeable differences are in the unconjugated carbonyl band region. Hardwoods have two more separated bands compared to softwoods. These two bands were integrated in one band at 30 and 80 °C for conifers after 16-h UV irradiation. The elevated temperature partly separated the bands in the unconjugated carbonyl region for Scots pine (Fig. 5.25) and for spruce as well (Fig. 5.24). The band at 1748 cm−1 (5721 nm) was growing faster than the band at 1724 cm−1 (5800 nm) for Scots pine by increasing temperature. The unconjugated carbonyl bands of poplar were partly separated at low temperatures (Fig. 5.26). The band at 1698 cm−1 (5889 nm) is visible only as a shoulder at

5.3 Examination of Chemical Changes Generated by Photodegradation Fig. 5.25 Difference IR spectra of Scots pine generated by 16 h of irradiation by a mercury lamp as a function of temperature

Fig. 5.26 Difference IR spectra of poplar generated by 16 h of irradiation by a mercury lamp as a function of temperature

Fig. 5.27 Difference IR spectra of beech generated by 16 h of irradiation by a mercury lamp as a function of temperature

251

252

5 Applications of IR Spectrum Measurement in Wood Research

30 °C. The intensity of this band was growing slowly by increasing temperature. In contrast, the intensity of the other band around 1760 cm−1 (5682 nm) was growing fast between 30 and 80 °C and the increase slowed down afterward. The peak position moved towards higher wavenumbers with increasing temperature. The maximum is at 1754 cm−1 (5701 nm) in case of the treatment at 30 °C, and it shifted to 1763 cm−1 (5672 nm) at 160 °C. It means that the band around 1760 cm−1 is not a single band but it is the superposition of two bands. The left-side band rose faster than the rightside band pushing the united peak position towards higher wavenumbers. The real peak positions of the two individual bands are probably around 1770 and 1750 cm−1 (5650 and 5714 nm). Here the band at 1770 cm−1 emerged faster with increasing temperature. This phenomenon is visible as shift of the united band (see Fig. 1.27). The two main unconjugated carbonyl bands are more separated for beech than for other investigated species at all applied temperatures (Fig. 5.27). Here also the leftside band was growing faster between 30 and 80 °C. The right-side band showed only a little intensity increase with increasing temperature. The peak position of left-side band moved towards higher wavenumbers by increasing temperature representing that it is an integrated band with individual peaks at 1770 and 1750 cm−1 (5650 and 5714 nm). The band intensities and peak positions showed that the intensity of the individual band at 1750 cm−1 (5414 nm) was small and did not show considerably increase by elevated temperature. The peak position of the left-side main band was 1760 cm−1 (5682 nm) at 30 °C. It is the middle position between 1500 and 1700. It suggests that the intensities of the two individual bands are close to equal at 30 °C. The calculated maximum values of these each individual bands are less than 0.09 units. Knowing this intensity value of the individual band at 1750, it is not surprising that the peak position of the united band is at 1769 cm−1 (5653 nm) after the irradiation at 160 °C. This phenomenon shows that the individual band at 1700 cm−1 is strongly dominant in case of an irradiation above 30 °C. The results show that four bands around 1770, 1750, 1720 and 1690 cm−1 (5650, 5714, 5814 and 5917 nm) emerge in the unconjugated carbonyl region as the consequence of the oxidation of the lignin degradation products during the photodegradation. New results show that the intensity changes of the right-side integrated band (around 1700 cm−1 , 5882 nm) do not correlate with the intensity changes of lignin bands. This band is attributed to the photodegradation of hemicelluloses (Preklet et al. 2021b; Hofmann et al. 2022). The band at 1770 cm−1 is weak for softwoods, and the band at 1720 cm−1 hardly exists for hardwoods. In addition, water leaching reduced the IR spectrum intensity of natural (untreated) beech wood at 1750 cm−1 demonstrating that this band exists originally in wood (Csanady et al. 2015). In the literature of wood photodegradation the unconjugated carbonyl band is usually mentioned as one band with a maximum at around 1740 cm−1 (5747 nm) and the maximum is usually shifted towards the higher wavenumbers during photodegradation. However, the reality is that the unconjugated carbonyl band is a complex band and the location of the maximum is determined by the intensity ratios of the sub bands. If the intensity ratio changes, the place of maximum will shift. The conjugated carbonyls in lignin and in extractives, in phenolic molecules and in quinone carbonyl groups absorb between 1590–1690 cm−1 (6289–5917 nm). The

5.3 Examination of Chemical Changes Generated by Photodegradation

253

degradation of extractives decreases the absorption, the creation of new phenolic and quinoid structures increases the absorption, while the degradation of syringyl lignin (found mostly in hardwoods) generates absorption decrease in the conjugated carbonyl region. Recently published results show that the slight decrease of 1640 cm−1 (6098 nm) band for Chinese ash could be attributed to thermally induced rearrangements in the structure of lignin as well (Xin et al. 2017). The conifers show temperature dependent absorption increase in the conjugated carbonyl region. This is only a small absorption increase around 1596 cm−1 for Scots pine (Fig. 5.25), but intensive for spruce (Fig. 5.24). In the 1590–1690 cm−1 region, beech and poplar present a broad decrease with two minimums at 1653 and at 1596 cm−1 (6050 and 6266 nm). The absorption decrease at 1596 cm−1 shows the degradation of aromatic ring in syringyl lignin. The complexity of changes is clearly visible for poplar (Fig. 5.26) where the absorption increases at 160 °C lifted up the negative band at 1595 cm−1 . The other difference between hardwoods and softwoods is apparent in the ether bridge splitting at 1175 and 1139 cm−1 (8511 and 8780 nm). The negative intensity of the peak at 1175 cm−1 is smaller than the same of peak at 1139 cm−1 for softwoods, but it is just the opposite for beech (it is close to equal for poplar). The band at 1116 cm−1 (8961 nm) shows an intensity increase, attributed to glucose ring stretching vibration in cellulose and to aromatic C–H vibration in lignin. It is visible only for hardwoods. This effect may be due to the percentage increase of crystalline cellulose due to the cleavage and dehydration of amorphous carbohydrates and/or crystallization of the para-crystalline region of cellulose (Forsthuber et al. 2013). The meaning of peaks between 1095 and 1120 cm−1 (9132 and 8929 nm) is questionable because these develop completely opposite positions for softwoods than for hardwoods. The absorption change in 1025–1030 cm−1 (9756–9709 nm) region was attributed to C–O ester stretching vibrations in methoxyl and β-O-4 linkages in lignin. Kubovsky and Kacik (2014) irradiated lime wood with IR laser, and the increase in absorption around 1030 cm−1 was identified as the deformations of the pyranose ring in cellulose. The aromatic ring vibrations of lignin around 1508 cm−1 (6631 nm) show absorption decrease. The light irradiation generated much larger decrease at 80°C than at 30°C. Higher temperatures could be expected to produce even greater changes, but this is not the case above 80 °C. The reason could be that the thermal effect created chemicals with structure similar to lignin. Earlier works (Forsthuber et al. 2013; Tjeerdsma et al. 1998; Boonstara and Tjeerdsma 2006; Chen et al. 2012; Kacikova et al.2013; Kacik et al. 2023) reported absorption increase due to thermal treatment in the 1508 cm−1 region, as consequence of the splitting of aliphatic side chains, the cleavage of β-O-4 linkages in the lignin structure, followed by the condensation reactions. Our results also show a small absorption increase in the 1508 cm−1 region generated by pure thermal treatment at 160 °C (as it will be described later in the discussion of Figs. 5.30 and 5.31, dotted lines). These negative and the positive changes partly counterbalance each other. This might be the reason why the negative band intensity at 1508 cm−1 did not increase with rising temperature above 80 °C.

254

5 Applications of IR Spectrum Measurement in Wood Research

There is absorption decrease and absorption increase in the OH region (3000–3700 cm−1 , 3333–2703 nm) observed in Scots pine in Fig. 5.28, and in beech in Fig. 5.29. Both are wide bands as a result of the superposition of many individual absorption changes belonging to the OH groups located at different positions. The maximum of the absorption decrease is around 3600 cm−1 (2778 nm) and the maximum of the absorption increase is between 3330 and 3350 cm−1 (3003 and 2985 nm). The absorption decrease is considerably larger up to 80 °C than the absorption increase. At higher temperatures the positive band grows fast with increasing temperature. This positive band expands not only in height but in width as well. It becomes so wide that it eliminates the right side of the neighbouring negative band of absorption decrease. The broad band around 3600 cm−1 presents the absorption decrease of the intramolecular hydrogen bond in the phenolic group in lignin. This absorption decrease can partly be the effect of drying, (i.e., the elimination of the water in wood). This intensity loss can also be interpreted as the rupture of intramolecular OH bonds of cellulose, comparable to that of the ether bridge. The absorption decreases in the OH Fig. 5.28 Difference IR spectra of Scots pine in the OH and CH region generated by 16 h of irradiation by a mercury lamp as a function of temperature

Fig. 5.29 Difference IR spectra of beech in the OH and CH region generated by 16 h of irradiation by a mercury lamp as a function of temperature

5.3 Examination of Chemical Changes Generated by Photodegradation

255

Fig. 5.30 Difference IR spectra of spruce generated by the mercury lamp and thermal treatment during 16-h irradiation (Figs 5.30 and 5.31 are in wrong positions

Fig. 5.31 Difference IR spectra of poplar generated by the mercury lamp and thermal treatment during 16-h irradiation

region can also be interpreted by the loss of OH groups followed by dehydration upon UV irradiation. Dehydration belongs to the plausible reaction, in the course of which double bonds—responsible for coloured compounds—are formed. The formation of esters and lactone rings, also common features of photodegradation, occurs via water loss and OH group consumption (Tolvaj and Faix 1995). The band around 3340 cm−1 (2994 nm) is a mixture of intermolecular and intramolecular hydrogen bonds (interand intramolecular H-bonds in cellulose 1 and lignin, O3-H3…O5 intramolecular Hbonds, O2–H2…O6 intramolecular H-bonds, O6–H6…O3’ intermolecular H-bonds in carbohydrates, aliphatic and phenolic inter- and intra-molecular hydrogen bonds in lignin) (Popescu et al. 2013). There are two instances of small absorption decrease in the CH region between 2840 and 2940 cm−1 (3521 and 3401 nm) belonging to the symmetric and asymmetric methoxyl stretching. These small changes are thought to be triggered by the demethylation, elimination of CH2 OH groups both from lignin and polysaccharides. There is an undefined lifting effect visible in this region at 120 and 160 °C.

256

5 Applications of IR Spectrum Measurement in Wood Research

The effects of discrete and simultaneous UV and thermal treatments can be seen in Figs. 5.30 and 5.31. The dotted lines represent the difference IR spectra generated by pure thermal treatment at 160 °C in total darkness. All investigated species showed similar changes at 160 °C in total darkness represented here by spruce (Fig. 5.30) and poplar (Fig. 5.31). These species were chosen for comparison because of their low extractive content. The absorption of unconjugated carbonyls increased around 1728 cm−1 (5787 nm) for spruce and around 1749 cm−1 (5718 nm) for poplar during pure thermal treatment. The absorption of conjugated carbonyls decreased around 1640 cm−1 (6098 nm). There is a small absorption increase at 1510 cm−1 representing the improvement of the number of lignin-type molecules. The thermal treatment generated the greatest absorption increase at 1081 cm−1 (9251 nm). This absorption increase is probably the same as was found by Popescu et al. (2013) at 1058 cm−1 (9452 nm) and attributed to C–O stretching vibrations in cellulose and hemicelluloses. The reason of the wavenumber difference might be that 1058 cm−1 was determined by Popescu et al. (2013) using the complex IR spectrum of wood. The complex IR spectrum, however, contains highly overlapping bands where the apparent peak is usually not the real one. In our case, the location of the peak (1081 cm−1 ) was determined using the difference spectrum, where the band stands alone (Figs. 5.30 and 5.31, Therm 160 °C, dotted line). The results suggest that the band having maximum at 1066 cm−1 (9381 nm) is a complex band generated by UV irradiation at 160°C. It consists of the band at 1081 cm−1 generated only by the thermal treatment, and another band on the right side. The band width is also large enough to include two sub bands. The irradiation at 30 °C is thought to be caused by the UV irradiation alone because the thermal treatment at 30 °C is rather slow. Practically, no thermal degradation happens during a 16-h treatment. Our findings strengthened this hypothesis since only noise was measured after the 16-h pure thermal treatment at 30 °C. Therefore, the UV irradiation at 30 °C is considered as pure UV treatment. Examining the changes in the unconjugated carbonyl region for spruce (Fig. 5.30), it is obvious that the simultaneous thermal and UV treatment generated much greater absorption increase than the sum of the absorption increases generated by the individual thermal treatment and UV irradiation (Therm 160 °C + UV 30 °C). These results show that the thermal effect during photodegradation is not only the simple addition of two effects, but the elevated temperature multiplies the effect of the photodegradation. Furthermore, this multiplication effect is not the only one observed. Additional chemical changes occurred in poplar as well. An additional peak emerged around 1770 cm−1 (5650 nm) during UV radiation at 160 °C compared to the irradiation at 30 °C (Fig. 5.31). There are bands where both absorption increase and decrease occurred. Good examples are the band of conjugated carbonyls around 1640 cm−1 and the band of aromatic skeletal vibration at 1508 cm−1 (6631 nm). Here, the interpretation of changes is even more complicated. The quantitative analysis according to the Arrhenius equation will presumably give more detailed information. The Arrhenius equation is introduced in Sect. 3.2. The Arrhenius plots of the individual absorption bands show the logarithm (Ln) of individual peak intensities of difference spectrum versus reciprocal temperature (in Kelvin). If the Arrhenius

5.3 Examination of Chemical Changes Generated by Photodegradation

257

plot is a straight line, the examined change is a single rate-limited thermally activated process. In this case the temperature dependence of the change is exponential. Seven characteristic absorption bands were selected for testing the validity of the Arrhenius equation in case of photodegradation at elevated temperatures. The measured date of 16-h UV irradiation at different temperatures were used to prepare the Arrhenius plots. The results are presented here according to a previously published paper (Preklet et al. 2018) with written permission from Elsevier. Most of Arrhenius plots presented here are curved lines showing that the examined changes were induced by multiple chemical processes. In some cases, the Arrhenius plots are straight lines. These trendlines are dotted lines on the figures. Most of the nonlinear lines have one breaking point at around 100 °C. It is in good correlation with previous results regarding the Arrhenius plots of colour coordinates (Tolvaj et al. 2015). Figure 5.32 shows the Arrhenius plots of the band which has a peak in the 1135– 1139 cm−1 (8811–8779 nm) interval, for the four examined species. This absorption decrease belongs to the symmetric stretching of ether bond, to the aromatic C–H deformation and to the glucose ring vibration. Therefore, this absorption decrease is a result of the superposition of 3 different chemical changes. The Arrhenius plot is a straight line only when the chemical change is a single rate-limited thermally activated process. That is why it is not surprising that the plots are strongly curved. The lines illustrate that the three different types of absorptions have individual temperature dependence. Any of these changes might have exponential temperature dependence, however. The Arrhenius plot is not the proper method to determine the nature of the chemical change in this case. Figure 5.33 illustrates the Arrhenius plots of the band which has a peak in the 1172–1176 cm−1 (8532–8503 nm) interval for the four examined species. This absorption decrease belongs to the asymmetric stretching of ether bond in cellulose

Fig. 5.32 Arrhenius plots of the absorption peak intensity change between 1135 and 1139 cm−1 (8811–8779 nm)

258

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.33 Arrhenius plots of the absorption peak intensity change between 1172 and 1176 cm−1 (8532–8503 nm). (The thin dotted straight lines are trend lines.)

