High Performance Coatings for Automotive and Aerospace Industries [1 ed.] 9781616683764, 9781608765799

Citation preview

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

MATERIALS SCIENCE AND TECHNOLOGIES

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

HIGH PERFORMANCE COATINGS FOR AUTOMOTIVE AND AEROSPACE INDUSTRIES

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in High Performance Coatings for Automotive and Aerospace Nova Scienceservices. Publishers, Incorporated, 2010. ProQuest Ebook Central, rendering legal, medical or anyIndustries, other professional

MATERIALS SCIENCE AND TECHNOLOGIES Additional books in this series can be found on Nova‘s website under the Series tab.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Additional E-books in this series can be found on Nova‘s website under the E-books tab.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

MATERIALS SCIENCE AND TECHNOLOGIES

HIGH PERFORMANCE COATINGS FOR AUTOMOTIVE AND AEROSPACE INDUSTRIES

ABDEL SALAM HAMDY MAKHLOUF Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

EDITOR

Nova Science Publishers, Inc. New York

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010 by Nova Science Publishers, Inc. All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers‘ use of, or reliance upon, this material.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA High performance coatings for automotive and aerospace industries / editor, Abdel Salam Hamdy Makhlouf. p. cm. Includes index. ISBN H%RRN 1. Protective coatings. 2. Corrosion and anti-corrosives. 3. Automobiles--Painting. 4. AirplanesPainting. 5. Metals--Finishing. I. Makhlouf, Abdel Salam Hamdy. TA418.76.H53 2009 629.2'32--dc22 2009043042

Published by Nova Science Publishers, Inc. New York High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

CONTENTS

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Preface

vii

Chapter 1

Novel Silicone Ceramer Coatings for Aluminum Protection Atul Tiwari and L. H. Hihara

Chapter 2

Thermally Stable Coatings for the Corrosion Protection of Magnesium Alloys: Double Layered Coatings Consisting of a Nanoparticulate Primer and a Sol-Gel Sealing Florian Feil and Wolfram Fürbeth

Chapter 3

Sol-Gel Enhanced Ni-P Composite Coatings Weiwei Chen, Wei Gao and Yedong He

Chapter 4

Protective Coatings Based on Sol-Gel Chemistry: A Review FX Perrin

Chapter 5

High-Speed Laser-Assisted Surface Modification Igor V. Shishkovsky

Chapter 6

An Attempt for Designing Economically Attractive Chrome-Free Conversion Coatings for Magnesium Alloys Abdel Salam Hamdyand Mahmoud Farahat

Chapter 7

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers for Space Applications I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Chapter 8

Plating of Nano-composites- Overview and Trends Subir Kumar Ghosh and Jean-Pierre Celis

Chapter 9

Improvement of the Reinforcements Distribution in the Composite Matrices Using Powder Coating Process Walid M. Daoush, Byung K. Lim, Hee S. Park, Sayed F. Moustafa and Soon H. Hong

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

1

47 57

83 111

129

141

229

301

vi Chapter 10

Chapter 11

Contents Mechanical and Electrical Characterization of Mwcnt/Cnp Filled Polymer Composites M S Vinod and Dr. V J Sundaram Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission Spectrometry (GD-OES) Tamás I. Török, Gábor Lévai, Mária Szabó and József Pallósi

Chapter 12

Corrosion Monitoring using Impedance Data D. M. Bastidas, E. Cano, E. M. Mora and J. M. Bastidas

Chapter 13

EIS Technique used to Protective Performance Assessment of Organic Coatings: Applicability in Car Component Manufacturing Pierluigi Traverso and Giorgio Luciano

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Index

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

325

337

353

387 401

PREFACE

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

DESCRIPTION The book of High Performance Coating for Automotive and Aerospace Industries is both a reference and a tutorial for understanding the most advanced coatings used in transportation industries. It discusses the basics of the materials used in such industries, including the durability to resist corrosion in severe environments. Also, it covers most of the coating processes, testing performance and surface characterization techniques. It addresses important questions frequently posed by end-user design engineers, coaters, and coatings suppliers in their quest for superior coating qualities for such industries. This book focuses on the industrial manufacturing and application aspects of various coating, thin film and surface engineering technologies for automotive and aerospace applications to fulfill the demands in the manufacturing industry. Also of particular interest are surface treatments before and after coating to enhance the performance of engineered surfaces, innovations in manufacturing practices and advances in industrial deposition equipment. Current coatings used in automotive and aerospace industries must be tuned to operate effectively under extreme conditions like high temperature, high pressure, intense chemical or mechanical wear, and a combination of those. Therefore, contributions in this book will emphasize corrosion, tribological and wear aspects of hard, superhard, nanocomposite and lubricious coatings with respect to their mechanical and physical properties.

SCOPE OF STUDY This book contains:    

An in-depth analysis of the technologies used for high-performance nano, composites and ceramic coatings. An overview of materials for high-performance coatings and their properties. New developments in high-performance coating techniques. Current and potential applications for high-performance coatings.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: High Performance Coatings for Automotive and Aerospace… ISBN: 978-1-60876-579-9 Editor: Abdel Salam Hamdy Makhlouf, pp. 1-46 ©2010 Nova Science Publishers, Inc.

Chapter 1

NOVEL SILICONE CERAMER COATINGS FOR ALUMINUM PROTECTION Atul Tiwari and L. H. Hihara Hawaii Corrosion Laboratory, Department of Mechanical Engineering, University of Hawaii at Manoa, Honolulu, HI 96822, USA

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION Coatings encompass a relatively diverse class of materials and predominantly serve the purpose of preventing metallic materiel from reverting back to their natural oxide state by corrosion. Studies coordinated by the German Research Society of Surface Treatment, the National Coating and Paint Association, and other independent agencies demonstrated the need for research and development on different coating materials [1-4]. Today, coatings are used to impart a variety of properties and characteristics to metal substrates such as corrosion resistance; aesthetics; antifouling, antibacterial, self-sensing, and self-cleaning abilities; wear and scratch resistance; thermal conductivity; UV resistance; etc. Moreover, coatings should ideally be easy to apply, be environmentally friendly, and allow for the incorporation of fillers for other desired characteristics. Sol-gels, which are an interesting class of materials, have significant promise to meet many of the coating challenges of the 21st century. The sol-gel process, however, has been known for a very long time, and was first investigated in 1845 by Ebelmen [5]. The major thrust to this field was later provided by Roy [6]. Several recent and excellent reviews are available on sol-gel science [7-12]. The sol-gel process has drawn significant attention due to its advantages such as low cost, ease of reaction between different constituents, control over the final structure at nano-dimensional scales, and room temperature processing. The sol-gel process can be controlled and involves hydrolysis of alkoxy metal compounds followed by a condensation reaction and hardening process. One particular system, namely silicon-based sol-gel coatings, will be the focus of this chapter. Preparation of silicon-based sol-gel coatings begins with mixing alkoxysilane with water and a desired volatile solvent. Hydrolysis of alkoxysilane results into intermediate silanol (Si-OH) moieties that condense with the release of water molecules and the evaporation of the solvent. The silanol moieties further condense