and hemicellulose. The conifers follow the Arrhenius law which is proved by the straight trend lines with high value (over 0.98) of coefficient of determination. This can be translated as the splitting of this type of ether bond is exponentially temperature dependent for conifers. On the other hand, the examined deciduous species do not follow the Arrhenius law. The reason might be that the anatomical structure of deciduous species is more complex than that of the conifers. The chemical structures are also different, mostly due to the differences in the types and quantities of hemicellulose molecules. The dissimilarities of hemicelluloses might be the reason for the differences in Arrhenius plots. The exact explanation of this phenomenon needs further chemical investigations. Figure 5.34 shows the Arrhenius plots of the absorption band of guaiacyl lignin ring breathing. The place of peaks located within the 1267–1274 cm−1 wavenumber region (7893–7862 nm) depending on the species. This was the only photodegradation induced absorption decrease, where the Arrhenius plots of all species outlined as straight lines. The coefficient of determination was fairly high for all four species. Based on the figure, it can be stated that this absorption decrease is a single ratelimited thermally activated process for all examined wood species. The temperature dependence of this change is exponential. Figure 5.35 shows the Arrhenius plots of the absorption band belonging to the aromatic skeletal vibration of guaiacyl lignin between 1507 and 1510 cm−1 (6636– 6623 nm). This absorption decrease is a typical effect of photodegradation of wood (Tolvaj and Faix 1995, Pandey and Vourinen 2008;, Huang et al. 2012; Cogulet et al. 2016, Bonifazi et al. 2016). The place of maxima is located in a narrow interval (1507–1510 cm−1 ) for the four examined species. These Arrhenius plots should be linear, because the photodegradation of guaiacyl lignin has exponential temperature dependence according to Fig. 5.34. All four Arrhenius plots are, however, curved, and almost parallel to the axis above 80 °C. This happens because the photodegradation

5.3 Examination of Chemical Changes Generated by Photodegradation

259

Fig. 5.34 Arrhenius plots of the absorption peak intensity change between 1267–1274 cm−1 (7893– 7862 nm). (The thin dotted straight lines are trend lines.)

of lignin results in absorption decrease. At the same time, the increase of the intensity for this band (generated by thermal treatment) suggests the degradation of amorphous carbohydrates (especially hemicelluloses) and increases the number of lignin type molecules (Boonstara and Tjeerdsma 2006; Esteves et al. 2008; Popescu et al. 2013; Timar et al. 2016a; Liu et al. 2017). At 160 °C a third phenomenon must be taken into consideration as well. Being above the glass transition temperature of lignin, re-organisation of lignin also can change the value of absorption. Earlier works (Forsthuber et al. 2013; Tjeerdsma et al. 1998; Chen et al. 2012; Kacikova et al.2013) reported absorption increase due to thermal treatment in the 1508 cm−1 region, as consequence of the splitting of aliphatic side chains, the cleavage of β-O-4 linkages in the lignin structure, followed by the condensation reactions. These opposite changes are almost equal above 80 °C. Although the Arrhenius plots clearly show that this absorption change is a superposition of two (or more) opposite changes, they are not suitable to determine the kinetics of IR change at the 1507–1510 cm−1 wavenumber region. Figure 5.36 shows the Arrhenius plots of the absorption band at 1595–1596 cm−1 wavenumbers (6270–6266 nm). These are the absorption regions of syringyl lignin, phenolic molecules and quinones. The plots are straight lines for spruce and Scots pine, but strongly curved lines for beech and poplar. Conifers hardly contain syringyl lignin, and their absorption increase was induced mostly by the newly generated quinones. These quinones are partly responsible for the colour change. The straight trend lines demonstrate that the temperature dependence of this absorption increase is exponential. Hardwoods contain syringyl lignin and these molecules were degraded by the light irradiation. The generation of quinones and the degradation of syringyl lignin modify the intensity of absorption in the opposite direction. That is why the Arrhenius plots of deciduous species are curved.

260

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.35 Arrhenius plots of the absorption peak intensity change between 1507 and 1510 cm−1 (6636–6623 nm)

Fig. 5.36 Arrhenius plots of the absorption peak intensity change between 1595–1596 cm−1 (6270– 6266 nm). (The thin dotted straight lines are trend lines.)

Figure 5.37 shows the Arrhenius plots of the band having its peak in the 1698– 1724 cm−1 (5889–5800 nm) interval for the four examined species. The maxima are located at 1698; 1698; 1719 and 1724 cm−1 (5889, 5889, 5817 and 5800 nm) for beech, poplar, Scots pine and spruce, respectively. These maxima of conifers and deciduous species are located relatively fare to each other. This is the absorption area of unconjugated carbonyl groups generated during the photodegradation. The Arrhenius plots are straight lines for poplar, spruce and Scots pine, but slightly curved line for beech. This behaviour of beech needs further chemical investigations. Figure 5.38 shows the Arrhenius plots of the absorption bands belonging to the unconjugated carbonyl groups of the investigated species. The place of peaks located

5.3 Examination of Chemical Changes Generated by Photodegradation

261

Fig. 5.37 Arrhenius plots of the absorption peak intensity change between 1698 and 1724 cm−1 (5889–5800 nm). (The thin dotted straight lines are trend lines.)

within the 1740–1770 cm−1 wavenumber (5747–5650 nm) interval depending on the species. This was the only absorption increase generated by the photodegradation, where the Arrhenius plots of all species were straight lines. The coefficient of determination was fairly high for all four species. Based on the figure, it can be stated that this absorption increase is a single rate-limited thermally activated process for all examined wood species. The temperature dependence of this change is exponential. The results showed that the same light irradiation generated much greater absorption change at 160 °C than at 30 °C. Simultaneous thermal and UV treatment resulted in a much larger increase in absorbance than the sum of the absorption increases

Fig. 5.38 Arrhenius plots of the absorption peak intensity change between 1740–1770 cm−1 (5747– 5650 nm). (The thin dotted straight lines are trend lines.)

262

5 Applications of IR Spectrum Measurement in Wood Research

generated by thermal treatment and UV treatment separately. Softwoods were more sensitive to the light irradiation at elevated temperatures than hardwoods. The results showed that four bands emerged around 1770, 1750, 1720 and 1690 cm−1 (5650, 5714, 5814 and 5917 nm) in the unconjugated carbonyl region as a result of the treatments. The intensity change of these sub bands strongly depends on the wood species. Instances of absorption increases were found at 1066 and 1035 cm−1 (9381 and 9662 nm) only at elevated temperatures (above 80 °C). The changes were attributed to C–O stretching vibrations in cellulose and hemicelluloses. The results indicated that the Arrhenius law was useful for the interpretation the kinetics of some photochemical reactions during the photodegradation of wood. Some IR absorption bands were found where the temperature dependence was exponential. These were the ether bond in cellulose and hemicellulose, guaiacyl lignin, quinones and unconjugated carbonyl groups absorbing at 1174; 1270, 1596 and 1698–1770 cm−1 (8518, 7874, 6266 and 5889–5650 nm), respectively. The results show that most of the absorption changes of wood generated by light irradiation at elevated temperatures are not single rate-limited thermally activated processes but complex ones. In most of the cases, at least two different (and sometimes antagonistic) types of chemical changes generate absorption change in the same wavenumber region. In such cases, the Arrhenius equation is unable to indicate whether the temperature dependence is exponential or not.

5.3.3 Effect of Air Humidity Content In general, the presence of water can accelerate the chemical processes. To test the air humidity dependence of photodegradation beech and Scots pine samples were put into a closed quartz tube. A water bath under the samples ensured 100% relative air humidity within the tube (wet condition). The temperature inside the tube was measured by thermocouple. For comparison, samples were closed into a similar dry tube (dry condition, without water bath). Mercury vapour lamp was used for UV irradiation. The data of mercury vapour lamp are described at the end of Sect. 4.3. The temperature was 32 °C for the first series and 53 °C for the second series within the wet tube during the irradiation. (The temperature within the irradiation chamber was set to 30 and 50 °C.) The irradiation was interrupted after 6, 16, 36 and 72 h for IR measurements. The results are presented here according to a previously published paper (Tolvaj et al. 2016). Figure 5.39 shows the difference spectra of Scots pine samples generated by 36h UV irradiation at 32 °C. The absorption intensity decreases for lignin at 1509 and 1277 cm−1 (6627 and 7831 nm) were much greater in wet condition than in dry condition. These results show clearly that the presence of water steam intensified the degradation of lignin. Other study also demonstrated that the presence of water significantly enhanced the degradation of lignin during accelerated weathering (Kanbayashi et al. 2018). This finding is in good correspondence with the colour change data (Fig. 4.30). A broad band was grown up in the unconjugated

5.3 Examination of Chemical Changes Generated by Photodegradation

263

Fig. 5.39 Difference spectra of Scots pine wood samples after 36 h UV irradiation at 32 °C

carbonyl region between 1800 and 1670 cm−1 (5555 and 5988 nm) as a consequence of lignin degradation. The wet condition resulted in a greater absorption increase in the unconjugated carbonyl region than the dry condition confirming the catalytic effect of water molecules. The spectrum measured in dry condition shows that the broad band consists of two sub-bands and the left-side band is higher than the right-side one. The wet condition generated close to equal intensities for both sub-bands. The absorption intensity of C–O–C stretching in cellulose and hemicelluloses decreased at 1171 cm−1 (8540 nm) and the presence of water generated greater change compared to the effect of dry condition. Figure 5.40 shows the difference spectra of Scots pine samples generated by 72h UV irradiation at 53 °C. The absorption intensity decreases for guaiacyl lignin at 1509 and 1277 cm−1 (6627 and 7831 nm) were much greater in wet condition than in dry condition. These results show clearly that the presence of water steam increased the degradation of lignin. Two positive absorption peaks were grown up in the unconjugated carbonyl region with maxima at 1728 and 1750 cm−1 (5787 and 5714 nm). The wet condition produced greater absorption increase in the unconjugated carbonyl region than the dry condition confirming the catalytic effect of water molecules. It is interesting to mention that the intensity ratio of the two unconjugated carbonyl bands were the opposite in wet condition compared to dry condition. Usually, the left-side band grows faster during photodegradation, as it is presented by dry condition. Probably the same situation happened in wet condition as well, but the steam can remove some of the degradation products of lignin. This kind of leaching was more intensive for the left-side band than for the right-side band. The leaching effect was close to equal for the two unconjugated carbonyl bands after 36-h irradiation at 53 °C. (Detailed discussion of the leaching effect of water can be found in Sect. 5.3.4.)

264

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.40 Difference spectra of Scots pine wood samples after 72 h UV irradiation at 53 °C

It is interesting to compare the absorption increases in the unconjugated carbonyl region presented in Figs. 5.39 and 5.40. Shorter irradiation at lower temperature resulted in less separated sub-bands compared to the longer irradiation at higher temperature. The comparison suggests that there was leaching effect at 32 °C as well. Under dry condition, the left-side band is higher in both cases. This should have been the situation also under wet condition, but the leaching reduced the intensity of the left-side band according to the temperature. The higher temperature resulted in greater reduction than the lower temperature. This phenomenon is much more obvious in the case of beech samples. The unconjugated carbonyl sub-bands of beech are much more separated than those of Scots pine. Figure 5.41 shows the difference spectra of beech samples generated by 36-h UV irradiation at 32 °C. The wet condition resulted in slightly greater lignin degradation than the dry condition represented by the absorption decrease of syringyl lignin at 1597 cm−1 (6262 nm) and guaiacyl lignin at 1510 cm−1 (6623 nm). The degradation products of lignin underwent rapid oxidation creating carbonyl groups. These carbonyl groups absorb the IR radiation between 1680 and 1820 cm−1 (5952 and 5495 nm). According to the lignin degradation, the wet condition should have been resulted in greater absorption increase in the carbonyl region than the dry condition. It happened only to a small extent. The peak at 1708 cm−1 (5855 nm) was a little higher in wet condition than in dry condition. The behaviour of the other peak at 1762 cm−1 (5675 nm) was the opposite. The leaching effect of the water molecules is visible for the left-side band. The phenomenon is much more intensive at higher temperature (Fig. 5.42). Figure 5.42 shows the difference spectra of beech samples generated by 36-h UV irradiation at 53 °C. The wet condition resulted in greater lignin degradation than the dry condition represented by the absorption decrease of syringyl lignin at 1597 cm−1 (6262 nm) and guaiacyl lignin at 1510 cm−1 (6623 nm). The difference was greater

5.3 Examination of Chemical Changes Generated by Photodegradation

265

Fig. 5.41 Difference spectra of beech wood samples after 36-h UV irradiation at 32 °C

Fig. 5.42 Difference spectra of beech wood samples after 36-h UV irradiation at 53 °C

for guaiacyl lignin compared to syringyl lignin. According to the lignin degradation, the wet condition should have been resulted in greater absorption increase in the carbonyl region than the dry condition. In contrast, the intensity of absorption increase at 1762 cm−1 (5675 nm) was considerably smaller in wet condition than in dry condition, representing the leaching effect. The elevated temperature (53 °C) produced a more intensive removal than the lower one (32 °C). The intensity of absorption increase at 1762 cm−1 (5675 nm) under wet condition was half of that under dry condition. The reason is that the vapour molecules move faster at higher temperature.

266

5 Applications of IR Spectrum Measurement in Wood Research

There are both absorption increases and decreases in the 1100–1300 cm−1 wavenumber region (9090–7692 nm). A huge absorption increase is visible in the case of wet condition, which is not a real absorption increase (Tolvaj et al. 2011). The increased roughness elevated the spectrum in this region. This lifting effect reduced the intensity of the negative peak at 1176 cm−1 (8503 nm). The intensity of this negative peak should increase with elapsed irradiation time, but the band almost disappeared because of the artificial positive band around 1200 cm−1 (8333 nm). Figure 5.43 shows the difference spectra of beech samples generated by 72-h UV irradiation at 53 °C. The wet condition resulted in greater lignin degradation than the dry condition. The doubled irradiation time amplified the leaching effect in the unconjugated carbonyl region. The intensities of both absorption bands were reduced but in quite different portions. There was a small reduction at 1708 cm−1 (5855 nm), which was not visible at lower temperatures and shorter irradiation intervals. There is a more pronounced absorption difference between the effects of wet and dry conditions at 1762 cm−1 (5675 nm). The irradiation under dry condition resulted in three times greater absorption increase than under wet condition. The reduction was generated by the leaching effect of the water vapour. Experimental results show that the photodegradation under strongly humid condition can lead to a greater lignin degradation than under dry condition. The rate of lignin degradation is more temperature and irradiation time dependent under wet condition than under dry condition. The presence of high air humidity generates leaching effect in the unconjugated carbonyl absorption region. This leaching effect is strongly temperature and irradiation time dependent. The increasing temperature and treatment time amplify the leaching effect. This finding is especially important as the air humidity in tropical countries is much higher than in continental climate. Consequently, results of different outdoor weathering tests are not comparable unless the air humidity data are carefully monitored.