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

2

Atul Tiwari and L. H. Hihara

to form SiOSi polymeric chain macromolecules. The three-dimensional network formed as a result of crosslinking between the silanol moieties consists of a porous structure that condenses further at higher temperatures or with the passage of time, resulting in a dense ceramic material [13]. The sol-gel process also allows the synthesis of hybrid ceramicpolymer materials or ceramers by the combination of ceramic and polymers. Ceramers represent a network system prepared by a sol-gel reaction of functional organic polymers with a metal alkoxy compound. These materials are also called ―Ormosil‖ when organic polymers reacts with alkoxysilanes [14]. This chapter focuses on prior work on ormosil or ceramer coatings, with the intent of informing the reader of the many aspects and facets of synthesizing, characterizing, and testing the coatings.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1.1 Background Corrosion of metals can be prevented with the use of protective coatings, which can be either barrier, inhibitive, or sacrificial. The selection of a coating generally depends on the substrate and intended application. The effectiveness of barrier coatings in preventing ingress and accumulation of moisture and corrosives at the substrate generally improves with better packing of polymeric chains, minimization of coating defects such as pinholes, and better adhesion to the substrate. Inhibitive coatings contain corrosion inhibitors to mitigate corrosion. One example is chromate conversion coatings that are very effective on aluminum, but may be banned in the future by the Environmental Protection Agency [15] due to health issues. These chromate conversion coatings are applied on aluminum using mixtures of hexavalent chromium salt and chromic acid. The oxidation-reduction reactions between the chromium compounds and the aluminum surface results in a thin film of insoluble trivalent chromium and soluble hexavalent chromium compounds. These hexavalent compounds leach out when the coating is breached and form insoluble trivalent chromium at the site of damage [16]. Sacrificial coatings contain active metals such as zinc or magnesium that acts as a sacrificial anode in a galvanic couple. Corrosion protection is afforded to the substrate as long as a sufficient amount of sacrificial coating is still present. The ceramer silicone-based coating is primarily a barrier coating. The silicone resin combines the advantage of inorganic ceramics with that of organic polymers. Silicone constitutes a special class of coating materials that shows superior properties required for a wide variety of applications. Silicones are prepared by controlled hydrolysis of silane compounds that results in silanols and finally condenses to silicones. Silicones are frequently used as material of choice for different coating applications. Those that have alkyl or phenyl groups generally have desirable properties. For example, the ethyl group displays better flexibility, hardness, water repellency, and chemical resistance. These could be used when fast-curing rates, resistance to thermal shock, and low temperature application are desired. Phenyl-bearing silicones show good oxidative resistance and heat resistance as well as a longer shelf life and less thermo-plasticity compared to other members of the silicone family. However, properties of silicone coatings are affected by the amount of hydrocarbon they contain. A greater amount of organic hydrocarbons will increase flexibility, cure times, thermo-plasticity, and tackiness when cured in ambient conditions [17].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

3

Figure 1. Different applications of silicone clear coatings.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2. SYNTHESIS AND DEVELOPMENTS Formulation of a coating consists of three vital components - the solvent, pigment, and binder - which are crucial to curing and hardening. Solvents are usually a mixture of aromatic and aliphatic hydrocarbons, which control hydrolysis and prevent gelation. Pigments/fillers such as titanium dioxide are added to create whiteness as well as UV resistance; or to create redness, iron oxide or cadmium red is utilized. The binder and its curing process are very important in controlling the final properties of the coating. Most silicones cure completely in 60 min. at approximately 200 oC. However, the curing time and temperature can be reduced by adding an appropriate metal catalyst such as zinc or tin compounds. Silicone formulations are often combined with organic vehicles such as acrylic, epoxy, or alkyds to give them additional functionalities such as improved film forming capabilities, higher hardness, reduced thermo-plasticity, and faster curing times. The following subsections review different kinds of silicone coatings.

2.1 Silicone-Rich Coatings Coating formulations that have a silicone backbone without hydrocarbon are ideal glassy materials for coatings. However, they are brittle and may crack, rendering a less than adequate performance. Proper selection of coating components is therefore needed for the development of effective barrier coatings. Silicone-rich coatings can be prepared by adopting controlled hydrolysis followed by end termination with functional groups [18-20]. Polysilsesquixones are synthesized by acid catalyzed hydrolytic polycondensation of trimethoxymethylsilane, triisopropoxymethylsilane, or triisobutoxymethylsilane. Polysilsesquixones so obtained could be diluted with a volatile

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

4

Atul Tiwari and L. H. Hihara

solvent for dip coating. However, the performance of the coating depends on the polarity and crystallinity of the substrate. Also, polycondensation in coating occurs between hydroxyl groups or between hydroxyl and alkoxy groups only by elevated-temperature curing [21, 22]. Hybrid coatings could be prepared without an acid catalyst by reacting -glycidoxypropyltrimethoxysilane (GPTS) and tetraethoxysilane in ethanol using hexamethylenediamine (HMDA) and 3-aminopropyltrimethoxysilane (APTS) as a crosslinker, followed by the addition of a dibutyltindilaurate-hardening catalyst and ultrapure water [23]. An oligomeric polyfunctional sol-gel precursor that is obtained by combining cyclic siloxane with chlorosilane followed by hydrolysis or an ethanolysis process can be used to create a low molecular-weight monomeric compound that results in a dense silicone hybrid coating [24]. Such a coating provides good UV protection, improved abrasion resistance, and enhanced stability against acid compared to a traditional automotive clear coat. Barrier properties of 3-GPTS pretreatment on copper and aluminum alloys followed by powder coating of epoxy-polyester show that silane-treated substrates provide better corrosion resistance than floreo-zirconate-treated substrates [25]. Barrier properties of a water-based silane mixture prepared by reacting vinyltriacetoxysilane with bis-(treimethoxypropylsilane) amine display good adhesion to 2024Al-T3 and 6061-T6 Al surfaces. The precursor to the coating forms stable covalent bonding with the aluminum surfaces [26].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.2 Coatings via Silicone-Polymer Blending Silicones can be coupled with epoxy resin to create an interpenetrating coating system in which hydroxyterminated polydimethylsiloxane acts as a modifier, -amino-propyltrimethoxysilane (-APTMS) as a crosslinker, and dibutyltindilaurate as a catalyst. Polyamidoamine and aromatic polyamine adducts can be used to cure the coating. Thermal properties of such siliconized epoxy coating systems are higher than that of the unmodified epoxy coating systems. Morphological studies show heterogeneity in the siliconized epoxy coating, a characteristic that was amplified as the silicone concentration in the epoxy network increased [27]. Siliconized epoxy coatings can also be prepared using hydroxyterminated polydimethylsiloxane as a modifier, -APTMS as a crosslinker, and dibutyltindilaurate as a catalyst. Polyamidoamine, aromatic amine adducts, and phosphorous-containing diamines are used as catalysts in such a case. The corrosion and fouling resistance behavior of such siliconized epoxy coatings when evaluated using potentiodynamic polarization, electrochemical impedance spectroscopy, salt spray, and antifouling tests exhibit a lower corrosion current and higher paint film resistance than does the pure epoxy coating [28]. A coating suspension for improved scratch resistance of a polymethylmethacrylate surface was prepared by reacting silatrane with 3-GPTS in the presence of an acid catalyst. Scratch resistance of the coated surface increased with the increase in the alkoxysilane content in the coating. It is important to note that the curing time and temperature affects the scratch resistance and adhesion properties of the coating layer [29].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

5

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.3 Nano- and Molecular Composite Coatings Nano-composite coatings have at least one of the components with dimensions less than 100 nm. Such coatings show a high degree of crosslinking in the network structure due to the presence of nano dimensional modifiers [30-32]. Such nano particulates have a large aspect ratio (surface/volume) for reaction and can form strong bonds with other components, resulting in a stronger and tougher material with fewer defects [33]. Molecular composites on the other hand consist of a flexible moiety dispersed in a rigid macromolecular network. Rigid nano particulates can provide support to the flexible macromolecular unit, and can impart desired characteristics for specific applications. Sol-gel hybrid nano-composite coatings can be prepared by evaporation of solvent and induced partitioning of phases. Several hybrid polymer layers self-assembled to form strong covalent bonds between organic-inorganic moieties, resulting in an optically-transparent coating, displaying increased indentation hardness (i.e, 1.0 GPa), corresponding to a dense silica film. This hierarchical nano-composite coating displayed an oriented, nano-laminated under-layer bonded to an isotropic worm-micellar top-layer [34]. A scratch-resistance coating was developed by incorporating nanoparticles into a polymeric matrix. For example, methacroyloxy-propyl-trimethoxysilane modified nano-sized silica and alumina particles were used as fillers in a polyacrylate matrix nano-composite coating [35]. Silica nanoparticles can be fabricated via controlled hydrolysis of TEOS in the presence of acid. The size of silica nanoparticles depends on the duration of the hydrolysis reaction. Silica nanoparticles of different particle sizes were fabricated by functionalizing them with 3-(trimethoxysilyl)propyl methacrylate. The methoxysilane group terminated the particle growth and stabilized the nanoparticles [36]. Incorporation of uniformly-distributed nanoparticles may enhance thermo-mechanical stability of the resultant nano-composite coatings. For example, uniform distribution of nanoSiO2 enhance the thermal stability of an acrylic nano-composite coating, while an agglomeration of these nanoparticles in the polymer deteriorated the properties of the coatings. Formation of the Si-O-Si network by evenly distributed nano-SiO2 contributes to the antioxidation process, char-accumulation, and stable char architecture [37]. Similarly, a super hydrophobic surface was obtained by using a coating material that has low surface energy and adequate roughness at the micro/nano meter scale. A silicone nano-filament coating was prepared using an equimolar amount of trichloromethylsilane and water vapor displaying a high water contact angle (~higher than 150o) on coated substrates, an indication of the super hydrophobic nature of the coating [38]. Similarly, water-based, hybrid nano-composite coatings were prepared that showed good mechanical as well as other unique properties [39]. Molecular composites having a low dielectric constant, polyimide rigid phase, coupled with flexible polysiloxane phases, can be obtained using a solution-blending technique. Partly miscible coating compositions displayed uniform distribution of polysiloxane microspheres in a continuous polyimide matrix. The low surface energy of polysiloxane occupied an entire surface of the molecular composite coating that hindered the in-diffusion of water molecules. Such molecular composite coatings has the potential to effectively reduce corrosion in microelectronic and other electronic devices [40].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

6

Atul Tiwari and L. H. Hihara

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.4 Coating Containing Fillers and Pigments Active functionalities of fillers and pigments could modify characteristics of the coating. A hybrid coating was prepared by reacting tetraethoxysilane, tetrabutylorthosilane, tetrabutyorthotitanate, and hydroxylterminated polydimethysiloxane. The tetrabutylorthotitanate filler affected specific surface areas, pore size, and pore volume, but did not affect pore morphology. Specific surface area and pore volume decreased with the decrease in tetrabutylorthotitanate [41]. Similarly, a flexible hybrid material was obtained by reacting polydimethylsiloxane with alkoxide of zirconium and tantalum. It was discovered that the inorganic component in the material was in the form of oxide clusters and attached to polydimethylsiloxane through metal-oxygen bonds [42]. In another study, phenyltrimethoxysilane, 2-(3, 4-epoxycyclohexyl)trimethoxysilane, and tetramethoxysilane were reacted in ethanol. Hexamethylenediamine was used as crosslinker while zincacetate, carbon nanotubes, and Nanosil silica nanoparticles were used as nanofillers. A small amount of water was added to the formulation to induce a hydrolysis reaction and dibutyltindilaurate was used as a catalyst. The reactive sol formulation was left overnight to allow homogenization. This coating provided excellent corrosion protection of aluminum in chloride-containing atmospheres and solutions [23]. Another coating was synthesized by the reaction of polydimethylsiloxane, titaniumtetraisopropoxide, and ethylacetoacetate along with variable amounts of colloidal silica. The hardness of the coating remained proportional to the silica filler content and curing temperature, and inversely proportional to the concentration of polydimethysiloxane. Hydrophobicity was determined by measuring the water contact angle that increased with the increase in polydimethylsiloxane content in the coating. However, the hardness value decreased with the increase in polydimethylsiloxane concentration in the coating [43]. A sol-gel composition was obtained by reacting a mixture of titania nanoparticles in an epoxy polymer with glycidoxypropyltrimethoxysilane (GPTS) and TEOS. An increase in the TiO2 concentration decreased the rate of polymerization while the epoxy group conversion was induced — effects that may have been caused by the UV absorption competition between the TiO2 nano-particles and the photo initiator. An increase in the concentration of dispersed TiO2 gave rise to an opaque coating, while a transparent film was obtained when TiO2 nanoparticles were generated in-situ in the presence of a suitable coupling agent. Also, the UV photo curing of these coatings resulted in a homogeneous dispersion of inorganic particles into the organic matrix without macroscopic phase separation [44]. Organic corrosion inhibitors are also used in sol-gel organosilicate hybrid coatings. Coatings with incorporated organic corrosion inhibitors were examined by potentiodynamic polarization to investigate the inhibition activity of the entrapped compounds. Scanning vibrating electrode and electrochemical impedance techniques determined the effectiveness of corrosion inhibitors in the coating. Incorporation of several organic inhibitors in the coating reduced the corrosion on aluminum substrates [45]. The effect of cerium salt, 3-mercapto-propyl-trimethoxysilane, and octadecyltrimethoxysilane treatment was investigated. It was discovered that treating the metal substrate with cerium salt did not improve the corrosion resistance. However, treating the substrate with a hydroalcoholic solution or octadecyl-trimethoxysilane was effective in reducing corrosion. Moreover, treating the substrate with 3-mercapto-propyl-trimethoxysilane had less effect than in the former case [46]. The self-healing capability of surface pretreatments formulated from

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

7

cerium-nitrate doped (triethoxysilylpropyl) tetrasulfide silane solutions was evaluated. Precipitation of cerium oxides and hydroxides in anodic areas inhibited corrosion. The inhibitive mechanism of cerium-doped silane provided better corrosion resistance than did the base coating. However, the optimum/critical concentration of cerium nitrate in the coating appeared important in controlling corrosion [47]. Sol-gel coatings composed of dimethoxydimethylsilane, methyltriethoxysilane, and tetrapropoxyzirconium with an added organic corrosion inhibitor tertachloro-p-benzoquinone were prepared. Electrochemical impedance spectroscopy, atomic force microscopy, and glow discharge optical emission spectroscopy determined that increasing the concentration of the organic inhibitor did not provide enough corrosion protection due to the disorganization of the sol-gel system. However, lowering the concentration of the inhibitor resulted in homogenous coating structures with increased corrosion inhibition characteristics [48]. A coating formulation was synthesized using an inorganic salt in the sol of tetramethoxysilane, organic polymeric additives, and well-dispersed chromium oxide fillers. SEM, differential scanning calorimetry, and X-ray diffraction patterns demonstrated that chromium oxide increased the flexibility of the tetraethoxysiloxane-derived ceramer coating [49].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.5 Effect of Catalyst in Coating Formulations Methyltriethoxysilane and tetraethoxysilane were reacted in the presence of an acid catalyst to form a sol-gel coating [50]. An increase in the Si-CH3 bond concentration increased the overall thickness of the coating. Fourier Transform Infrared (FTIR) spectroscopy and ellipsometry established a correlationship between the methyl content and microstructure in the coating. Methyltrimetoxysilane and 3-GPTMS with 3-APTMS and N(2-aminoethyl)-3-aminopropyltrimethylsiloxane reacted in the presence of an acid catalyst to form a corrosion resistant coating. FTIR spectroscopy confirmed the molecular structure of the coatings. Atomic force microscopy showed the surface morphology did not reveal surface structures, suggesting a very dense packing arrangement within the molecular chains. Results from various analytical techniques showed that the optimum concentration of the hardener was the key factor in controlling the corrosion resistance over aluminum alloys [51]. Dibutyltindilaurate catalyst played an important role in the synthesis of the tetraethylorthosilicate sol-gel system. Various analytical techniques used on the gel and solid coating indicated that lauric acid derived from the hydrolysis of dibutyltindilaurate contributed significantly to modifying the condensation path of the Si moiety. Lauric acid acted as a catalyst for the hydrolysis and condensation of silicones. FTIR and small angle X-ray analysis indicated the possibility of hetro-condensation between Sn and Si moieties [52].