Fig. 5.43 Difference spectra of beech wood samples after 72-h UV irradiation at 53 °C

5.3 Examination of Chemical Changes Generated by Photodegradation

267

5.3.4 Effect of Water Leaching The two most destructive abiotic impacts for outdoor wooden applications are UV radiation coming from the Sun and the leaching effect of the rain. Photodegradation of wood is a widely investigated phenomenon. The water can leach out extractives (Kannar et al. 2018; Bejo et al. 2019) and consequently the colour of wood turns towards grey during long-term outdoor exposure (Tolvaj and Papp 1999). The effect of water leaching is a much less investigated phenomenon than the photodegradation. Some papers deal with simultaneous light exposure and water leaching during artificial weathering (Kamdem and Greiler 2002; Hansmann et al. 2006; Temiz et al. 2005, 2007; Fufa et al. 2013). In these studies, the combined effect of the photodegradation and water leaching was investigated, but the individual impact of light exposure and water leaching was not separated. The present study investigates systematically the individual alteration effect of UV light exposure and water leaching separately during artificial weathering. Significant differences have been found between the weathering properties of hardwood and softwood species; therefore, they are discussed separately, one after the other. Sun radiation and leaching effect of the rain were simulated by the following experiments. Samples were irradiated by mercury lamp, then plunged into distilled water at 22 °C (wet treatment). (The data of mercury vapour lamp are described at the end of Sect. 4.3). The temperature in the irradiation chamber was 50 °C during the treatment. In the first cycle, UV radiation time was 24 h, followed by water leaching for 24 h, while in the second cycle UV radiation time was doubled; 48-h UV radiation was followed by 24-h water leaching. The shorter UV radiation time at the beginning of the treatment was chosen because the degradation effect of UV radiation is very intensive at the beginning of the treatment. The second cycle was repeated up to 20 days UV radiation and 10 days water leaching (Table 5.2). The other series of specimens were subjected to UV radiation without water leaching (dry treatment). The initial weight of the samples was measured at the beginning of the treatments. Wet samples were dried at room temperature up to the initial weight to generate equal moisture content for all IR measurements. (Drying time was relatively short because the samples were small (30 × 10 × 5 in mm) Diffuse reflectance infrared Fourier transform spectrum of the samples was measured before and after treatments. (Details of the IR measurement can be found at the end of Sect. 5.1.) Hardwood species with different extractive content were chosen for the test, namely black locust (Robinia pseudoacacia L.), beech (Fagus sylvatica L.), aspen (Populus tremula L.) and sessile oak (Quercus petraea Liebl.). While black locust and sessile oak timbers have high but diverse extractive content, beech timber is featured with medium and aspen with low extractive content. Tangential surface of the specimens, containing only earlywood in the case of black locust, aspen and oak, was used for IR measurement. The results are presented here according to a recently published paper (Varga et al. 2020).

268

5 Applications of IR Spectrum Measurement in Wood Research

Table 5.2 The order of UV irradiation and water leaching during the treatments UV irradiation (UV) 1. cycle

1 day

2. cycle

2 days

Water leaching (w) 1 day

3. cycle

2 days

5. cycle

2 days

1 day

3UV + 2w

1 day

5UV + 3w

5UV + 2w 7UV + 3w 1 day

6. cycle

2 days

8. cycle

2 days

1 day

9UV + 5w

1 day

11UV + 6w

11UV + 5w 13UV + 6w 1 day

9. cycle 10. cycle

7UV + 4w 9UV + 4w

2 days

7. cycle

1UV + 1w 3UV + 1w

2 days

4. cycle

Result 1UV

13UV + 7w 15UV + 7w

1 day

15UV + 8w

1 day

17UV + 9w

17UV + 8w

2 days

20UV + 9w

3 days 1 day

20UV + 10w

The investigated species were chosen because of their diverse extractive content. Water-soluble extractives were in focus of our interest. Coldwater soluble extractive content of black locust ranges from 1.9 to 8.3%. Similar data for oak and beech are 3.6–5.7% and 1.9%, respectively (Wagenführ 2007). Aspen has data only for hot water-soluble extractives (2.9%) in Holzatlas. The main extractive components of black locust heartwood are the flavonoids. Flavonoids give 89% of the total extractive content. Within flavonoids, dihydrorobinetin is the main component covering 58% of total flavonoid content. The robinetin content covers 14% of total flavonoid content (Sanz et al. 2011). Sessile oak is less rich in extractives than most of other oak species. The total extractive content of sessile oak was found to be 3.4% in a recent study (Baar et al. 2020). The difference IR spectrum of wood gives information about the chemical changes generated by the applied treatment. The fingerprint area is presented in this work because the main differences were found in this region. The fingerprint region consists of the highly overlapped absorption bands of the main components of wood. The difference spectrum was calculated to present the chemical changes generated by UV irradiation and water leaching. The advantage of difference spectrum method

5.3 Examination of Chemical Changes Generated by Photodegradation

269

Fig. 5.44 Difference IR spectra of the investigated hardwood species generated by one day UV irradiation

is that only the absorption changes are visible. Figure 5.44 shows the effect of oneday UV irradiation on the investigated hardwood species. The main changes were similar for all investigated species. The absorption around 1509 cm−1 (6627 nm) decreased representing the degradation of guaiacyl lignin. The exact places of the minima were 1512, 1505, 1506 and 1512 cm−1 (6614, 6645, 6640 and 6614 nm) for black locust, beech, aspen and oak, respectively. This negative peak is usually visible together with the absorption decrease of the aromatic C–H deformation at around 1469 and 1428 cm−1 (6807 and 7003 nm) and with the absorption decrease of the guaiacyl ring breathing at around 1267 cm−1 (7893 nm). Comparing the species according to the degradation rate of guaiacyl lignin, black locust showed the smallest degradation followed by oak. The guaiacyl lignin in beech and aspen samples suffered greater degradation than in black locust and oak. This tendency remained during the whole 20-day irradiation. Absorption decrease of syringyl lignin is visible at around 1597 cm−1 (6262 nm). The exact places of the minima are at 1603, 1595, 1593 and 1597 cm−1 (6238, 6270, 6277 and 6262 nm) for black locust, beech, aspen and oak, respectively. The absorbed UV photons create free phenoxyl radicals by splitting the aromatic ring of lignin. These free radicals react with oxygen to generate carbonyl groups. As a consequence of lignin degradation, the intensity of unconjugated carbonyl band increased between 1680 and 1820 cm−1 (5952 and 5495 nm). These bands represent the absorption of C=O stretching for aliphatic carboxyl groups, unconjugated ketones and gamma lactones. The unconjugated carbonyl region consists of two visible bands at around 1705 and 1765 cm−1 wavenumbers (5865 and 5666 nm). These two bands partly overlap each other generating the shift of the maxima towards each other during the treatments.

270

5 Applications of IR Spectrum Measurement in Wood Research

Two absorption decreases are visible at around 1174 and 1137 cm−1 (8518 and 8795 nm). The first decrease belongs to the asymmetric stretching of ether bond while the second one belongs to the symmetric stretching of ether bond, the aromatic C–H deformation and to the glucose ring vibration. These absorption decreases indicate the ether splitting and the degradation of cellulose. There is a small positive peak at 1066 cm−1 (9381 nm) for beech and aspen. This positive peak is associated with the C–O and C–C stretching in cellulose and hemicelluloses. Popescu et al. (2013) found that the intensity increases of the band at 1058 cm−1 (9452 nm) indicates the formation of aliphatic alcohols during thermal treatment at 140 °C. There are differences among the investigated species regarding the absorption intensity change around 1174 cm−1 . It is an intensive negative band for black locust and oak. In contrast, this band is visible as a shoulder for beech and aspen. The other differences among the investigated species are visible in the 1590–1720 cm−1 wavenumber (6289–5814 nm) interval. The shorter wavenumber part of this interval is the absorption region of conjugated carbonyl groups, to be found mainly in the extractives. These groups are responsible for the colour of wood. The investigated species present similar moderate absorption decreases around 1645 cm−1 (6079 nm). The difference IR spectra of oak is shifted towards the negative values in the 1680– 1720 cm−1 wavenumber (5952–5814 nm) interval. This negative shift disappeared after 3-day UV irradiation. The reason of this shift is unknown but the extractives might be responsible for this phenomenon. Figure 5.45 shows the effects of the first UV treatment and water leaching for beech samples. The thin solid line shows the chemical changes after 1-day UV treatment (1UV). The thick solid line represents the effect of one-day water leaching after the first UV irradiation (1UV + 1w), and the dotted line presents the effect of 2-day UV irradiation after the first leaching (1UV + 1w + 2UV). The leaching hardly affected the lignin bands at 1505 and 1595 cm−1 wavenumbers (6645 and 6270 nm). (Alterations of these lignin bands will be discussed in detail later.) The 1-day water leaching reduced the absorption intensity of the unconjugated carbonyl groups at 1710 and 1763 cm−1 (5848 and 5672 nm) in different way. The leaching by water removed almost the whole band at 1763 cm−1 but affected only a little the band at 1710 cm−1 . The following 2-day UV irradiation (after the water leaching) generated a little greater absorption increase in the unconjugated carbonyl region than the one-day UV irradiation at the beginning of the treatment. A lifting effect is visible in the 1000–1200 cm−1 wavenumber (10,000–8333 nm) interval in Fig. 5.45 if we compare the three spectra. The water leaching and the further UV irradiation generated this anomaly (this effect is even more pronounced in Figs. 5.46, 5.47 and 5.48). However, these high positive intensities cannot be real absorption increases in a broad wavenumber interval. This is because the (K-M) equation does not provide the absorption spectrum properly if the surface roughness changes and the absorption is high enough. The K-M equation calculates the quotient of the absorption coefficient and the scattering coefficient. The shape of K-M function gives the absorption function properly if the light scattering remains constant during a treatment. However, photodegradation and water leaching increases the surface roughness (Tolvaj et al. 2014). The swelling lifts up the fibre edges during

5.3 Examination of Chemical Changes Generated by Photodegradation

271

Fig. 5.45 Difference IR spectra of beech (B) generated by UV irradiation and water leaching. Legends: 1UV = 1-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

water leaching, and these fibres never return to the original position during drying. This phenomenon is also responsible for the change of surface roughness. Detailed discussion of this lifting phenomenon can be found in a previous work (Tolvaj et al. 2011). The history of the negative peak at 1092 cm−1 (9158 nm) during the leaching periods shows properly the lifting effect in Figs. 5.44, 5.45, 5.46, 5.47 and 5.48. The well visible negative peak at 1092 cm−1 in Fig. 5.44 is already a positive valley between two positive bands in Figs. 5.45, 5.46, 5.47 and 5.48. The large K-M value increase overlaps the real absorption changes, preventing the evaluation of the KM spectrum in the 1000–1200 cm−1 wavenumber interval. This is the reason why the interpretation of this wavenumber region will be neglected during the further spectrum analysis. The other species showed similar changes as beech after one-day UV irradiation and water leaching. In the case of aspen, however, leaching reduced the absorption spectrum in the unconjugated carbonyl region where the originally two bands were visible as a single band with a maximum at 1723 cm−1 (5804 nm). Another difference was that the lifting effect in the 1000–1200 cm−1 wavenumber interval was negligible for black locust species. Moreover, black locust was an exception during the whole test. The lifting effect for black locust was considerably smaller in all cases compared to the other investigated species (see Fig. 5.48 as well). The reason might be that black locust is the hardest species with extremely high extractive content. The lumens in black locust wood are filled with tyloses. The presence of tyloses reduces the roughness increase generated by the photodegradation. No significant difference was found between the effects of the 5-day pure UV treatment and the 5UV + 2w combined UV radiation and leaching treatment for aspen (Fig. 5.46). Only the band with the maximum at 1761 cm−1 (5679 nm) was partly removed by the previous two-days water leaching. The 2-days leaching did not affect

272

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.46 Difference IR spectra of aspen (A) generated by UV irradiation and water leaching. Legends: 5UV = 5-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

Fig. 5.47 Difference IR spectra of oak generated by UV irradiation and water leaching. Legends: 9UV = 9-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

the lignin bands at 1506 and 1593 cm−1 (6640 and 6277 nm). The additional one-day water leaching (5UV + 3w) reduced the absorption intensity of both unconjugated carbonyl bands considerably. The other examined species showed similar changes as aspen. Only for oak appeared a small negative band at 1745 cm−1 (5731 nm). (This alteration will be discussed later.) Additional two-day UV irradiation (7UV + 3w) generated only a little absorption increase in the unconjugated carbonyl region

5.3 Examination of Chemical Changes Generated by Photodegradation

273

Fig. 5.48 Difference IR spectra of black locust (BL) generated by UV radiation and water leaching. Legends: 15UV = 15-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

for aspen specimens, compared to the A5UV + 3w spectrum. Absorption increase generated by this two-day UV irradiation was a little greater for black locust and oak than for aspen. The previously discussed lifting effect is well visible in Fig. 5.46 as well. Not only the leaching but also the 5-day pure UV irradiation produced lifting around 1100 cm−1 (9091 nm). Further treatments generated additional changes. Figure 5.47 shows the effects of 9-day UV irradiation and 5-day water leaching in the case of oak wood. The degradation of both guaiacyl and syringyl lignin was considerably greater for leached samples than for non-leached samples during the 9-day UV treatments. The comparison basis is the Oak9UV + 4w spectrum in Fig. 5.47. The next one-day water leaching (9UV + 5w) completely eliminated the absorption values in the whole unconjugated carbonyl region (1680 and 1820 cm−1 , 5952 and 5495 nm). A large absorption decrease appeared at 1745 cm−1 wavenumber (5731 nm). This negative peak appeared after 4-day leaching for beech. This negative absorption difference was reported in previous works (Csanady et al. 2015; Bejo et al. 2019). This absorption decrease did not originate from the carbonyls created by photodegradation. This negative band shows that the water leached out molecules being originally in the wood independently of the applied treatments. To clarify the reason of the absorption decrease at 1745 cm−1 , wood samples were put into distilled water to find out the effects of pure water leaching (without UV irradiation). The negative band at 1745 cm−1 appeared first after 3-day leaching for oak. Comparing the investigated species, oak was the most sensitive species to leaching in terms of the unconjugated carbonyl region. In contrast, black locust was the most stable species to water leaching. Small negative band at 1745 cm−1 wavenumber (5731 nm) appeared first after 8-day leaching

274

5 Applications of IR Spectrum Measurement in Wood Research

(Fig. 5.48). Black locust samples showed the smallest lifting effect in the 1000–1200 cm−1 wavenumber (10,000–8333 nm) interval. Moreover, the guaiacyl lignin in black locust suffered the smallest degradation during the pure UV irradiation among the investigated species. The high extractive content partly protects the lignin against the photodegradation, and the degradation products of lignin are relatively stable to water leaching. These experiences demonstrate that black locust is the most durable wood among the investigated species and therefore is suitable for outdoor applications. There is a negative shift of the spectra in the 1440–1670 cm−1 wavenumber (6944– 5988 nm) interval generated by leaching. This negative shift started with one-day leaching and increased up to the 4th day of leaching, then stopped and remained steady during the investigated time interval. The shift in the 1540–1680 cm−1 wavenumber (6494–5952 nm) interval can be interpreted by the leaching of water-soluble extractives with conjugated double bond chemical systems. These extractives are responsible for the colour of wood. The reason of the negative shift in the 1440–1540 cm−1 wavenumber (6944–6494 nm) interval is unknown. Clarification of the reason of this phenomenon needs further investigations. The negative band with maximum at 1603 cm−1 (6238 nm) is not a single absorption band of the syringyl lignin, which appears usually at 1595 cm−1 (6270 nm). It must be the superposition of two (or more) bands. The peak position of the other main band is around 1610 cm−1 (6211 nm). None of the other investigated species presented this absorption decrease at 1610 cm−1 . Therefore, it is thought that one of the extractives of black locust is sensitive to photodegradation and the UV light degrades this molecule type. Probably this extractive has aromatic ring similar to syringyl lignin, because the absorption bands are close to each other. There are differences among the investigated species regarding the necessary time for total leaching of the photodegradation generated unconjugated carbonyls. The two positive bands of unconjugated carbonyls disappeared after 5, 6 and 8 days of water leaching of oak, beech and aspen, respectively. The applied maximum 10-day leaching was not enough to diminish completely the two positive bands of unconjugated carbonyls in black locust. The band with a maximum at 1765 cm−1 (5666 nm) disappeared but the more stable band with a maximum at 1710 cm−1 (5848 nm) remained as a small positive band even after 10-day water leaching for black locust. Figures 5.44, 5.45, 5.46, 5.47 and 5.48 display the difference IR spectra in various chosen situations during the applied 20-day treatments. The following figures show the absorption changes of individual bands throughout the whole treatment time. Relative absorption changes are presented, because the differences were calculated based on the normalised spectra to provide correct comparison. Figure 5.49 shows the absorption changes of oak at 1512 cm−1 wavenumber (6614 nm). This wavenumber belongs to the vibration of the aromatic ring in guaiacyl lignin. Negative values indicate the splitting of the aromatic ring caused by UV irradiation. The first empty column on the left represents the effect of one-day UV irradiation, followed by the columns of repeated one-day water leaching’s and two-day UV irradiations. The degradation of lignin was intensive at the beginning of the treatments

5.3 Examination of Chemical Changes Generated by Photodegradation

275

Fig. 5.49 Absorption band intensity change of guaiacyl lignin in oak at 1512 cm−1 (6614 nm). Empty columns represent the effect of pure UV treatment in dry condition. Legends: xUV = x-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

and the intensity of absorption decrease hardly changed after 11 days of UV irradiation for non-leached samples. After 5-day UV irradiation, leached samples suffered considerably greater lignin degradation compared to the samples getting only UV irradiation. The reason can be that water leaching gives further possibility for the UV radiation to degrade the aromatic rings of lignin in deeper layers of the samples. Yu et al. (2022) found that; “Water spray considerably accelerated deterioration by washing away the degraded fragments, thereby exposing the fresh substrate underneath.” Similar result was found by Kanbayashi et al. (2018) for sugi wood. Beside oak, leached black locust samples suffered considerably greater lignin degradation compared to non-leached samples. These two species have high extractive content. Although, extractives partly protected the lignin against the UV degradation, leaching of extractives clearly reduced this protection. This phenomenon can be the second reason why the leached samples suffered greater lignin degradation than non-leached ones. In the case of aspen and beech, lignin degradation difference between leached and non-leached samples was small. This phenomenon can be explained by the low extractive content of aspen and beech. The absorption change around 1600 cm−1 (6250 nm) consists of the absorption region of syringyl lignin and that of conjugated carbonyl groups. Positive absorption increase is due to the generation of conjugated carbonyl groups as a result of the photodegradation of lignin and extractives. Degradation of syringyl lignin reduces the absorption in this wavenumber region. Figure 5.50 shows the absorption intensity change in the absorption place of syringyl lignin at 1595 cm−1 wavenumber (6270 nm). Positive values diminish the intensity of greater negative values during the photodegradation at 1595 cm−1 wavenumber. At the beginning of the treatments, lignin degradation was faster than the generation of conjugated carbonyl groups in

276

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.50 Absorption band intensity change in the absorption place of syringyl lignin at 1595 cm−1 (6270 nm) for aspen wood. Empty columns represent the effect of pure UV treatment in dry condition. Legends: xUV = x-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

non-leached samples (empty columns) up to the 11th day of UV irradiation. After that, the trend changed; generation of conjugated carbonyls became more intensive reducing the growing intensity of the absorption decrease at 1595 cm−1 . Water leached samples suffered considerably greater lignin degradation than nonleached samples, after 5-day UV irradiation. The reason was described in the interpretation of Fig. 5.49. Leached samples showed continuously increasing absorption decrease at 1595 cm−1 . This phenomenon presents the dominance of lignin degradation compared to the generation of conjugated carbonyl groups. The other reason of the great decrease is that the conjugated carbonyl groups were also leached out by water. The other investigated species showed similar absorption change around 1595 cm−1 as aspen. Absorption changes of unconjugated carbonyls of beech at 1713 cm−1 (5838 nm) are presented in Fig. 5.51. The absorption change of these unconjugated carbonyl groups can be evaluated only for beech samples because this band is visible as a shoulder in the case of the other investigated species. Absorption changes of nonleached samples (empty columns) increased continuously during the whole investigated time interval. Leaching reduced the absorption intensity continuously. The absorption change values turned into negative values after 7-day water leaching. Negative values were generated by the leaching of carbonyl groups being originally in wood. Figure 5.52 shows the absorption change of unconjugated carbonyl groups in black locust samples absorbing at around 1766 cm−1 (5663 nm). Pure UV irradiation generated continuous absorption increase (empty columns). Only the last

5.3 Examination of Chemical Changes Generated by Photodegradation

277

Fig. 5.51 Absorption band intensity change of the unconjugated carbonyls at 1713 cm−1 (5838 nm) for beech wood. Empty columns represent the effect of pure UV treatment in dry condition. Legends: xUV = x-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

two empty columns show slight decrease. Leaching reduced the absorption intensity continuously. The absorption change value turned into negative values generated by the tenth day of water leaching. As for the other investigated species, water leached out earlier the total band around 1760 cm−1 (5682 nm). This carbonyl band disappeared after 4-day leaching in the case of oak. The two-day UV irradiation (after water leaching) always increased the absorption considerably (dotted columns). This increase was similar to the absorption increase of non-leached samples up to the fifth day of UV irradiation. The intensity of dotted columns decreased after the fifth day of UV irradiation. This means that the photodegradation was unable to supply as many unconjugated carbonyl groups as many was leached out by the previous water treatment. The results present the diverse properties of the two investigated unconjugated carbonyl groups in case of pure UV irradiation. Time dependence of the absorption change at 1766 cm−1 (5663 nm) was similar to the time dependence of the guaiacyl lignin degradation. In contrast, the time dependence of the absorption increase around 1710 cm−1 (5848 nm) did not follow the changing tendency of the lignin band around 1510 cm−1 (6623 nm). Consequently, the absorption increase around 1710 cm−1 must be related to the degradation of hemicelluloses and/or cellulose. The origin of the absorption increase around 1710 cm−1 needs further chemical investigation. The results can be summarised as follows: Greater lignin degradation was found in the case of leached samples than in purely UV treated samples. Unconjugated carbonyl groups generated by the photodegradation proved to be the most sensitive chemical components to leaching. The photodegradation generated two absorption

278

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.52 Absorption band intensity change of the unconjugated carbonyls in black locust wood at 1766 cm−1 (5663 nm). Empty columns represent the effect of pure UV treatment in dry condition. Legends: xUV = x-day UV treatment, xUV + yw where x is the duration of UV treatment and y is the duration of water leaching treatment in days

bands of unconjugated carbonyl groups around 1710 and 1760 cm−1 wavenumbers. The band at 1760 cm−1 was much more sensitive to leaching by water than the band at 1710 cm−1 . Three to ten days of water leaching was enough to remove all unconjugated carbonyls generated by the photodegradation, depending on the species. The most sensitive species to water leaching was sessile oak. The most stable species for both photodegradation and water leaching was black locust. Individual effects of UV light irradiation and water leaching was systematically investigated and compared during accelerated weathering of conifers involving spruce and Scots pine species. The test conditions were the same as for hardwoods. The IR measurement was made on the tangential surface of the samples. The measured surface of the samples contained only earlywood. The results are presented here according to a recently published paper (Preklet et al. 2021a). Figure 5.53 shows the difference spectrum of Scots pine and spruce species created by one day and by nine-day UV irradiations. The fingerprint region is presented in this study, because this region reveals most of the chemical changes caused by photodegradation. One-day UV irradiation was enough to generate clearly visible changes. The absorption of guaiacyl lignin around 1510 cm−1 (6623 nm) decreased. This negative change was visible together with the absorption reduction of the aromatic C–H deformation at 1469 cm−1 (6807 nm) and with the absorption reduction of the guaiacyl ring breathing at 1269 cm−1 (7880 nm). The free radicals produced by the photodegradation react with oxygen and produce unconjugated carbonyl groups. Many studies have analysed the large number of carbonyl compounds formed during

5.3 Examination of Chemical Changes Generated by Photodegradation

279

Fig. 5.53 Difference IR spectra of Scots pine (P) and spruce (S) species generated by one-day and nine-day UV irradiation

the photodegradation of lignin (Lanzalunga and Bietti 2000; Chang et al. 2014; Mattonai et al. 2019). The newly generated unconjugated carbonyl groups absorb between 1650 and 1820 cm−1 (6061 and 5495 nm). This broad band is a sum of individual absorption bands. The spectrum of spruce consists of 3 absorption bands generated by oneday UV irradiation. Two maxima around 1706 and 1764 cm−1 (5862 and 5669 nm) and a shoulder around the 1690 cm−1 wavenumbers are clearly visible. The band at 1764 cm−1 belongs to the absorption of CO stretching for unconjugated ketones and γ lactones produced by the oxidation after the degradation of the aromatic ring. The aliphatic carboxyl groups absorb at 1706 cm−1 . The absorption increase of Scots pine shows only one real maximum and two shoulders after one-day UV irradiation. The prolonged irradiation time amplified the absorption changes of both the aromatic ring and the unconjugated carbonyl groups. The nine-day treatment produced one integrated absorption increase with a maximum at 1748 cm−1 (5721 nm). The shoulders are hardly visible. The visible band is the mathematical integration of the overlapping individual bands. The figure clearly illustrates that the visible band is sometimes not a real one but occurs if the individual bands highly overlap each other (see Figs. 1.26 and 1.27). Three absorption decreases can be seen at 1174, 1137 and 1096 cm−1 (8518, 8795 and 9124 nm) belonging to the ether bond stretching in cellulose and hemicellulose. The first decrease represents the asymmetric stretching of the ether bond while the second decrease shows the symmetric stretching of the ether bond together whit the aromatic C–H deformation and the glucose ring vibration. The third absorption decrease belongs to the C–O–C stretching in cellulose and hemicelluloses. The nineday irradiation generated unusual changes in the 1000–1200 cm−1 wavenumber (10,000–8333 nm) region. Absorption increases are visible for both examined

280

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.54 Difference IR spectra of earlywood (E) and latewood (L) for Scots pine (P) and spruce (S) species generated by one-day UV irradiation

conifers in this whole wavenumber region. This is not a real absorption increase in a wide wavenumber interval. The Kubelka–Munk equation does not generate the absorption spectrum correctly for intensive absorption bands if the surface roughness alters. This phenomenon lifts up the difference spectrum (Tolvaj et al. 2011). Nine-day UV irradiation produced much greater absorption decrease at 1137 cm−1 (8795 nm) than one-day treatment did. Though it is clearly visible for spruce, this negative band seems to be a positive valley between two positive bands for pine. The lifting effect created this anomaly. The current study does not discuss the changes in the 1000–1200 cm−1 (10,000–8333 nm) interval. It is important to find out the real position of sub-bands in the unconjugated carbonyl region. Figure 5.53 demonstrates that lower intensity generates greater separation between the bands. The photodegradation properties of spruce latewood can help to find the exact place of the individual maximum of the bands. Figure 5.54 presents the difference spectra of both earlywood and latewood for Scots pine and spruce species produced by one-day UV irradiation. The latewood part of spruce is less sensitive to photodegradation than earlywood, and the two bands are well separated. That is why this latewood is suitable for finding the proper place of the two maxima. These two maxima are located at the 1706 and 1764 cm−1 wavenumbers (5862 and 5669 nm). This figure assuredly demonstrates that the increasing rate of overlapping pushes the places of maxima towards each other. The intensity differences between the two bands also determine the place of the visible maximum. The shapes of the absorption bands of unconjugated carbonyls for Scots pine and spruce are mainly identical. This finding gives us the possibility to use the same two wavenumbers for both investigated conifer species as the place of maxima. Figures 5.55 and 5.56 present the effects of UV treatment and water leaching for Scots pine and spruce samples, respectively. The thin solid line presents the chemical

5.3 Examination of Chemical Changes Generated by Photodegradation

281

Fig. 5.55 Difference IR spectra of Scots pine (P). Legends: 1 UV = one-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

changes after one-day UV treatment (1UV). The thick solid line shows the effect of one-day UV irradiation and one-day water leaching (1UV + 1w), and the dotted line represents the effect of two-day UV irradiation after the first water leaching (3UV + 1w). (The intensity interval of the vertical axes is the same for Figs. 5.55, 5.56, 5.57 and 5.58 to keep the correct comparison.) The leaching did not modify the intensity decrease of the lignin band at the 1510 cm−1 wavenumber (6623 nm), but the subsequent two-day UV irradiation generated further lignin degradation for both Scots pine and spruce specimens. Detailed alteration of this lignin band will be discussed later. The pure UV irradiation produced greater absorption increase in the unconjugated carbonyl region (between 1650 and 1820 cm−1 , 6061 and 5495 nm) for Scots pine than for spruce. The presence of two absorption bands is decidedly visible for both Scots pine and spruce species. The first water leaching reduced the absorption intensity of the unconjugated carbonyl groups at 1706 and 1764 cm−1 (5862 and 5669 nm) in different way. The band at 1764 cm−1 showed greater leaching effect than the band at 1706 cm−1 . The band at 1764 cm−1 became hardly visible for spruce. Spruce samples presented greater absorption decrease than Scots pine samples in the whole unconjugated carbonyl region caused by water leaching. The two-day UV irradiation (after water leaching) created further lignin degradation and unconjugated carbonyl group production, presented by negative absorption increase at 1510 cm−1 (6623 nm) and positive absorption increase between 1650 and 1820 cm−1 (6061 and 5495 nm). Scots pine showed greater changes than spruce from both the viewpoint of lignin degradation and the viewpoint of unconjugated carbonyls generation. The negative intensity of absorption change for the aromatic ring at 1510 cm−1 (6623 nm) increased continuously during further treatment cycles. This increase was

282

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.56 Difference IR spectra of spruce (S). Legends: 1 UV = one-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

Fig. 5.57 Difference IR spectra of Scots pine (P). Legends: 11 UV = eleven-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

slow for spruce, but the increase was a little faster in the first 11 days of UV irradiation for Scots pine than for spruce. Figures 5.57 and 5.58 present the absorption changes after 11-day UV treatment for Scots pine and spruce, respectively. These figures do not clearly show the intensity differences for the negative lignin band at 1510 cm−1 (6623 nm) because the individual curves highly overlap each other. However, the figures visibly show that Scots pine suffered greater lignin degradation than spruce.