2.6 Factors Affecting the Coating Process The following section reviews some critical factors in the coating process. Aside from the synthesis of coating formulations, there are other factors that can affect the coating quality.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

8

Atul Tiwari and L. H. Hihara

2.6.1 Surface Preparation The adhesion of a coating is highly dependent on the condition of the substrate surface, and, therefore, surface cleaning and preparation play a vital role in the coating process. In aluminum alloys, for example, the formation of an Al2O3 layer is dependent on time and temperature (Fig. 2), and, hence, the surface condition would be dependent on prior processing and heat treatments. Since the surface bonding with coatings may be dependent on the oxide layer, surfaces generally should be etched or abrasive grit blasted to remove the existing oxide layer to obtain consistent results.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 2. Thickness growth of inert aluminum oxide layer over aluminum surface as a function of time and temperature. Adopted from Ref. [53] and reproduced after permission from Elsevier Publishing.

It should be noted that due to the high reactivity between aluminum and oxygen, a thin passive oxide layer will immediately reform on the aluminum surface after cleaning. The abrasive grit-blasting technique will also introduce surface roughness that may be beneficial to adhesion for some coatings.

2.6.2 Pretreatments Surface pretreatments are often used to enhance bonding. The interaction of silicones with inorganic metal surfaces depends on the nature and preparation of the surfaces prior to coating. For example, the amount of hydroxyl groups available over an aluminum surface for the reaction with coating functionalities depend on pretreatment conditions [54]. Some examples of surface pretreatments that have been used are summarized below. Adhesion was enhanced between aluminum and an epoxy coating through a porous pseudoboehmite oxyhydroxide layer (Fig. 3) that was formed by hydration of the aluminum substrate in boiling water [55]. A thin layer resulting from phosphate treatments promoted the adhesion of silicone coatings on aluminum [56]. In another case, silicone coatings were applied to a tacky epoxy primer layer to ensure the in-diffusion of silicones into the epoxy matrix that acted as a coupling agent between the silicone coating and metal substrate [57]. A few chromium-ion based primers such as strontium chromate were also used as a pretreatment [16, 58]. Review on surface treatment can be found in reference [59].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

9

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 3. TEM image of a cross section of epoxy-psuedoboehmite-aluminum layers. Thin hairs of a pseudoboehmite oxyhdroxide layer, which facilitate the strong bonding, are clearly visible. Image acquired from [55] and reproduced after permission from Elsevier Publications.

2.6.3 Method of Application Coatings can be applied by various techniques such as spraying or dipping, which can affect the coating thickness and hence its properties. Base-catalyzed particulate sols were used to prepare silica hybrid coatings. A crack-free hybrid silica coating results from either electrophoretic deposition or dip coating in a mixture of methyltrimethoxysilane and ethytriethoxysilane containing sodium hydroxide. Corrosion resistance improved when the coating thickness was more than two microns. The barrier properties of the coating obtained by eletrophoretic deposition were superior to that achieved with dipping [60]. 2.6.4 Drying and Curing The drying and curing processes involve a hydrolysis reaction followed by condensation and polymerization of monomers (Fig. 4). In the early stages, particles grow by nucleation followed by agglomeration and network formation that finally leads to gelling. Rapid drying of the gel at elevated temperatures may 1) introduce residual stresses in the coating network, resulting in cracking, or 2) lead to the formation of pores and cavities. Also, when curing at elevated temperatures, the mismatch in coefficient of thermal expansion between the substrate and coating may also lead to delamination. Slow heating or ambient temperature hardening is preferred to enable the gradual release of volatile organic components minimizing pore formation. Drying and hardening processes in coatings can be analyzed using spectroscopic techniques [61]. The effects of drying and curing temperature on GPTS-coated oxidized aluminum substrates was studied. Changes in the curing temperature induced significant changes in the final coating morphology. Coating thickness increased with the increase in temperature, possibly due to a higher crosslinking density introduced between the silanol groups. [62].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

10

Atul Tiwari and L. H. Hihara

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 4. Plot of coating viscosity as a function of time during the application and hardening process. Reproduced from Ref. [63] after permission from Marcel Decker Publications.

3. CHARACTERIZATION OF SILICONE COATINGS The development of novel ceramer coatings relies on extensive material characterization to gain a thorough understanding of coating synthesis, curing, and performance. Various analytical techniques used in research and development of silicone ceramer coatings are summarized below with examples from various studies. The information is not intended to be a comprehensive guide, but rather a collection of work showing how others have used the techniques in their research. 3.1 Fourier Transformation Infra-Red Spectroscopy FTIR can be used to study coating composition as well as to monitor the reaction process as a function of time. Several available FTIR techniques such as attenuated total reflection [64-66] for coatings hardened on a metal surface, diffuse reflectance [67] for ultra-thin films, and transmittance [68, 69] for clear or transparent coatings are useful. FTIR is useful for determining parameters such as residual porosity and molecular bonding. If used in concert with a microscope, excellent spatial resolution can be obtained. Rapid kinetics or in-situ observation of the curing reaction mechanism using timeresolved FTIR spectroscopy [70] can provide valuable information about the volume collapse

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

11

or hardening route in the complex ceramer network. The residual porosity Vp in the coating structure can be determined by FTIR analysis using the following equation [71]:

Vp  1   xd  A

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

while,

d

d

; and

A  xd . d  x. .

Here xd and d (1.0 x 104 cm-1for dense silica film) are the film thickness and the absorption coefficient of the dense film, respectively, while x and  are the film thickness and the absorption coefficient of a porous film; A is absorbance. The effect of polytetrahydrofuron in TEOS hybrid film was studied by FTIR analysis, which displayed a defect band at 560 cm-1 assigned to the skeletal vibration of 4-fold siloxane rings [72]. The intensity if this band was a function of the polymer content and molecular weight increment. Changes in the deconvoluted bands at 1080 cm-1 suggested that porosity of the hybrid film depends on the polymer content and average molecular weight. Polytetrahydrofuraon hinders the reactivity of the silanol groups and retains the 4-fold fold siloxane rings in the hybrid gel film that leads to porous structure compared to the pure silica film. FTIR was used to monitor the hydrolysis of phenylaminomethyl trimethoxysilane (PAMS) and variable concentrations of TEOS [73]. The band at 1167 cm-1 corresponding to SiOSi recorded progress of the hydrolysis reaction. Formation of H3O+ ion increased the hydrolysis of alkoxysilanes while OH- ion formation increased the condensation between different hydroxyl functionalities. The area ratio of bands at 600 cm-1 and 1070 cm-1 monitored the amount of cyclic species forming in the solution at a given time. The rate of hydrolysis increased in solutions containing pure TEOS compared to those having a mixture of TEOS and PAMS. An increase in the pH of the solution due to PAMS dissociation and steric hindrance from the bulky groups in the mixture lowered the rate of the hydrolysis reaction. Moreover, the formation of cyclic species increased in pure PAMS solution but decreased in the case of mixed silicones, probably due to the reaction of TEOS with PAMS before it could react each other to form cyclic. The change in the amount of cyclic species, polymerization rate, and functionalities of the precursor changed the gelation time of the precursors. FTIR spectra recorded on ceramer gel as a function of time is shown in Fig. 5. Peaks appearing at approximately 2900 cm-1 corresponding to symmetric and asymmetric –CH2 stretching [74] were found consistent and independent of time. Si-O-Si vibrations appeared at about 1080 cm-1 [75]. Spectral assignments indicated that the hydrocarbon portion of the composition did not take part during the initial phases of the reaction. Free and hydrogenbonded hydroxyl stretching appeared in the region of approximately 3500 cm-1 [74]. The intensity of the stretching decreased as the temperature increased, probably because the water molecules de-bonded from amine or hydroxyl groups in the coating. It is likely that these water molecules and those obtained from the condensation of silanol groups participated in the hydrolysis of the alkoxy groups in silanes. A possible kinetic pathway can be estimated by monitoring the time-dependency of the peak heights at 1350 cm-1 and 1950 cm-1, which

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

12

Atul Tiwari and L. H. Hihara

literature shows corresponds to hydroxyl/amines and Si-O bond stretching modes [76, 77], respectively. The drop in intensity of other peaks in the spectra could be attributed to various condensation and cross-linking reactions occurring simultaneously.

Figure 5. Ambient conditions FTIR analysis of silicone ceramer coating [78].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

3.2 Raman Spectroscopy Raman spectroscopy has been particularly useful in studying the chemical changes that occur during the conversion of solution to gel, gel to glass, and glass to ceramic [79-82]. If used in concert with a microscope, excellent spatial resolution can be obtained. The hydrolysis reaction of tetraethoxysilane (TEOS) that contained variable amounts of water molecules was recorded with Raman spectroscopy [83]. Bands at 650 cm-1 corresponding to TEOS, and at 600 cm-1 corresponding SiOSi were used to monitor the hydrolysis reaction. The intensity of the band at 650 cm-1 decreased with the increase in H2O/TEOS ratio in the mixture. Band intensity at 600 cm-1 increased with the reduction intensity of 650 cm-1 band, suggesting the hydrolysis and subsequent condensation of silicones. GPTS and APTS systems displayed a doublet at 643 cm-1 corresponding to SiO3 symmetric and anti-symmetric stretching vibrations of GPTS [84]. Intensity of these bands decreased with the increase in time of hydrolysis. Bond intensities were observed at 881cm-1 and 1035 cm-1 corresponding to CCO- stretching vibrations from ethyl alcohol and COvibrations from methanol suggesting the formation of alcohol during the hydrolysis of silane (Fig. 6). The hydrolysis reaction velocity decreased at elevated temperatures due to a decrease in the pH of the medium. The pH dropped due to the neutralization of acidic silanol groups. The insertion or presence of hydrocarbon between the silicon atoms could be investigated by looking into the peaks at 1350 cm-1 and 1030 cm-1 that appears due to the scissoring and wagging mode of –CH2- groups, respectively [85]. Aluminum coupons coated with a quasi-ceramic silicone coating were exposed for 4 months at Mauna Loa Observatory on the Big Island of Hawai‗i that receives high levels of solar radiation. In another experiment, a similar set of coated coupons were exposed for 60 hours in artificially-simulated sunlight. No visible damage was detected after the prolonged exposure periods. Raman spectra recorded before and after the exposures indicate that the coatings were unaffected after 60 hours of intense UV radiation in the lab, while coatings

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

13

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

exposed for 4 months at the outdoor site showed distinct changes in structural patterns. The structural changes in the coatings could be due to coating degradation or to extended reactions occurring between different chemical entities as a result of prolonged exposure to solar radiation [86].

Figure 6. Raman spectral assignments of the GPTS/APTS hybrid material. The spectra were recorded after 0 min. (dotted line), 15 min. (dashed line) and 14 hours (solid line) after the start of the sol-gel reaction. Plot acquired from ref [81] after permission from Springer Publication.

3.3 Nuclear Magnetic Resonance Spectroscopy The reaction mechanism between different chemical entities in a polymeric system can be effectively studied using Nuclear Magnetic Resonance Spectroscopy (NMR) [87-89]. Analysis can be conducted on the coating precursors as well as on the solidified coating. Organic silicones such as glycidoxypropyltrimethoxysilane tend to undergo a ringopening reaction in acidic or basic conditions to form diol and other hydroxyl compounds. Such reactions can be effectively monitored with NMR technique [90]. GPTMS-TEOS ormosil system displayed following 13C peaks in NMR analysis [91]. 8.09.4 ppm [Si-CH2], 22.0-23.4 ppm [Si-CH2CH2CH2], 73.2-74.2 ppm [Si-CH2CH2CH2-O-CH2], 71.4-72.0 ppm [Si-CH2CH2CH2-O-CH2], 51.2-52.3 ppm [Si-CH2CH2CH2-O-CH2CH[O]CH2], and 44.2-45.2 ppm [Si-CH2CH2CH2-O-CH2CH[O]CH2]. Similarly, following 29Si NMR peak assignments has been reported [92, 93]: -49.4 to 50.7 ppm [T1= R-Si(O)2(OH)], -54.6- to -59.9 ppm [T2=R-Si(OR)(OH)2], -65.1 to-68.8 ppm [T3=R-Si(OH)3], -92.8 to -94.0 ppm [Q2=Si(OSi)2(OH)2], -102.7 to -103.4 ppm [Q3=Si(OSi)3(OH)] and -111.1 to -113.0 ppm [Q4=Si(OSi)3].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

14

Atul Tiwari and L. H. Hihara

Moreover, 17O NMR signals have been reported (Table 1) for hydrolyzed pure dimethyldiethylsiloxane (DMDES), methyltriethoxysilane (MTES), and tetraethoxysilane (TEOS), as well as their reaction products [94]. Here D refers to bi-functional Me2SiO2 units, T to tri-functional units and Q to tetra-functional SiO4 units. Table 1:17O NMR assignments for different alkoxy systems. Reproduced [94] with permission from American Chemical Society Sample DMDES MTES TEOS DMDES/TEOS

DMDES/MTES

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

MTES/TEOS

Chemical shift (ppm)

Assignment

33.7 63.9, 67.5 29.7 56.9 15.3 31.0 15.0 47.5 64.0, 67.5 31.7 60.4 64.1 15.4 30.3 44.7 57.1

D-OH D-O-D T-OH T-O-T Q-OH Q-O-Q Q-OH D-O-Q D-O-D D-OH / T-OH D-O-T D-O-D Q-OH T-OH or Q-O-Q T-O-Q T-O-T

NMR analysis shows that water in the synthesis process is used for hydrolysis as well as an epoxy ring opening. However, when less water is used, it is preferably consumed in the hydrolysis process. Hydrolysis with large amount of water leads to a product that displayed poor corrosion inhibition probably due to the formation of a less well-connected network structure. Poor silicone-particle packing and the formation of channels could allow the percolation of corrosive media to the substrate. Moreover, the formation of hydroxyl-bearing groups due to the ring opening of epoxide group induces hydrophilicity in the coating structure, which attracts electrolyte into the coating, promoting corrosion. Coatings containing low hydrocarbon content that were synthesized with low water content during hydrolysis displayed better packing arrangement and barrier properties than those containing higher hydrocarbon content. Similarly, 29Si and 13C NMR studies showed that metal alkoxide such as titanium alkoxide play a vital role in alkoxysiliane formulations and enhanced the degree of condensation. A higher degree of condensation in titanium alkoxide and availability of secondary condensable entity enhances the reactivity in such sols. Titanium alkoxide also facilitates the ring‘s opening of an epoxy group. For example, in the case of GPTS and titanium ethoxide formulations, there was as much as 78% ring cleavage. However, an excess of water is required for ring cleavage and hydrolysis since the reactivity of titanium alkoxide with water is low. Ether-bond formation between titanium alkoxide and epoxide rings was only possible when a sufficient amount of titanium hydroxide was present in the sol. Preexisting ether linkages in the titanium sol was probably responsible for such anomalous behavior [90].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

15

3.4 X-Ray Photoelectron Spectroscopy XPS is a versatile tool and provides both elemental and chemical information. Surface survey scans, dedicated element scans, and elemental mapping can be done using the XPS technique [95, 96]. For example in Fig. 7, the deconvoluted Si2p signal reveals pure silicon at 99.69 eV, a small amount of Si2O at 100.64 eV, a Si2O3 moiety at 102.72 eV, and a weak SiO2 peak at 103.67 eV [97].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 7. Si2p de-convolution of high resolution XPS analysis of silicone. Adopted from Ref. [97].