5.3 Examination of Chemical Changes Generated by Photodegradation

283

Fig. 5.58 Difference IR spectra of spruce (S). Legends: 11 UV = eleven-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

The number of newly generated unconjugated carbonyl groups was greater for Scots pine than for spruce. The maximum intensity value of the unconjugated carbonyl band was 1.01 and 0.72 after 11-day pure UV irradiation (11UV) for Scots pine and spruce, respectively. These intensity values were reduced considerably by leaching (11UV + 6 w). The bands at 1764 cm−1 (5669 nm) mostly disappeared and remained visible as shoulder after 6-day leaching. The 6-day leaching reduced the band intensity at 1706 cm−1 (5862 nm) as well. The intensity values were reduced up to 0.27 and 0.19 by 6-day leaching for Scots pine and spruce, respectively. The next 2-day UV irradiation (13UV + 6w) generated new unconjugated carbonyl groups producing absorption increase. This absorption increase was able to compensate the effect of the previous water leaching, generating the same curve as (11UV + 5w) produced. This tendency remained up to the end of treatments, only the peak intensities decreased slightly. Figures 5.55, 5.56, 5.57 and 5.58 presented selected difference IR spectra of wood species in the fingerprint region generated by the 20-day treatments. The next figures demonstrate the absorption changes of individual bands. Figure 5.59 shows the absorption changes of spruce at the 1510 cm−1 wavenumber (6623 nm). This absorption represents the vibration of the aromatic ring in guaiacyl lignin. Negative values imply the breaking of the aromatic ring generated by UV irradiation. The first empty column in Fig. 5.59 represents the effect of one-day UV irradiation, followed by the columns of repeated one-day water leaching and two-day UV irradiations. The photodegradation of guaiacyl lignin was intensive during the first day of UV irradiation. The subsequent two-day irradiations produced moderate absorption intensity decreases up to the thirteenth day for spruce. There is an abnormal intensity decrease at 7-day UV irradiation. This anomaly is visible in all similar figures (Figs. 5.55. 5.56, 5.57 and 5.58). The reason might be that the mercury lamp did not operate for a few

284

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.59 Absorption change of the unconjugated carbonyls at 1510 cm−1 for spruce. Empty columns show the effect of pure UV irradiation. Legends: xUV = x-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

hours due to a power outage. (This anomaly was ignored during the further evaluation.). Leached specimens showed considerably larger lignin degradation compared to the samples receiving pure UV irradiation. The reason of this phenomenon can be that water leaching generated further opportunity for the UV radiation to degrade the lignin in deeper layers of the specimens. Scots pine samples presented similar changes at the 1510 cm−1 wavenumber compared to spruce, only the intensities were greater. The maximum absorption intensity changes were − 0.36 for pure UV irradiated samples and − 0.45 for UV irradiated + leached samples for spruce. The same type of intensities for Scots pine were − 0.42 and − 0.52. For pure UV irradiated samples, 15 days were enough to generate the maximum lignin degradation. In contrast, leached samples produced continuously increasing lignin degradation during the 20-day UV irradiation. This result confirms that leaching opens deeper layers for photodegradation. The chosen UV irradiation and leaching times were not long enough to determine the maximum of lignin degradation. The degradation of lignin raised the number of unconjugated carbonyl groups. Time dependence of this phenomenon is presented in Figs. 5.60, 5.61 and 5.62. The time dependence of unconjugated carbonyls absorbing at 1706 cm−1 (5862 nm) is presented in Fig. 5.60 for Scots pine. The absorption increase was rapid up to the third day of pure UV treatment. This was followed by a continuous but moderate absorption increase for pure UV treated samples. The leached samples followed the changing tendency of pure UV treated samples up to the third day of leaching. After this period, the two types of changes diverged from each other. The new two-day UV irradiations always produced new unconjugated carbonyl groups, but the next water leaching reduced the number of carbonyl groups up to a constant level during the total examined period. This constant level was 0.2 unit for both investigated species.

5.3 Examination of Chemical Changes Generated by Photodegradation

285

Fig. 5.60 Absorption change of the unconjugated carbonyls at 1706 cm−1 for Scots pine. Empty columns show the effect of pure UV irradiation. Legends: xUV = x-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

The first leaching removed 18% of the carbonyls absorbing at 1706 cm−1 and the last (tenth) leaching also reduced by 29%. The maximum reduction was 44% during the third leaching period. Spruce samples showed similar properties to those of Scots pine. The leaching data of spruce were: first leaching 49% (it was the maximum as well), and last leaching, 38%. These data show that spruce species were more sensitive to leaching than Scots pine. Figures 5.61 and 5.62 show the absorption changes of unconjugated carbonyl groups absorbing at 1764 cm−1 wavenumber (5669 nm) for Scots pine and spruce, respectively. Changing tendencies of absorption intensities were similar for both investigated species. Differences were found in the intensity values. Samples receiving only UV irradiation presented a rapid absorption increase during the first five days of treatment. This intensity increase was slow up to the thirteenth day of irradiation and continued to slow down until the end of treatment. The maximum absorption intensity changes were 0.79 for pure UV irradiated samples and 0.53 for UV irradiated + leached samples for Scots pine. Spruce samples produced smaller absorption increase at 1764 cm−1 than Scots pine. The same type of intensities for spruce were 0.58 and 0.36. The leached samples followed the changing tendency of pure UV treated samples up to the third day of treatment. After this period, the two types of changes diverged. The subsequent two-day UV irradiations always produced new unconjugated carbonyl groups, but the following water leaching reduced the number of carbonyl groups considerably. This reduction produced almost constant absorption change values (slightly below 0.1 unit) for spruce (Fig. 5.62). The similar intensity change values were between 0.2 and 0.1 for Scots pine showing slow intensity decrease. The first leaching removed 54% of the unconjugated carbonyls

286

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.61 Absorption change of the unconjugated carbonyls at 1764 cm−1 for Scots pine. Empty columns show the effect of pure UV irradiation. Legends: xUV = x-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

Fig. 5.62 Absorption change of the unconjugated carbonyls at 1764 cm−1 for spruce. Empty columns show the effect of pure UV irradiation. Legends: xUV = x-day UV irradiation; xUV + yw is an abbreviation where x is the extent of UV irradiation, and y is the extent of water leaching in days

absorbing at 1764 cm−1 , and the last leaching removed 48%. The maximum reduction was 65% during the second leaching period. Spruce samples showed similar leaching properties as Scots pine, but the leaching was much more effective. The leaching data of spruce were the following: first leaching, 77%, last leaching 66%,

5.3 Examination of Chemical Changes Generated by Photodegradation

287

and the maximum was 79% during the second leaching period. These data show that spruce species were much more sensitive to leaching than Scots pine. The results present the diverse properties of the two investigated unconjugated carbonyl groups in case of pure UV irradiation. Time dependence of the absorption change at 1764 cm−1 (5669 nm) was similar to the time dependence of the guaiacyl lignin degradation. In contrast, the time dependence of the absorption increase at 1706 cm−1 (5862 nm) did not follow the changing tendency of the lignin band at 1510 cm−1 (6623 nm). Consequently, the absorption increase at 1706 cm−1 must be related to the degradation of hemicelluloses and/or cellulose. The origin of the absorption increase at 1706 cm−1 needs further chemical investigation. The results can be summarised as follows: Leached samples produced greater lignin degradation than the purely UV treated samples. This finding presents that the leaching effect of rain opens deeper layers for UV degradation raising the rate of degradation of the timber. Scots pine suffered greater lignin degradation than spruce and produced higher absorption increase in the absorption region of unconjugated carbonyls. Unconjugated carbonyl groups showed the greatest sensitivity to leaching. Spruce was more susceptible to leaching of unconjugated carbonyl groups than Scots pine. These results show that both Scots pine and spruce timbers need proper surface finishing in case of outdoor applications to prevent UV light and water leaching induced decomposition. For the samples, that were subjected only to UV irradiation, the 15-day treatment was enough to generate maximum lignin degradation. In contrast, leached samples produced continuously increasing lignin degradation during the 20-day UV irradiation. Nevertheless, the applied UV irradiation and leaching duration was not long enough to determine the maximum time interval of lignin degradation. These results highlight that the combined UV irradiation and water leaching generates degradation of wood in deeper layers than the pure UV irradiation. The photodegradation produced two absorption increases in the unconjugated carbonyl region at 1706 and 1764 cm−1 wavenumbers. The absorption band at 1764 cm−1 showed much greater leaching effect than the band at 1706 cm−1 . These results strengthen the different origin of these two types of absorption increases.

5.3.5 Photodegradation Properties of Steamed Wood. Steaming is an environmentally friendly colour modification method to create a warm brown wood colour. There are some wood species that have unattractive or highly inhomogeneous natural colour. This appearance reduces the usability of these species for visible constructions. Colour modification by steaming can help to modify the unattractive original colour or highly inhomogeneous colour to attractive brown colour. The colour stability of steamed wood is an important practical issue. This phenomenon is discussed in Sect. 4.8.

288

5 Applications of IR Spectrum Measurement in Wood Research

It is also of scientific interest to investigate chemical changes of steamed wood during UV irradiation to reveal whether steaming can improve the UV stability of the studied species. Black locust (Robinia pseudoacacia L.), beech (Fagus sylvatica L.), poplar (Populus x euramericana cv. Pannonia) and spruce (Picea abies Karst.) specimens were used for the tests. Black locust and beech were chosen because these species are the most frequently steamed species in wood industry so that their behaviour during UV irradiation could be of high importance from commercial point of view. Plantation poplar grows in large quantity and needs new application areas, whereas spruce was added for comparison of deciduous and conifer species. Initial moisture content of the specimens was between 9 and 10% before steaming. Wood specimens were placed in the autoclave with distilled water at the bottom for conditioning the air to maintain maximum (100%) relative humidity. The chosen steaming temperatures were 100, 110, and 120 °C, the steaming time was 2 days. Steam-treated specimens were subjected to photodegradation together with the thermally untreated control specimens. Mercury vapour lamp was used as strong UV light emitter. (The data of mercury vapour lamp are described at the end of Sect. 4.3). The light power density was 76 J/m2 s on the surface of the specimens. The irradiation chamber was set to 60 °C guaranteeing ambient temperature conditions. The total irradiation time was 36 h. The irradiation was interrupted after 7, 16 and 36 h to measure the IR spectra. The radial surface of the samples was used for IR measurement. (Details of IR measurement can be found at the end of Sect. 5.1.) (Wavelength values are not indicated in the figures because too many numbers together may be confusing.) The results are presented here according to a recently published paper (Hofmann et al. 2022). The extractive content of the species plays an important role in colour and in colour change of the wood species. Most of the extractives are sensitive to thermal treatment, thus giving the possibility to modify the wood colour by steaming, without any chemicals. Moreover, it is important to discover, whether steaming can modify the photodegradation properties of timbers. It is worth to present first the photodegradation properties of the unsteamed species for basic comparison (Figs. 5.63 and 5.64). Figure 5.63 shows the absorption spectra of the untreated specimens. The spectra of the four species are very similar, showing peaks at the same wavenumbers. From the high number of peaks only those are assigned, which will be discussed in the present evaluation. Several overlapping absorption bands can be found between the 1100–1500 cm−1 wavenumber (9090– 6667 nm) range. The measured place of maxima of overlapped bands are usually not the real ones, small shifts are possible. (The difference spectrum gives more precise peak positions, because the changed bands are visible only.) The only wellseparated band between 1700 and 1800 cm−1 (5882 and 5556 nm) belongs to the absorption of unconjugated carbonyl groups. This is also a composed band, including the absorption bands of the carbonyl groups, found in different molecules at various positions. The next band at 1597 cm−1 belongs to the aromatic ring vibrations of syringyl lignin. This band is overlapped by the broad band of conjugated carbonyl groups between 1570 and 1700 cm−1 (6369 and 5882 nm) and the bound water also

5.3 Examination of Chemical Changes Generated by Photodegradation

289

Fig. 5.63 IR spectra of untreated black locust (BL), beech (B), poplar (P) and spruce (S) samples

Fig. 5.64 Difference IR spectra of unsteamed black locust (BL), beech (B), poplar (P) and spruce (S) specimens generated by 36-h UV irradiation

absorbs in this wavenumber range. Conjugated carbonyl groups are found mainly in extractives. Wood can contain a great variety of extractives in various amounts, depending on wood species (deciduous, coniferous) and tissue type (sapwood, heartwood), thus absorption changes in this region only gives information on the change of their amount but do not provide information on the type of the extractives. The typical lignin band is visible at 1506 cm−1 (6640 nm). The next band at 1463 cm−1 (6835 nm) is the superposition of the absorption for C–H deformation in lignin and C–H deformation in xylan. The peak at 1378 cm−1 (7257 nm) belongs to the C–H deformation in cellulose. This band remains stable during photodegradation and is used for spectra normalisation and comparison. The peak at 1276 cm−1 (7837 nm)

290

5 Applications of IR Spectrum Measurement in Wood Research

represents the absorption of Caryl –O, guaiacyl ring breathing with CO stretching. The two highest peaks represent the absorption of C–O–C stretching, both asymmetric and symmetric for cellulose and hemicelluloses at 1175 and 1133 cm−1 (8511 and 8826 nm), respectively. Figure 5.64 presents the difference IR spectra of the investigated unsteamed species generated by 36-h UV irradiation. The greatest absorption decrease is visible at 1173 and 1139 cm−1 (8525 and 8780 nm). The first decrease belongs to the asymmetric stretching of the ether bond in cellulose. The second decrease belongs to the symmetric stretching of the ether bond, the aromatic C–H deformation and to the glucose ring vibration. These absorption decreases indicate the ether bond splitting. Hemicelluloses have ether linkages in different positions. The determination of the exact positions of ether bond splitting needs further investigations using near infrared spectroscopy. The negative band at 1173 cm−1 for spruce is not visible. The positive band with maximum at 1186 cm−1 (8432 nm) overlaps and eliminates this negative band. The negative peak around 1510 cm−1 (6623 nm) belongs to the aromatic skeletal vibration of guaiacyl lignin. The exact places of this peak are 1512, 1510, 1509 and 1508 cm−1 (6614, 6623, 6627 and 6631 nm) for black locust, spruce, beech and poplar, respectively. This negative peak is detectable together with the absorption decrease of the guaiacyl ring breathing around 1270 cm−1 (7874 nm). Absorption decrease of syringyl lignin is visible around 1598 cm−1 (6258 nm). The exact places of the minima are 1621, 1598 and 1596 cm−1 (6169, 6258 and 6266 nm) for black locust, beech, and poplar, respectively. The peak maximum, observed for black locust is not the real one. The neighbouring negative band overlaps this lignin band and the superposition modifies the place of the maximum considerably. As softwoods contain insignificant amounts of syringyl lignin, the corresponding band is missing completely from the IR spectrum of spruce. Deciduous species presented an absorption decrease around 1650 cm−1 (6061 nm) wavenumber. This region represents the absorption of conjugated carbonyl groups located mainly in extractives. As a consequence of lignin degradation, the generated free radicals react with oxygen producing carbonyl groups and, in the end, a great variety of carbonyl compounds. The absorption increase of unconjugated carbonyls is visible in the 1680–1820 cm−1 wavenumber (5952–5493 nm) range. For deciduous species the increase of two additional bands was observed at 1705 and at 1764 cm−1 wavenumbers (5865 and 5669 nm). The band at 1764 cm−1 represents the absorption of CO stretching for unconjugated ketones and γ lactones generated by the oxidation after the splitting of the aromatic ring. The band at 1705 cm−1 represents the absorption of aliphatic carboxyl groups. These two bands were integrated into a broad band for spruce with a maximum at 1730 cm−1 (5780 nm). It seems that the absorption increase at 1764 cm−1 was small and the absorption increase at 1705 cm−1 was dominant for spruce. Deciduous species showed opposite changes. The band at 1764 cm−1 was dominant for deciduous species comparing to the band at 1705 cm−1 . Deacetylation of hemicelluloses during steaming is caused by the cleavage of acetyl groups linked as an ester group to the hemicelluloses (Carrasco and Roy 1992; Tjeerdsma and Militz 2005; Liu et al. 2016). Typical smell of acetic acid

5.3 Examination of Chemical Changes Generated by Photodegradation

291

Fig. 5.65 Difference IR spectra of black locust and spruce specimens generated by two-day steaming at 100 °C

comes out from the steaming chamber, confirming deacetylation. Figure 5.65 represents the difference IR spectra of black locust and spruce generated by two-day steaming at 100 °C. The IR spectra clearly show that deacetylation of hemicelluloses is the main chemical change during steaming. The difference IR absorption spectra of black locust and spruce show great differences. The reason lies in the different hemicellulose content of black locust and spruce. The highly acetylated xylan is the main polyose component in hardwoods, however, glucose units in spruce are rarely acetylated. The absorption of carbonyl groups being in the acetyl unit decreased at 1751 cm−1 (5711 nm) parallel with the absorption decrease of the ester bridge at 1174 cm−1 (8518 nm). These two main changes were accompanied by the absorption decreases at 1470 cm−1 (6803 nm) (C–H deformation in xylan). All these absorption decreases clearly present the deacetylation effect of steaming. Spruce specimens showed only small absorption decrease at the above-mentioned wavenumbers showing the low deacetylation rate of hemicelluloses. There is a wide negative absorption band around 1678 cm−1 (5959 nm) representing the degradation of extractives having conjugated carbonyl groups. A broad positive peak is visible between 950 and 1250 cm−1 (10,526 and 8000 nm), however, this peak does not represent real absorption increase. The Kubelka–Munk equation does not provide the absorption spectrum properly for intensive absorption bands if the surface roughness changes. This phenomenon elevates the difference spectrum intensities in this region. A detailed discussion of this phenomenon can be found in a previous work of the author (Tolvaj et al. 2011). Photodegradation properties of steamed specimens are presented in Figs. 5.66, 5.67, 5.68, 5.69 and 5.70. Figure 5.66 shows the irradiation time dependence of the chemical changes in poplar specimens steamed at 110 °C. Although the lignin content decreased (both syringyl and guaiacyl) during UV irradiation, the absorption