XPS analysis was conducted on the solid ceramer coatings to confirm the reaction pathways adopted by precursor macromolecules. Fig. 8 shows a surface survey scan obtained on the 6061Al-T6 coupon treated with a silicone coating. The deconvolution of Si2p spectra shows only one peak at 102.63 eV, which is due to 10.6 at.% silicon. The C1s spectra show four distinct peaks. The peak at 285.96 eV corresponds to a long carbon chain having 30.7 at. %, while that at 286.51 eV represents carbon joined to nitrogen (2.9 at.%) in aminosilane. Another peak positioned at 288.13 eV corresponds to carbon attached to different oxygen atoms (17.0 at.%), while the peak at 289.28 eV corresponds to an ester-type linkage (COOR ~1.1 at.%) of carbon. Looking to N1s spectra indicates nitrogen joined to carbon in aminosilane at 400.22 eV and nitrogen bonded to aluminum metal at 402.39 eV. These findings suggest that the backbone chain of aminosilane was intact and amine functionality reacted with surfacial hydroxyl groups of aluminum metal resulting in permanent covalent bonding between the coating and metal substrate. The O1s deconvoluted spectra show three peaks. The peak at 530.71 eV could be explained by Si-O-Al linkage, while the strong peak at 532.56 eV was probably due to an oxygen atom involved in the Si-O-Si linkage. An oxygen atom attached to a metal, probably SnO2, explains the weak peak at 534.07 eV [23]. Surface contamination and the cause of coating delamination could be effectively studied using the XPS technique. Silicone adhesives when utilized with organic-film adhesion may undergo oxidative degradation. The degraded byproduct could remain on the metal surface or diffuse into the organic film leading to delamination [98]. In grit-blasted aluminum and

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

16

Atul Tiwari and L. H. Hihara

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

GPTS-treated aluminum, both coated with an epoxy-base adhesive, XPS showed that chemisorption that was covalent in nature was the mode of adhesion for both substrates. However, on the GPTS-coated aluminum surfaces, covalent bonds formed between the curing agent and epoxy rings from GPS or the epoxy resin, while on grit-blasted aluminum, covalent bonds formed between different components of the adhesive [99].

Figure 8. XPS analysis of silicone ceramer coating. Adopted from Ref. [23] and reproduced after permission from Elsevier Publications.

3.5 Secondary Ion Mass Spectroscopy Secondary Ion Mass Spectroscopy (SIMS) utilizes a high-energy primary ion beam to sputter and eject secondary ions from solid surfaces. The ejected ions are analyzed with a mass spectrometer. Elemental and chemical information can be obtained. SIMS also has excellent spatial resolution. Grit-blasted aluminum, as well as grit-blasted and GPTS pre-treated aluminum, were coated with a commercial epoxy adhesive. The reactions between the different coating components were analyzed using SIMS [99]. Several different hydrocarbon fragments were found at low mass; for example, C2H5+ (mass/charge (m/z)=29 daltons (Da)), C3H5+ (m/z=41 Da) and C3H7+ (m/z=43 Da). These ions were contributions from the substrate as well as GPTS coating. The fragments at m/z=28 Da were assigned to the presence of Si+ as well as a small contribution from C2H4+. Similarly, other fragments at m/z= 77, 91, 128, and 178 Da originated from the phenyl group in the epoxy resin. Ions with even mass numbers indicate nitrogen molecules with an odd number of nitrogen atoms as per ―Nitrogen Rule‖ suggested elsewhere [100]. It was realized that fragmentation from the cross-linked structure was

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

17

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

difficult to compare with the uncross-linked material. Also, a covalent bond was formed between the oxidized aluminum and a silanol group from hydrolyzed GPTS. The covalent bonds were formed between amine-silanol-epoxy groups when epoxy adhesive was applied over GPTS [99]. Permanent bonds of this nature are highly desirable as they are not susceptible to attack from corrosive media, including water. The SIMS technique was used to analyze reactions and interaction in a silicone ceramer coating hardened over 6061Al-T6 aluminum alloy. Although SIMS only samples the uppermost mono-layers, reactions between the coating and aluminum substrate could still be reflected in the spectra due to mixing and in-diffusion of reaction products during the curing process, when the coating was still in the liquid state. A positive ion static SIMS spectrum is shown in Fig. 9. The spectrum displays a majority of CH fragments from a long chain hydrocarbon present in the coating backbone. A NH2+ fragment from aminosilane is visible at m/z=16 Da, suggesting that there was still some aminosilane left in the coating that had not reacted and was presented as hydrogen-bonded moiety. Another peak at m/z=15 Da was from a CH3+ fragment from a hydrocarbon portion of the chain. An additive effect from Al+ and C2H3+ fragments could explain the high intensity of signal appearing at m/z=27 Da. Several different peaks such as Si+ (m/z = 28 Da), SiH+ (m/z = 29 Da), SiC+ (m/z = 41 Da), SiCH2+ (m/z = 42), SiCH3+ (m/z = 43 Da), SiO+ (m/z =44 Da), and SiOH+ (m/z =45 Da) appeared due to the fragmentation of silicones [101-103]. The signal at m/z = 56 Da corresponds to a SiOC+ fragment from methoxysilane, indicating the likelihood of the incomplete hydrolysis of the alkoxy group. The high intensity of the signal at m/z=18 Da (from water molecules) caused all other signals to appear relatively weak [23].

Figure 9. Positive ions SIMS spectra of silicone ceramer coating.

3.6 X-Ray Diffraction X-ray diffraction (XRD) is a technique that can be used to analyze crystal structure, chemical composition, and physical properties of coatings and their substrates. This technique utilizes the impingement of an X-ray beam of a known wavelength onto the sample under analysis. The X-rays are then diffracted by the crystal lattice giving a pattern of peaks of various intensity as a function of the diffraction angle. Materials with a crystalline structure

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

18

Atul Tiwari and L. H. Hihara

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

generate sharp peaks; whereas, amorphous materials may generate broad and ill-defined peaks. Results from XRD may also be used to extract various forms of information. The clear presence of pure silica (as SiO2) may not be easily detected in hybrid coatings due to confinement or encapsulation of it by organic polymer chains. Therefore, XRD patterns obtained on such ceramer materials display a single homogenous amorphous phase without any evidence of the silica filler. However, the absence of sharp peaks in the XRD pattern and the presence of a broad peak at 21° 2 value was evidence that amorphous silica was present with homogeneity in the sample, assuring good adhesion between inorganic and organic moieties [104]. XRD analysis also showed that the broad peak width (Fig. 10) centered at approximately 22o (2 value) changed with the organosilicone concentration. Increase of polyorganosilicone in the composition shifted the diffracted peak to a lower angle, indicating the increase in inter-atomic distance in silicone [105]. X-ray diffraction has been indirectly used to obtain other information. For example, the diffusion of ions from the metal substrate into the coating has been studied [106]. A silica coating was applied over a copper substrate, and the in-diffusion of atmospheric oxygen through the silica coating resulted in the formation of copper oxide at the silica-copper interface. Diffusion coefficient values have been calculated using this method. The D value of 10-19 cm2/s has been determined for silica glasses at 1000oC, and 10-3 cm2/s at room temperature for largely porous silica gel glasses [107]. Calculations using room-temperature diffusion coefficient values obtained from the XRD technique suggested that the metal oxide layer may start building at the coating-metal interface after approximately 10 years of the exposure to the atmosphere [106].