292

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.66 Difference IR spectra of steamed (at 110 °C) poplar specimens generated by 7, 16 and 36-h UV irradiation

of guaiacyl ring breathing at 1270 cm−1 (7874 nm) did not decrease after 7-h irradiation. The reason might be that an unidentified band was grown up with increasing treatment time at the left side eliminating the decrease in absorption at 1270 cm−1 . The absorption values of unconjugated carbonyl groups increased around 1710 and 1760 cm−1 wavenumbers (5848 and 5682 nm) by prolonged irradiation time. In general, the values of absorption changes were proportional to the UV irradiation time. The only exception was the absorption decrease of the ether bonds at 1171 cm−1 (8525 nm). The absorption intensity decrease of this band hardly changed after this 7-h period. The reason for this might be that the previous steaming generated great absorption decrease at 1171 cm−1 (see Fig. 5.65). Time dependence of the chemical changes was similar in the case of the other investigated deciduous species as well. Figure 5.67 shows the difference IR spectra of unsteamed and steamed black locust specimens generated by 36-h UV irradiation. Steaming reduced the degradation rate of guaiacyl lignin absorbing at 1512 cm−1 (6614 nm). Steaming at 120 °C produced a slightly better lignin protection compared to the steaming at 100 °C. The photodegradation induced yellowing level of steamed black locust also indicated the protective effect of steaming (Varga et al. 2021). The spectra do not clearly show the photodegradation stability of syringyl lignin absorbing around 1600 cm−1 wavenumber (6250 nm). The conjugated carbonyl groups have absorption here as well, and the neighbouring bands overlap each other. Steaming reduced the rate of ether bond splitting (absorbing at 1174 cm−1 (8518 nm)) compared to the unsteamed specimens. The reason might be that steaming produced ether bond splitting during this pre-treatment as well. The alteration of steaming temperature does not affect the intensity of this absorption band. Two bands developed in the unconjugated carbonyl region. The band of CO stretching for unconjugated ketones and γ lactones is located at 1764 cm−1 (5669 nm). Pre-steaming did not modify the intensity of this band.

5.3 Examination of Chemical Changes Generated by Photodegradation

293

Fig. 5.67 Difference IR spectra of unsteamed and steamed (St) black locust specimens generated by 36-h UV irradiation

Fig. 5.68 Difference IR spectra of unsteamed and steamed (St) beech specimens generated by 36-h UV irradiation

Unsteamed specimens suffered similar absorption changes to steamed specimens. Only the place of the maximum shifted somewhat towards shorter wavenumbers for the steamed samples. This maximum shift was produced by the growth of the adjacent overlapping band around 1705 cm−1 (5865 nm). The intensity change of the carbonyl band at 1705 cm−1 (5865 nm) was modified by the steam pre-treatment. Steamed specimens showed greater absorption increase than the unsteamed one. Previous studies demonstrated that the absorption increase at 1705 cm−1 does not originate from the lignin degradation (Preklet et al. 2021a, b). Research papers demonstrated that photo-oxidation of cellulose and hemicelluloses

294

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.69 Difference IR spectra of unsteamed and steamed (St) poplar specimens generated by 36-h UV irradiation

Fig. 5.70 Difference IR spectra of unsteamed and steamed (St) spruce specimens generated by 36-h UV irradiation

results in the formation of aldehyde and ketone groups on carbon atoms C2 and C3 of the pyran or furan units, and just these carbonyls may contribute to the increase in absorbance in the unconjugated carbonyl region around 1730 cm−1 85,780 nm) (Müller et al. 2003; Reinprecht et al. 2018). These papers studied the integrated carbonyl band having maximum at 1730 cm−1 . Our results were determined by difference spectra. Figure 5.65 shows that deacetylation of glucose rings was the only difference between the steamed and unsteamed black locust specimens. Consequently, the deacetylated glucose units must be more susceptible to photodegradation than the others. The reason that steamed specimens showed greater absorption

5.3 Examination of Chemical Changes Generated by Photodegradation

295

increase than the unsteamed ones at 1705 cm−1 might be that deacetylated hemicelluloses underwent photo-oxidation during the UV treatment. This statement needs further chemical investigations. All of the investigated deciduous species showed this property. This finding supports that deacetylated hemicelluloses play an important role in the photodegradation generated absorption increase at 1705 cm−1 . Figures 5.68 and 5.69 show the difference IR spectra of unsteamed and steamed beech and poplar specimens, respectively. The changes were generated by 36-h UV irradiation. The IR absorption changes of beech and poplar were similar. Guaiacyl lignin absorbing at 1509 cm−1 (6627 nm) produced similar degradation in case of both unsteamed and steamed specimens. It means that steaming was unable to reduce the photodegradation sensitivity of lignin in both beech and poplar. Syringyl lignin absorbing at 1598 cm−1 (6258 nm) showed the same degradation properties as guaiacyl lignin. Unsteamed samples produced somewhat greater absorption decrease than steamed ones. The deviation was generated by the neighbouring overlapping band at 1650 cm−1 (6061 nm). Steaming reduced the rate of ether bond splitting absorbing at 1174 cm−1 (8518 nm) compared to the unsteamed specimens, because the applied heat treatment already induced ether bond splitting. The photo-oxidation of deacetylated hemicelluloses enlarged the absorption in the unconjugated carbonyl region at 1705 cm−1 (5865 nm) for beech and poplar specimens similar to black locust. Figure 5.70 shows the difference IR spectra of unsteamed and steamed spruce specimens generated by 36-h UV irradiation. Curves of unsteamed and steamed specimens are almost the same. This phenomenon shows that steaming does not modify the photodegradation properties of spruce timber. Figure 5.71 shows the absorption decrease of guaiacyl lignin of unsteamed and steamed species generated by 36-h UV irradiation. Degradation of lignin macromolecules is the main chemical alteration during the photodegradation of wood. The data of unsteamed species (empty columns) reveal that lignin was most stable in black locust compared to the other investigated species. A previous study demonstrated that extractives play an essential role in the photodegradation of wood, and the rate of wood degradation was lowered by the presence of extractives (Chang et al. 2010). The extractive content of black locust heartwood can reach 9%. Flavonoids make up 89% of the total extractive content. Within flavonoids, dihydrorobinetin is the main component covering 58% of total flavonoid content, followed by robinetin with 14% (Sanz et al. 2011). The UV light degrades the extractives followed by rapid oxidation of the degradation products. The modified chromophores act as a form of energy trap which slows down the photodegradation of lignin (Nemeth et al. 1992; Anish et al. 2023). Steaming widely modifies the extractives. This statement is supported by the intensive colour change during steaming. The modification rate increases with elevated temperature according to the Arrhenius law. Steam induced colour change demonstrates this modification of extractives. The modified extractives provide even higher protection for lignin. Steaming of black locust at 120 °C doubled (48%) the protection of lignin compared to the unsteamed specimens. As for the other investigated wood species, only poplar showed a 15% decrease in lignin degradation among the steamed at 120 °C and the untreated specimen. Beech and spruce species did not

296

5 Applications of IR Spectrum Measurement in Wood Research

Fig. 5.71 Absorption decrease of guaiacyl lignin of unsteamed and steamed (St) species generated by 36-h UV irradiation

show the protective effect of the extractives for the steamed specimens. In the case of black locust, the main extractive is dihydrorobinetin, which might play the main role in protecting lignin during UV irradiation. This phenomenon needs further chemical investigation. Experimental results show that steaming was able to reduce considerably the rate of UV degradation of lignin only for black locust. In line with this finding, a previous study (Varga et al. 2021) demonstrated that steaming was able to increase the colour stability against UV irradiation only for black locust among the investigated species. Dzurenda and Dudiak (2022) published that steaming increased the colour stability of alder wood samples during UV irradiation. Steamed hardwood specimens presented greater absorption increase at 1705 cm−1 (5865 nm) than unsteamed ones. This increase refers to the photodegradation of hemicelluloses. The current findings demonstrate that in the case of outdoor wooden applications the steam generated attractive brown colour and the lignin in the timber requires effective protection using a proper finishing layer.

References Agresti, G., Bonifazi, G., Calienno, L., Capobianco, G., Lo Monaco, A., Pelosi, C., Picchio, R., Serranti, S.: Surface investigation of photo-degraded wood by colour monitoring, infrared spectroscopy, and hyperspectral imaging. J. Spectrosc. 1(1), ID 380536 (2013). https://doi.org/10. 1155/2013/380536 Anish, MC., Giridher, BN., Nair, S., Anantha, NS., Pandey, KK.: Influences of extractives and thermal modification on the UV resistance of Albizia lebbeck wood. Wood Mat. Sci. Eng. 18(2), 540-548 (2023). https://doi.org/10.1080/17480272.2022.2053586

References

297

Baar, J., Paschová, Z., Hofmann, T., Koláˇr, T., Koch, G., Saake, S., Rademacher, P.: Natural durability of subfossil oak: wood chemical composition changes through the ages. Holzforschung 74, 47–59 (2020). https://doi.org/10.1515/hf-2018-0309 Bejo, L., Tolvaj, L., Kannar, A., Preklet, E.: Effect of water leaching on photodegraded spruce wood monitored by IR spectroscopy. J. Photochem. Photobiol. A: Chem. 382, ID 111948 (2019). https://doi.org/10.1016/j.jphotochem.2019.111948 Bonifazi, G., Calienno, L., Capobianco, G., Lo Monaco, A., Pelosi, C., Picchio, R., Serranti, S.: A new approach for the modelling of chestnut wood photo-degradation monitored by different spectroscopic techniques. Environ. Sci. Pollut. Res. 24, 13874–13884 (2016). https://doi.org/ 10.1007/s11356-016-6047-0 Boonstara, M.J., Tjeerdsma, B.: Chemical analysis of heat treated softwoods. Holz Roh Werkstoff 64, 204–211 (2006). https://doi.org/10.1007/s00107-005-0078-4 Bourgois, P.J., Guyonnet, R.: Characterization and analysis of torrified wood. Wood Sci. Technol. 22, 143–155 (1988). https://doi.org/10.1007/BF00355850 Carrasco, F., Roy, C.: Kinetic study of dilute-acid prehydrolysis of xylan-containing biomass. Wood Sci. Technol. 26, 189–208 (1992). https://doi.org/10.1007/BF00224292 Chang, T.C., Chang, H.T., Wu, C.L., Lin, H.Y., Chang, S.T.: Stabilizing effect of extractives on the photooxidation of Acacia confusa wood. Polym. Degrad. Stab. 95, 1518–1522 (2010). https:// doi.org/10.1016/j.polymdegradstab.2010.06.012 Chang, T.C., Lin, H.Y., Hongchang, L., Chang, S.T.: Study on inhibition mechanisms of lightinduced wood radicals by Acacia confusa heartwood extracts. Polym. Degrad. Stab. 105, 42–47 (2014). https://doi.org/10.1016/j.polymdegradstab.2014.03.036 Chen, Y., Fan, Y., Gao, J., Stark, M.S.: The effect of heat treatment on the chemical and color change of black locust (Robinia pseudoacacia) wood flour. BioResources 7(1), 1157–1170 (2012) Cirule, D., Kuka, E., Andersone, I., Andersons, B.: Wood discoloration patterns depending on the light source. Heritage Sci. 10, ID 158 (2022). https://doi.org/10.1186/s40494-022-00795-2 Cogulet, A., Blanchet, P., Landry, V.: Wood degradation under UV irradiation: a lignin characterization. J. Photochem. Photobiol. B: Biol. 158, 184–191 (2016). https://doi.org/10.1016/j.jphoto biol.2016.02.030 Colom, X., Carrillo, F., Nogues, F., Garriga, P.: Structural analysis of photodegraded wood by means of FTIR spectroscopy. Polym. Degrad. Stab. 80, 543–549 (2003). https://doi.org/10.1016/S01413910(03)00051-X Csanady, E., Magoss, E., Tolvaj, L.: Quality of Machined Wood Surfaces. Springer, Heidlberg New York London (2015). https://doi.org/10.1007/978-3-319-22419-0 Dietrichs, H.H., Sinner, H., Puls, J.: Potential of steaming hardwoods and straw for feed and food production. Holzforschung 32, 193–199 (1978) Dzurenda, L., Dudiak, M.: Colour change of steamed Alder Wood caused by UV radiation. Drvna Industrija 73(3), 289–297 (2022). https://doi.org/10.5552/drvind.2022.0019 Esteves, B.M., Gracia, J., Pereira, H.M.: Extractive composition and summative chemical analysis of thermally treated eucalypt wood. Holzforschung 62, 344–351 (2008). https://doi.org/10.1515/ HF.2008.057 Esteves, B., Marques, A.V., Domingos, I., Pereira, H.: Chemical changes of heat treated pine and eucalypt wood monitored by FTIR. Maderas, Sciencia Technol. 15(2), 245–258 (2013). https:// doi.org/10.4067/S0718-221X2013005000020 Fabiyi, J.S., Ogunleye, B.M.: Mid-infrared spectroscopy and dynamic mechanical analysis of heat treated obeche (Triplochiton scleroxylon) wood. Maderas, Ciencia Technol. 17(1), 5–16 (2015). https://doi.org/10.4067/S0718-221X2015005000001 Forsthuber, B., Müller, U., Teischinger, A., Grüll, G.: Chemical and mechanical changes during photooxidation of an acrylic clear wood coat and its prevention using UV absorber and micronized TiO2 . Polym. Degrad. Stab. 98, 1329–1338 (2013). https://doi.org/10.1016/j.pol ymdegradstab.2013.03.029 Forsthuber, B., Illy, A., Grüll, G.: Photo-scanning colorimetry of wood and transparent wood coatings. Eur. J. Wood Prod. 72, 487–495 (2014). https://doi.org/10.1007/s00107-014-0808-6

298

5 Applications of IR Spectrum Measurement in Wood Research

Fufa, S.M., Jelle, B.P., Hovde, P.J.: Weathering performance of spruce coated with water based acrylicpaint modified with TiO2 and clay nanoparticles. Progr. Org. Coat. 76, 1543–1548 (2013). https://doi.org/10.1016/j.porgcoat.2013.06.008 George, B., Suttie, E., Merlin, A., Deglise, X.: Photodegradation and photostabilisation of wood. Polym. Degrad. Stab. 88, 268–274 (2005). https://doi.org/10.1016/j.polymdegradstab.2004. 10.018 Hansmann, C., Deka, M., Wimmer, R., Gindl, W.: Artificial weathering of wood surfaces modified by melamine formaldehyde resin. Holz Roh Werkstoff 64, 198–203 (2006). https://doi.org/10. 1007/s00107-005-0047-y Hill, C.A.S.: Wood Modification: Chemical, Thermal and Other Processes. Wiley, London (2006) Hofmann, T., Tolvaj, L., Visi-Rajczi, E., Varga, D.: Chemical changes of steamed timber during short-term photodegradation monitored by FTIR spectroscopy. Eur. J. Wood Prod. 80, 841–849 (2022). https://doi.org/10.1007/s00107-022-01814-6 Hon, D.N.S.: Weathering and photochemistry of wood. In: Hon, D.N.S., Shiraishi, N. (eds.) Wood and Cellulose Chemistry. Marcel Dekker, New York (2001) Huang, A., Zhou, Q., Liu, J., Fei, B., Sun, S.: Distinction of three wood species by Fourier transform infrared spectroscopy and two-dimensional correlation IR spectroscopy. J. Mol. Struct. 883–884, 160–166 (2008). https://doi.org/10.1016/j.molstruc.2007.11.061 Huang, X.: Study on the degradation mechanism of heat-treated wood by UV light. Ph.D. thesis, Universitá du Québec a Chicoutimi, Canada (2012) Huang, X., Kocaefea, D., Kocaefea, Y., Boluk, Y., Pichette, A.: Study of the degradation behavior of heat-treated jack pine (Pinus banksiana) under artificial sunlight irradiation. Polym. Degrad. Stab. 97, 1197–1214 (2012). https://doi.org/10.1016/j.polymdegradstab.2012.03.022 Kacik, F., Kubovsky, I., Boucek, J., Hrcka, R., Gaff, M., Kacikova, D.: Colour and chemical changes of black locust wood during heat treatment. Forests 14(1), ID 73 (2023). https://doi.org/10.3390/ f14010073 Kacikova, D., Kacik, F., Cabalová, I., Durkovic, J.: Effects of thermal treatment on chemical, mechanical and colour traits in Norway spruce wood. Biores. Technol. 144, 669–674 (2013). https://doi.org/10.1016/j.biortech.2013.06.110 Kamdem, P., Greiler, S.: Surface roughness and color change of cooper amine and UV absorbertreated red maple (Acer rubrum) exposed to artificial ultraviolet light. Holzforschung 56, 473– 478 (2002). https://doi.org/10.1515/HF.2002.0736 Kanbayashi, T., Kataoka, Y., Ishikawa, A., Matsunaga, M., Kobayashi, M., Kiguchi, M.: Confocal Raman microscopy reveals changes in chemical composition of wood surfaces exposed to artificial weathering. J. Photochem. Photobiol. B: Biol. 187, 136–140 (2018). https://doi.org/10. 1016/j.jphotobiol.2018.08.016 Kanbayashi, T., Matsunaga, M., Kobayashi, M.: Effects of natural weathering on the chemical composition of cell walls in sapwood and heartwood of Japanese cedar. Wood Sci. Technol. 55, 1013–1024 (2021). https://doi.org/10.1007/s00226-021-01301-w Kannar, A., Tolvaj, L., Magoss, E.: Colour change of photodegraded spruce wood by water leaching. Wood Res. 63(6), 935–946 (2018) Kataoka, Y., Kiguchi, M., Fujiwara, T., Evans, P.D.: The effect of within-species and betweenspecies variation in wood density on the photodegradation depth profiles of sugi (Cryptomeria japonica) and hinoki (Chamaecyparis obtusa). J. Wood Sci. 51, 531–536 (2005). https://doi. org/10.1007/s10086-004-0685-4 Kataoka, Y., Kiguchi, M., Williams, R.S., Evans, P.D.: Violet light causes photodegradation of wood beyond the zone affected by ultraviolet radiation. Holzforschung 61, 23–27 (2007). https://doi. org/10.1515/HF.2007.005 Kocaefe, D., Poncsak, S., Boluk, Y.: Effect of thermal treatment on the chemical composition and mechanical properties of birch and aspen. BioResources 3(2), 517–537 (2008) Kubovsky, I., Kacik, F.: Colour and chemical changes of the lime wood surface due to CO2 laser thermal modification. Appl. Surf. Sci. 321, 261–267 (2014). https://doi.org/10.1016/j.apsusc. 2014.09.124

References

299

Lanzalunga, O., Bietti, M.: Photo- and radiation chemical induced degradation of lignin model compounds. J. Photochem. Photobiol. B: Biol. 56(2–3), 85–108 (2000). https://doi.org/10.1016/ S1011-1344(00)00054-3 Li, M.Y., Cheng, S.C., Li, D., Wang, S.N., Huang, A.M.: Structural characterization of steam-heat treated Tectona grandis wood analyzed by FT-IR and 2D-IR correlation spectroscopy. Chin. Chem. Lett. 26(2), 221–225 (2015). https://doi.org/10.1016/j.cclet.2014.11.024 Liu, X.Y., Timar, M.C., Varodi, A.M., Sawyer, G.: An investigation of accelerated temperatureinduced ageing of four wood species: colour and FTIR. Wood Sci. Technol. 21, 357–378 (2016). https://doi.org/10.1007/s00226-016-0867-4 Liu, R., Pang, X., Yang, Z.: Measurement of three wood materials against weathering during long natural sunlight exposure. Measurement 102, 179–185 (2017). https://doi.org/10.1016/j.measur ement.2017.01.034 Matsuo, M., Yokoyama, M., Umemura, K., Gril, J., Yano, K., Kawai, S.: Color changes in wood during heating: kinetic analysis by applying a time-temperature superposition method. Appl. Phys. A 99, 47–52 (2010). https://doi.org/10.1007/s00339-010-5542-2 Matsuo, M., Umemura, K., Kawai, S.: Kinetic analysis of color changes in keyaki (Zelkova serrata) and sugi (Cryptomeria japonica) wood during heat treatment. J. Wood Sci. 60, 12–20 (2014). https://doi.org/10.1007/s10086-013-1369-8 Mattonai, M., Watanabe, A., Shiono, A., Ribechini, E.: Degradation of wood by UV light: a study by EGA-MS and Py-GC/MS with on line irradiation system. J. Anal. Appl. Pyrol. 139, 224–232 (2019). https://doi.org/10.1016/j.jaap.2019.02.009 Mikleˇci´c, J., Jirous-Rajkovi´c, V., Antonovi´c, A., Špani´c, N.: Discoloration of thermally modified wood during simulated indoor sunlight exposure. BioResources 6, 434–446 (2011) Mitchell, P.H.: Irreversible property changes of small loblolly pine specimens heated in air, nitrogen, or oxygen. Wood Fiber Sci. 20(3), 320–355 (1988) Müller, U., Rätzsch, M., Schwanninger, M., Steiner, M., Zöbl, H.: Yellowing and IR changes of spruce wood as result of UV-irradiation. J. Photochem. Photobiol. B: Biol. 69, 97–105 (2003). https://doi.org/10.1016/S1011-1344(02)00412-8 Nemeth, K., Vano, V., Faix, O.: The effect of wood extractives on the photodegradation of wood. In: EWLP conference, Grenoble, France, pp 191–192 (1992) Nemeth, R., Hill, CAS., Takats, P., Tolvaj, L.: Chemical changes of wood during steaming measured by IR spectroscopy. Wood Mat. Sci. Eng. 11(2), 95–101 (2016). https://doi.org/10.1080/174 80272.2014.961169 Okon, K.E., Lin, F., Chen, Y., Huang, B.: Effect of silicone oil heat treatment on the chemical composition, cellulose crystalline structure and contact angle of Chinese parasol wood. Carbohyd. Polym. 164, 179–185 (2017). https://doi.org/10.1016/j.carbpol.2017.01.076 Pandey, K.K.: Study of the effect of photo-irradiation on the surface chemistry of wood. Polym. Degrad. Stab. 90, 9–20 (2005). https://doi.org/10.1016/j.polymdegradstab.2005.02.009 Pandey, K.K., Vuorinen, T.: Comparative study of photodegradation of wood by a UV laser and a xenon light source. Polym. Degrad. Stab. 93, 2138–2146 (2008). https://doi.org/10.1016/j.pol ymdegradstab.2008.08.013 Pastore, T.C.M., Santos, K.O., Rubim, J.C.: A spectrocolorimetric study on the effect of ultraviolet irradiation of four tropical hardwoods. Biores. Technol. 93, 37–42 (2004). https://doi.org/10. 1016/j.biortech.2003.10.035 Persze, L., Tolvaj, L.: Photodegradation of wood at elevated temperature: colour change. J. Photochem. Photobiol. B: Biol. 108, 44–47 (2012). https://doi.org/10.1016/j.jphotobiol.2011. 12.008 Popescu, C.M., Popescu, M.C., Vasile, C.: Structural analysis of photodegraded lime wood by means of FT-IR and 2D IR correlation spectroscopy. Int. J. Biol. Macromol. 48(4), 667–675 (2011). https://doi.org/10.1016/j.ijbiomac.2011.02.009 Popescu, M.C., Froidevaux, J., Navi, P., Popescu, C.M.: Structural modifications of Tilia cordata wood during heat treatment investigated by FT-IR and 2D IR correlation spectroscopy. J. Molecul. Struct. 1033, 176–186 (2013). https://doi.org/10.1016/j.molstruc.2012.08.035

300

5 Applications of IR Spectrum Measurement in Wood Research

Preklet, E., Tolvaj, L., Bejo, L., Varga, D.: Temperature dependence of wood photodegradation. Part 2: evaluation by Arrhenius law. J. Photochem. Photobiol. A: Chem. 356, 329–333 (2018) Preklet, E., Tolvaj, L., Visi-Rajczi, E., Hofmann, T.: Effect of water leaching on photodegraded scots pine and spruce timbers monitored by FTIR spectroscopy. Forest 12, Paper ID:833 (2021a). https://doi.org/10.1016/j.jphotochem.2018.01.008 Preklet, E., Tolvaj, L., Tsuchikawa, S., Varga, D.: Photodegradation properties of earlywood and latewood spruce timber surfaces. Acta Silv. Lign. Hung. 17, 9–21 (2021b). https://doi.org/10. 37045/aslh-2021-0001 Reinprecht, L., Mamonová, M., Pánek, M., Kacik, F.: The impact of natural and artificial weathering on the visual, colour and structural changes of seven tropical woods. Eur. J. Wood Prod. 76, 175–190 (2018). https://doi.org/10.1007/s00107-017-1228-1 Sanz, M., Fernández de Simon, B., Esteruelas, E., Munoz, A.M., Cadahia, E., Hernández, T., Estrella, I., Pinto, E.: Effect of toasting intensity at cooperage on phenolic compounds in Acacia (Robinia pseudoacacia) Heartwood. J. Agric. Food Chem. 59, 3135–3145 (2011). https://doi. org/10.1021/jf1042932 Srinivas, K., Pandey, K.: Photodegradation of thermally modified wood. J. Photochem. Photobiol. B: Biol. 117, 140–145 (2012). https://doi.org/10.1016/j.jphotobiol.2012.09.013 Stamm, A.J.: Thermal degradation of wood and cellulose. Ind. Eng. Chem. 48(3), 413–417 (1956) Temiz, A., Yldiz, U.C., Aydin, I., Eikenes, M., Alfredsen, G., Colakoglu, G.: Surface roughness and color characteristics of wood treated with preservatives after accelerated weathering test. Appl. Surf. Sci. 250(1–4), 35–42 (2005). https://doi.org/10.1016/j.apsusc.2004.12.019 Temiz, A., Terziev, N., Eikenes, M., Hafren, J.: Effect of accelerated weathering on surface chemistry of modified wood. Appl. Surf. Sci. 153(12), 5355–5362 (2007). https://doi.org/10.1016/j.apsusc. 2006.12.005 Timar, MC., Varodi, A., Hacibektasoglu, M., Campean, M.: Color and FTIR analysis of chemical changes in Beech Wood (Fagus sylvatica L.) after light steaming and heat treatment in two different environments. BioResources 11(4), 8325–8343 (2016a) Timar, M.C., Varodi, A.M., Gurau, L.: Comparative study of photodegradation of six wood species after short-time UV exposure. Wood Sci. Technol. 50, 135–163 (2016b). https://doi.org/10.1007/ s00226-015-0771-3 Tjeerdsma, M., Boonstra, M., Pizzi, A., Tekely, P., Millitz, H.: Characterisation of thermally modified wood: molecular reasons for wood performance improvement. Holz Roh Wekstoff 56, 149–153 (1998) Tjeerdsma, B.F., Militz, H.: Chemical changes in hydrothermal treated wood: FTIR analysis of combined hydrothermal and dry heat-treated wood. Holz Roh Werkstoff 63, 111–201 (2005). https://doi.org/10.1007/s00107-004-0532-8 Tolvaj, L., Faix, O.: Artificial ageing of wood monitored by DRIFT Spectroscopy and CIE L*a*b* color measurements. I. Effect of UV light. Holzforschung 49, 397–404 (1995). https://doi.org/ 10.1515/hfsg.1995.49.5.397 Tolvaj, L., Mitsui, K.: Light source dependence of the photodegradation of wood. J. Wood Sci. 51, 468–473 (2005). https://doi.org/10.1007/s10086-004-0693-4 Tolvaj, L., Mitsui, K., Varga, D.: Validity limits of Kubelka-Munk theory for DRIFT spectra of photodegraded solid wood. Wood Sci. Technol. 4, 135–146 (2011). https://doi.org/10.1007/s00 226-010-0314-x Tolvaj, L., Varga, D.: Photodegradation of timber of three hardwood species caused by different light sources. Acta Silv. Lign. Hung. 8, 145–155 (2012) Tolvaj, L., Molnar, Zs., Nemeth, R.: Photodegradation of wood at elevated temperature: Infrared spectroscopic study. J. Photochem. Photobiol. B: Biol. 121, 32–36 (2013). https://doi.org/10. 1016/j.jphotobiol.2013.02.007 Tolvaj, L., Molnar, Zs., Magoss, E.: Measurement of photodegradation-caused roughness of wood using a new optical method. J. Photochem. Photobiol. B: Biol. 134, 23–26 (2014). https://doi. org/10.1016/j.jphotobiol.2014.03.020

References

301

Tolvaj, L., Tsuchikawa, S., Inagaki, T., Varga, D.: Combined effects of UV light and elevated temperatures on wood discoloration. Wood Sci. Technol. 49, 1225–1237 (2015). https://doi.org/ 10.1007/s00226-015-0749-1 Tolvaj, L., Popescu, C.M., Molnar, Zs., Preklet, E.: Effects of air relative humidity and temperature on photodegradation processes in Beech and Spruce Wood. BioResources 11(1), 296–305 (2016) Varga, D., Tolvaj, L., Tsuchikawa, S., Bejo, L., Preklet, E.: Temperature dependence of wood photodegradation monitored by infrared spectroscopy. J Photochem. Photobiol. A: Chem. 348, 219–225 (2017). https://doi.org/10.1016/j.jphotochem.2017.08.040 Varga, D., Tolvaj, L., Molnar, Zs., Pasztory, Z.: Leaching effect of water on photodegraded hardwood species monitored by IR spectroscopy. Wood Sci. Technol. 54, 1407–1421 (2020). https://doi. org/10.1007/s00226-020-01204-2 Varga, D., Tolvaj, L., Preklet. E.: Colour stability of steamed black locust, beech and spruce timbers during short-term photodegradation. Wood Res. 66(4), 544–555 (2021) Wagenführ, R.: Holzatlas (Wood atlas). Fachbuchverlag Leipzig, Köln (2007) Windeisen, E., Stroble, C., Wegener, G.: Chemical changes during the production of thermotreated beech wood. Wood Sci. Technol. 41, 523–536 (2007). https://doi.org/10.1007/s00226007-0146-5 Xin, Y.L., Timar, M.C., Varodi, A.M., Sawyer, G.: An investigation of accelerated temperatureinduced ageing of four wood species: colour and FTIR. Wood Sci. Technol. 51, 357–378 (2017). https://doi.org/10.1007/s00226-016-0867-4 Yildiz, S., Tomak, E.D., Yildiz, U.C., Ustaomer, D.: Effect of artificial weathering on the properties of heat treated wood. Polym. Degrad. Stab. 98, 1419–1427 (2013). https://doi.org/10.1016/j.pol ymdegradstab.2013.05.004 Yu, Y., Xiao, Z., Liang, D., Wang, Y., Militz, H., Xie, Y.: Accelerated weathering performance of plantation-grown juvenile poplar and Chinese fir woods. Holzforschung 77(2), ID 0133 (2022). https://doi.org/10.1515/hf-2022-0133