Figure 10 XRD pattern of 3-(methacryloxypropyl)-trimethoxysilane hybrid silica powder with different mole ratios of TEOS/MEMO in the sols after heat treatment at 150 oC for 5 hours showing an amorphous coating network. Reproduced after Ref. [105] with permission from Elsevier Publications.

4. THERMAL ANALYSIS The durability of barrier coatings can to some degree be related to the thermal stability of the coating. Polymer chains that resist breaking up into smaller fragments as a result of heat

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

19

exposure will likely have more stable long-term mechanical properties—averting or delaying failure caused for example by increasing brittleness. It is therefore advantageous to analyze the stability of the coating spanning a range of temperatures. Such studies can provide information on coating durability in aggressive climatic conditions [108]. The following subsections summarize research related to the thermal analysis of silicone ceramer materials. The information is not intended to be a comprehensive guide, but rather a collection of work showing how others have used thermal techniques in their research.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.1 Thermogravimetry Thermogravimetric analysis (TGA) records degradation (weight loss) of material as a function of time and temperature. The onset of the decomposition temperature from the TGA curve suggests the thermal stability of the material while the ultimate residue that is obtained after the complete pyrolysis reaction gives the percentage of volatile or decomposable constituents in the compounds under investigation. Derivative thermogram (DTG) showing the rate of mass change is constructed from the weight loss curve as a function of time and temperature. Observations from the DTG give the decomposition profile and percentage of constituents decomposing at that particular time and temperature. DTG of methyltriethoxysilane and TEOS-based hybrid material shows that the decomposition temperature of Si-CH3 increases with increasing tetraethoxysilane concentration. This suggests that a dense inorganic network hinders the out-diffusion of decomposed products and thereby increases the overall thermal stability of the material [109]. The organic-inorganic hybrid containing nitrogen, phosphorous, and silicon in epoxy resin was investigated. The hybrid system was compared to pure epoxy resin. The residue obtained after complete decomposition was higher in the hybrid materials due to the synergistic effect of nitrogen, phosphorous, and silicon in the system. Also, the activation energy from the decomposition of the hybrid material was higher than that of pure epoxy, proving that the hybrid material had higher thermal stability than that of pure epoxy [110]. Interaction between tetraethoxysilane and polyols was studied using thermal analysis, FTIR, and solid state NMR spectroscopy. Thermal analysis demonstrates that the polyols were bound to the silica matrix through chemical, as well as hydrogen bonding [111]. Thermal behavior of epoxy-polyester powder coating containing YiO2 and SiO2 particles was investigated. The TG/DTA-GC/MS technique revealed that benzene and phenol were formed as a result of thermal decomposition. Also, the thermal stability of the cured coating was higher than that of the resin. Reticulation of resin chains was responsible for the higher thermal stability of the coatings compared to their parent compounds [112]. Thermal stability and corrosion resistance of a hybrid coating prepared by reacting 3-GPTS and aminoterminated siloxane with tetraethoxysilane was determined. Spectroscopic techniques characterized the different forms of silicone in the coating structure. Deconvolution of the DTG curve showed a four-step decomposition pattern under a nitrogen atmosphere. The thermal stability of the 3-GPTS-TEOS hybrid and the activation energy required its decomposition were higher than that of the polydimethylsiloxane-TEOS hybrid. The finding suggests that the degradation of the aliphatic n-propyl segment was enhanced by the higher thermal conductivity of silica [113].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

20

Atul Tiwari and L. H. Hihara

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

TGA was conducted on an organic/inorganic hybrid of a SiO2-matrix composite prepared by a sol-gel process utilizing tetramethoxysilane and TEOS as a precursor. Thermal stability of the hybrid material was enhanced due to the presence of tetramethoxysilane. A pure silica network was detected due to the thermal degradation and evaporation of the incorporated organic groups above 500 oC [105]. The thermal degradation kinetics of the methacryloyl group containing poly(methylphenylsiloxane) was studied using TGA. Friedman, FlynnWall-Ozawa, Coats-Redfern, and Phadanis-Deshpande methods were used to study the kinetics of degradation. They show that the first degradation stage of the poly(methylphenylsiloxane) material followed a three-dimensional diffusion mechanism, while the second degradation stage followed a nucleation and growth mechanism [114]. Ceramer coatings can be pyrolyzed at variable heating rates. The gases evolved as a result of material degradation can be analyzed using FTIR [115]. For example, Fig. 11 show TGA and FTIR of a solidified silicone ceramer coating at a heating rate of 20 oC/min in an inert atmosphere. The DTG plot suggests that material decomposed in four steps and that the final residue was 53%. The results indicate that material consists of at least 47% decomposable constituents that may evolve during the lifetime of the coating. The FTIR analysis of the evolved gas suggests that either the hydrocarbon or lower members of silicones were evaporating during the early stages of heating. The regime from 3026 to 2992 cm-1 shows symmetric and asymmetric vibrations from hydrocarbons. Strong peaks in the region beyond 3500 cm-1 correspond to the cleavage and decomposition of species containing hydrogen-bonded hydroxyl and amine functionalities. Furthermore, the hydrocarbon portion of the coating decomposed during several different heating windows. A detailed study of silicone ceramer coating decomposition and its kinetics has been conducted by Tiwari and Hihara [116].

(a)

(b)

Figure 11. Weight loss of solid silicone ceramer coating as a function of time in inert atmosphere pyrolytic conditions displaying multiple steps decomposition processes (a). Evolved gas FTIR analysis of solid ceramer coating at 20 oC/min heating rate (b).

4.2 Differential Scanning Calorimetry This technique records the amount of heat flowing between a sample and a reference under controlled thermal conditions. The DSC curve reflects transitions that are related to

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

21

specific heat, the glass transition temperature, exothermic reaction peaks for crystallization or crosslinking reactions, narrow endothermic peaks for fusion or melting, broad endothermic peaks related to decomposition, and volatilization, as well as dissociation or increase/decrease in heat flow due to oxidation or thermal decomposition [108]. The thermal properties of a polysiloxane-poly(tetrafluroethylene) semi-interpenetrating coating system was investigated using differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) techniques. Thermal stability of the coating system increased with the increase in the poly(tetrafluroethylene) content. Each blend had a single glass transition temperature (Tg), suggesting that the blends were comprised of a single phase. Also, transitions corresponding to the semi-IPN network were observed. DSC can reveal distribution of components at the molecular level [117]. For example, silicone containing a tri-functional epoxy monomer was mixed with conventional epoxy resin and then analyzed using the DSC technique. A single glass transition temperature was observed indicating that all the components in the material were in a single homogeneous phase. Moreover, low Tg values were recorded due to an increased free volume in the composition that had a high amount of phenyl rings in the composition [118]. A PDMS-modified epoxy coating displayed two transitions in a DSC thermogram. A peak appearing at -46 oC corresponds to the melting of the PDMS segment while Tg was seen at -16 oC. The value of Tg was found dependent on the mobility of the polymer chains. Higher Tg values were observed for stiffer samples that restricted the mobility of the chains [119].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.3 Dynamic Mechanical Analysis Dynamic Mechanical Analysis (DMTA) is a useful technique for determining the viscoelastic properties of coating materials. This technique helps determine the glass transition temperature (Tg) that appears as an alpha relaxation as well as other relaxation processes (e.g., beta, gamma, and delta). The visco-elastic response of a material can be measured as a function of applied periodic load in the form of fixed frequency or resonant frequency. The oscillatory applied load could be in the form of tension, compression, flexural, or torsion that provides information on storage modulus, loss modulus, or damping values [120, 121]. The phase morphology of a silicone biocidal coating was determined using DMTA (Fig. 12). A single glass transition temperature was obtained from the Tan  (represents the damping property of the material and defined as the tangent of the phase angle and the ratio of loss modulus over storage modulus) vs. temperature curve of the coatings, suggesting that there was no phase separation in the material. A higher value of a well-defined rubbery plateau region was seen in the amine-cured coating containing cyclic siloxane compared to those having linear siloxanes. Two clear transitions were seen in polybutadiene-cured siloxane biocidal coatings, indicating the incompatibility of silicones with polyalkanes. The presence of long hydrocarbon chains in siloxane acted as a compatibilizer that induced miscibility with the polyalkanes. However, microphase separation occurred due to an incompatibility between the two polymers. The modulus as well as the glass transition temperature were found to vary as a function of chain length in all coating formulations [122]. In the case of the polysiloxane-polyurea coating, the storage modulus (a measure of elastic response of the material different from Young‘s modulus) was independent of temperature