Index

A Acetyl group, 225–228, 290 African mahogany, 71 Air humidity, 58, 105, 163, 167, 171, 174, 188, 190, 205, 262, 266 Alder, 61, 87, 143, 145, 146, 148, 161, 174, 175, 207, 296 Aromatic ring, 38, 41, 166, 174, 226, 232–234, 240, 253, 269, 274, 275, 279, 281, 283, 288, 290 Arrhenius law, 91, 92, 104, 178, 186, 258, 262, 295 Arrhenius plot, 92, 104, 105, 113, 114, 184–188, 256–261 Ash, 20, 61, 82, 83, 87, 174, 175, 178, 179, 184–187, 206, 208, 237, 238, 253 Aspen, 208, 267–276

B Baseline correction, 31, 33, 34, 44, 45, 224, 227 Baseline shift, 33, 44–46, 229, 234, 238, 246 Beech, 34, 35, 37–40, 58, 59, 61, 87, 91, 117–122, 124, 128, 136, 142–147, 149–152, 164, 166, 174–177, 188–190, 197–202, 204, 206, 208, 227–239, 247, 248, 250–254, 259, 260, 262, 264–271, 273–277, 288–290, 293, 295 Beer-Lambert law, 4, 5, 14 Birch, 61, 78, 87, 143, 145–147, 149–152, 174, 175, 204, 208 Black locust, 26, 28, 32–34, 45, 46, 59–63, 75, 87, 91, 92, 94–102, 104–106,

108–116, 118, 120, 124, 127, 128, 136, 142, 143, 146, 147, 149, 151, 159–166, 168–170, 174–176, 197–204, 208–217, 223, 226, 246, 247, 267–271, 273–276, 278, 288–296 Blackwood, 71

C Carbonyl group, 34, 35, 38, 41, 43, 166, 229, 234, 237, 239–244, 250, 252, 260, 262, 264, 269, 270, 275–279, 281, 283–285, 287–292 Cellulose, 29–32, 34, 38, 41, 46, 174, 191, 208, 224–226, 231, 232, 234, 236, 240, 241, 250, 253–257, 262, 263, 270, 277, 279, 287, 289, 290, 293 Characteristic wavelength, 23, 66 Cherry wood, 84 Chroma, 51, 53, 54, 57, 58, 72–75, 98, 99, 105, 113, 120, 124, 129, 132, 133, 138–140, 147, 148, 152, 166 Chromaticity diagram, 51, 54, 55 Colour homogenisation, 91, 115–117, 120, 121, 124, 125, 133, 134, 136, 138–141 Colouring material, 60 Colour matching functions, 53, 65, 69, 70 Colour modification, 91, 92, 100, 105–107, 117, 130, 143, 150, 166, 168, 178, 191, 197, 206, 209, 248, 287 Colour space, 53, 54, 56–58, 68, 69, 73, 76

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 L. Tolvaj, Optical Properties of Wood, Smart Sensors, Measurement and Instrumentation 45, https://doi.org/10.1007/978-3-031-46906-0

303

304 D Deacetylation, 223, 226, 227, 229–231, 290, 291, 294 Density, 5, 31, 80, 82, 85, 86, 88, 117, 133, 144, 149, 164, 166, 172, 232, 238, 246, 288 Difference spectrum, 1, 13, 32, 36, 37, 41–43, 143, 223, 224, 226–230, 233–237, 239, 250, 256, 262–266, 268, 278, 280, 288, 291, 294 Diffuse reflection, 2, 4, 39, 76, 78 Discoloration, 163, 188, 208, 232 Douglas fir, 61

E Earlywood, 21, 79, 93, 101, 104, 125, 126, 130, 133–141, 164, 191, 232, 237–245, 249, 267, 278, 280 Emission spectrum, 52, 53, 171–173, 245 Equilibrium moisture, 31, 92, 100, 107 Ether bond, 231, 233, 234, 238, 240–242, 249, 257, 258, 262, 270, 279, 290, 292, 295 Eucalyptus, 20, 60, 80–82, 87 Extractives, 3, 26, 60, 61, 63, 87, 92, 94, 95, 97, 98, 100–103, 105, 107, 108, 110, 115, 118, 119, 126, 127, 130, 134, 144–146, 157, 160–162, 164, 170, 174–177, 181, 184, 186, 191, 195, 198, 200, 201, 203, 204, 208, 209, 211, 214–217, 229, 231, 238, 240, 242, 248, 252, 253, 256, 267, 268, 270, 271, 274, 275, 288–291, 295, 296

F Fourier transformation, 2, 8 Frequency, 2, 3, 5, 10, 27, 33

G Geometric mean, 65, 75 Gloss measurement, 51, 52, 77, 78, 86 Gloss unit, 77 Grain direction, 37, 79 Guaiacyl lignin, 26, 35, 38, 225, 226, 229, 234, 235, 237, 240, 242, 243, 246, 249, 258, 262–265, 269, 274, 275, 277, 278, 283, 287, 290, 292, 295, 296

Index H Haematin, 60 Hardwood, 34, 35, 61, 77, 108, 129, 164, 178, 226, 227, 231, 233, 248, 250, 252, 253, 259, 262, 267, 269, 278, 291, 296 Heartwood, 20, 21, 38, 60, 61, 79, 80, 92, 101, 108, 117–130, 133–141, 143, 164, 168, 174, 197, 237, 238, 268, 289, 295 Heat treatment, 78, 87, 100, 117, 181, 205–209, 211, 215, 216, 226, 248, 249, 295 Hemicelluloses, 34, 38, 92, 94, 97, 98, 100–103, 110, 118, 126, 127, 129, 174, 181, 186, 209, 225–227, 229–231, 234, 241, 250, 252, 256, 258, 259, 262, 263, 270, 277, 279, 287, 290, 291, 293, 295, 296 Hornbeam, 61, 208, 226 Hue, 21–24, 26, 51–55, 57–59, 61–66, 68–72, 85, 94, 97–99, 105, 108, 112, 113, 120, 121, 124–126, 128–130, 132–134, 136–142, 147, 148, 152, 160–162, 166, 167, 170, 171, 180, 181, 195, 196, 200–207, 211, 216, 217

I Illuminant, 52–54, 57, 58 Incident light, 3, 4, 52, 61, 76–79, 163, 172, 188 Infrared spectrum, 33, 223 Integrating sphere, 7, 8

K Kadam wood, 71, 72 Karri wood, 60, 80 Kashmir walnut, 71, 72 Ketone, 26, 225, 234, 240, 249, 269, 279, 290, 292, 294 K-M function, 5, 12, 13, 270 K-M spectrum, 5, 10, 13, 43, 44, 271 Kubelka-Munk equation, 5, 242, 291

L Lacquer, 141, 150–152 Lactone, 225, 234, 236, 240, 255, 269, 279, 290, 292 Lambertian scatterer, 7

Index Larch, 34, 35, 38, 61, 91, 92, 95–98, 101, 104, 105, 133–139, 143, 145–152, 174–176, 238, 239 Laser, 52, 58, 172, 253 Latewood, 21, 77, 79, 86, 93, 101, 104, 125, 126, 130, 131, 133–141, 191, 232, 238–245, 280 Lightness, 21, 51–66, 69–76, 94, 95, 100–102, 104–106, 108–110, 113, 118, 119, 122, 123, 125, 126, 130, 131, 134, 137, 141, 143–145, 147, 148, 151, 159, 160, 164, 166, 168, 169, 177, 179–186, 189, 191, 192, 195–198, 209–211, 216, 217, 232 Light source, 7–9, 52, 76, 157, 171–173, 178, 223, 232, 245, 249 Lignin, 3, 26, 29, 34, 35, 41, 43, 60, 103, 104, 118, 130, 144, 160–162, 166, 170, 174, 176, 177, 181, 183, 184, 186, 187, 194, 195, 200, 201, 203, 204, 208, 213–217, 223, 225, 226, 231, 232, 236–240, 242, 243, 245, 246, 249, 252–256, 259, 262–266, 269, 270, 272, 274–277, 279, 281, 282, 284, 287, 289–293, 295, 296 Linden, 87, 143–147, 174, 175, 226

M Maple, 61, 87, 161, 174, 175 Mercury lamp, 65, 173, 178, 188, 189, 191, 232, 236–238, 245, 247–249, 251, 254, 255, 267, 283 Methoxyl group, 225 Methyl group, 34 Michelson interferometer, 1, 2, 8 Moisture content, 29–31, 91, 108, 109, 125, 130, 134, 139, 152, 159, 189, 191, 224, 227, 267, 288 Myrtle, 61

N Nitrogen, 100, 206

O Oak, 61, 83, 85, 87, 143, 145–149, 159–161, 168–170, 174, 175, 267–270, 272–275, 277, 278 Oxygen, 100, 163, 166, 174, 206, 231, 240, 269, 278, 290

305 P Photodegradation, 36–38, 40–43, 45, 46, 58, 63–65, 87, 115, 149, 157–164, 170–174, 176–180, 182, 184, 186–188, 193, 195, 197–201, 203–206, 208, 209, 211–217, 223, 231, 232, 236–239, 242, 246, 248, 250, 252, 255–258, 260–263, 266, 267, 270, 271, 273–275, 277–280, 283, 284, 287–289, 291, 292, 294–296 Photon, 1–7, 20, 27, 28, 39, 52, 65, 66, 70, 141, 144, 149, 157, 158, 166, 167, 172, 178, 195, 223, 269 Physical lightness, 21, 22, 65–67 Poplar, 10–14, 16–19, 26, 28, 30, 31, 40–44, 61, 62, 72, 91, 92, 94–99, 101, 102, 104–107, 125–129, 143–147, 150–152, 161, 164, 166, 168–170, 174–176, 178, 181–184, 187, 197, 198, 200, 204–206, 208, 209, 211, 213–217, 226, 227, 230, 237, 238, 240, 248, 250, 251, 253, 255, 256, 259, 260, 288–292, 294, 295

Q Quercetin, 60

R Redness, 59, 95–98, 101–104, 106, 110, 111, 113–115, 118–120, 122, 123, 126–128, 130–132, 134–136, 144–147, 151, 152, 160, 161, 164–167, 169–171, 174–177, 182–184, 186–192, 194–196, 198, 199, 201–206, 211–213, 216, 217, 248 Red tingle, 81, 82 Reflection spectrum, 23, 24, 45, 52, 64, 65, 71, 74 Robinetin, 61, 97, 102, 108, 110–112, 115, 170, 203, 268, 295 Rosewood, 80, 81 Roughness, 6, 12, 37, 39–41, 43–46, 77, 79, 81–85, 87, 229, 242, 266, 270, 271, 280, 291

S Sappan wood, 60

306 Sapwood, 60, 61, 79, 80, 101, 118, 121–130, 133–141, 164, 168, 174, 238, 289 Saturation, 51–54, 58, 69, 72–75, 91, 161, 170 Scattering, 3–5, 33, 34, 43, 44, 81, 86, 126, 144, 149, 229, 242, 270 Scots pine, 20, 26, 28, 45, 46, 61, 62, 87, 88, 91, 92, 95–99, 101, 103–105, 130–132, 136, 143, 146, 150, 151, 168–170, 174, 175, 178, 179, 186, 187, 208, 235–239, 248–251, 253, 254, 259, 260, 262–264, 278–287 Shape factor, 23 Softwood, 34, 35, 38, 77, 130, 133, 139, 164, 178, 231, 233, 248, 250, 252, 253, 262, 267, 290 Spectrophotometer, 1, 2, 7–9, 13, 14, 17, 19, 20, 25, 28, 33, 35, 39, 45, 53, 224 Specular reflection, 7, 76 Spruce, 20, 34, 45, 61–63, 65, 75, 91–99, 101, 103–105, 130–133, 136, 143–148, 150, 164, 168, 170, 174–176, 178, 179, 181–184, 186–201, 204–208, 226, 227, 230, 231, 234–236, 238–245, 248–250, 253, 255, 256, 259, 260, 278–291, 294, 295 Standard deviation, 26, 88, 115, 116, 134 Standard observer, 53, 58, 69, 70 Steaming, 31, 32, 59, 60, 62, 63, 91, 92, 105–144, 149, 151, 161, 174, 175, 195, 197–207, 209, 217, 223, 226–231, 287, 288, 290–292, 295, 296 Sugi, 14–16, 79, 80, 91, 105, 133, 139–141, 164, 275 Sunlight, 4, 52, 88, 157, 158, 162–164, 166, 168, 170, 172, 245, 246 Sun radiation, 64, 157, 158, 163–173, 188, 223, 231, 245, 247, 248, 267 Syringyl lignin, 38, 225, 226, 233–235, 237, 253, 259, 264, 265, 269, 273–276, 288, 290, 292, 295 T Tannins, 26, 60, 133 Thermal effect, 92, 179, 180, 184, 253, 256 Thermal treatment, 32, 78, 86, 88, 91–107, 126, 136, 178–188, 197, 206, 208–217, 229, 248, 250, 253, 255, 256, 259, 262, 270, 288 Total reflection, 9, 17

Index Tracheid, 77 Transmission spectrum, 9, 166 Tristimulus values, 53–55, 63, 65, 69, 70, 72, 74, 75, 95 Turkey oak, 61, 91, 117, 121–125, 128, 136, 226

U Ultraviolet light, 59, 149, 157, 158, 172, 174, 178, 179, 182, 186, 188–191, 194, 198, 199, 206–209, 216, 217, 248, 267, 274, 278, 287, 288, 295 Ultraviolet spectrum, 19 Urushi wood, 60

V Viewing angle, 76, 77, 79, 83, 86 Visible light, 3, 10, 20, 26, 60, 64, 104, 149, 158, 173, 232 Visible spectrum, 56, 57, 113

W Walnut, 81, 82, 84, 85, 87, 174, 175, 177 Water leaching, 64, 157, 191–196, 252, 267, 268, 270–278, 280–287 Wavelength, 2–7, 9–11, 14, 16–23, 26, 27, 29, 31, 33, 34, 37, 38, 40, 41, 43, 52, 54, 56–58, 60, 65–69, 71, 72, 77, 93, 95, 105, 106, 113, 142, 151, 157, 158, 163, 166, 167, 171–173, 208, 225, 228, 232, 245, 249, 288 Wavenumber, 5, 10, 11, 19, 27, 29, 33, 34, 40, 43, 45, 225, 227–229, 231–233, 235, 240–243, 245, 250, 252, 256, 258, 259, 261, 262, 266, 269–271, 273–275, 278–281, 283–285, 287–293 Waxes, 61, 101, 163 Weathering, 87, 88, 152, 157–162, 171, 190, 207, 208, 238, 262, 266, 267, 278 Wetting, 30, 91, 141–152 Window glass, 163, 166, 167, 169, 170

X Xenon lamp, 171, 172, 208, 238, 245, 246 Xylan, 34, 225, 227, 229, 231, 289, 291

Index Y Yellowness, 59, 97, 98, 102–106, 110–115, 120, 124, 128, 131, 132, 136–138,

307 140, 145–147, 151–153, 160, 161, 165–167, 169–171, 176, 183–190, 192–196, 199–206, 213–215