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

22

Atul Tiwari and L. H. Hihara

before Tg and decreased after Tg was reached [123]. Tg was obtained from -relaxation, and the storage modulus was independent of the coupling-agent concentration. However, the cross-linking density increased with the coupling-agent concentration. Also, the cross-linking density was found dependent on the TEOS content. 10000

1000

100

10

1

0.1

0.01 -100

-80

-60

-40

-20

0

Temp (C)

20

40

60

80

100

(a) 1.2

1

0.8

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

0.6

0.4

0.2

0 -100

-80

-60

-40

-20

0

Temp (C)

20

40

60

80

100

(b) Figure 12. DMTA plot of coatings showing variation of storage modulus as a function of temperature (a). Tan  as a function of temperature (b). Reproduced after Ref. [122] with permission from Taylor & Francis Publications.

5.0 MECHANICAL PROPERTIES Mechanical properties such as modulus, hardness, tensile strength, compressive strength, etc. can affect the durability of coatings. Conventional testing techniques can be used to obtain mechanical properties from self-standing coatings, while advanced techniques are required to evaluate coatings that are bonded to substrates. Two important mechanical tests are discussed in the following subsections.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Novel Silicone Ceramer Coatings for Aluminum Protection

23

5.1 Peel Test The peel test is useful in evaluating the adhesive strength between a coating and substrate. A basic peel test involves peeling a flexible strip of thickness ―h‖ and width ―b‖ by applying a force ―F‖ at an angle ―‖ (Fig. 13). If residual stress is neglected and the peel arm is infinitely stiff, then the total energy per unit area ―G‖ dissipated in de-bonding the strip is given as follows: [124]

G

F (1  Cos  ) b

The true adhesive toughness ―GA‖ is given as:

GA  G  G p ( Ro )

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where ―Ro‖ is a local radius of curvature, and ―GP‖is the plastic bending. The value of ―GP‖ depends on the properties of the strip and conditions at the point of bonding.

Figure 13. Schematic of basic peel test employed in coating technology. Reproduced after Ref [124] with permission from Taylor & Francis Inc.

In one example, delamination occurred at the epoxy-silicone interface for an epoxy under-coated, duplex siloxane -butyl acrylate styrene silicone top coat system. Also, peeling took place due to the nucleation of cavities that grew until the coating failed. Moreover, it was discovered that the force required to peal the silicone-polymer duplex top coat from the epoxy undercoat decreased as the top and bond coats were made thicker [125]. An epoxy coating was peeled off of a polydimethylsiloxane-coated surface and a decohesion mechanism was recorded during the release of the epoxy film. Different peeling mechanisms were observed as a function of the thickness gradient and average coating thickness. For thin films with low thickness gradients, the peel mechanism was attributed to the formation of voids and their coalescence that started from the thinner section of the coating and proceeded to the thicker sections. In thicker coatings, peeling progressed radially-

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

24

Atul Tiwari and L. H. Hihara

inwards from the outer edges beginning in thinner sections and propagating to thicker sections. The formation of finger-like structures in certain regimes and peeling in thicker sections were also a mode of failure in some cases [126]. The Scotch® tape peel-off test was conducted on silanized 2024 aluminum, coated with a commercial silicone adhesive coating. XPS and FTIR analysis were utilized to study the silicone primer layer after the peel-off test. The primer layer remained integrated while failure occurred due to loss of adhesion between the commercial silicone adhesive and the silane primer [127]. Peel tests are very common in the automotive paint industry and are frequently used to analyze the strength of pressure-sensitive adhesives. Additives such as polysiloxanes and titanium dioxide are used in such adhesives and play an important role in formulation. Polysiloxanes are commonly used as an ingredient in clear coats and dramatically affect the performance of the coating system [128].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

5.2 Nano-indentation and Nano-scratch Tests Thin reactive coatings that are bonded to metal surfaces can be tested by using nanoindentation or nano-scratch techniques. Nano-indentation involves pressing an indenter of known geometry into the coating surface with a known load such that it penetrates through the coating layers. Hardness and modulus values of the coated materials can be determined by knowing the load and area under the indenter impression [129]. Especially for the case of thin coatings (e.g. z, the oxidation state of the metal). Silicon alcoxides, for which N=z, are thus not concerned by olation reactions. It must be noted that the overall reactions in Figure 4 does not deliver information about the reaction mechanism of each individual event. For example, it is usually proposed that hydrolysis proceeds through three successive steps: nucleophilic, addition-proton, and transfer-elimination. The reactivity and structure of the final oxide depend over the hydrolysis-polymerization parameters (nature of the alkoxide, solvent, pH, degree of hydrolysis, i.e., amount of water…) and /or the drying procedure. Some of the most important of these aspects will be emphasized below. M-OR + H2O

M-OH + ROH

M-OR + M-OH M-OH + M-OH H M-OH + M

O

M-OH + M

O

R

Hydrolysis

M-O-M + ROH

Alcoxolation

M-O-M + H2O

Oxolation

H M-O-M + ROH Olation

H H

H M-O-M + H2O

Figure 4. General reaction scheme in the course of sol-gel process of M(OR)x.. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

86

FX Perrin

2.2. Effect of the Hydrolysis Ratio Based on the partial-charge model, the effect of the hydrolysis ratio h=[H2O]/[M(OR)x] on the condensation process was investigated by Livage and co-workers [14]. The first two steps of hydrolysis of Ti(OPri)4 (h2) occur easily ((OPri)0). Protonation of OPri produces a more positively charged leaving group ((HOPri)>0) than protonation of OH ((H2O)>0).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.3. Acid and Base Catalysis The hydrolysis and condensation reactions of tetraalkoxy- and organoalkoxysilanes are acid- and base-catalyzed. It is generally believed that the acid-catalyzed hydrolysis and condensation mechanism involve the fast protonation of OR and silanol groups, respectively, followed by a slow attack of water and neutral silanol, respectively, on the intermediate. Electron- providing substituents that help stabilize the positively–charged transition states should increase the reaction (hydrolysis and condensation) rates. Thus, the decreasing electron providing power OR>OH>OSi will preferentially direct the condensation (as well as hydrolysis) towards the end rather than the middles of the chains (Figure 5), leading to an open network structure.

acid catalysis H+

basic catalysis SiOOR HO Si O OR

A

O O Si O OR

B

OR O Si O OR

C

OR O

Si

OR

OR

D

Figure 5. Typical partially-hydrolyzed polymer. The preferential site for condensation is indicated for acid and base-catalysis conditions. Based on the electronic effects of the substituents (see text), the basicity of alkoxy groups on the different sites decrease in the order D>A>C>B. In base-catalyzed conditions, the same electronic effects will preferentially direct the attack of silanolate in the middle of the chain.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Protective Coatings Based on Sol-Gel Chemistry: A Review

87

In contrast, the base-catalyzed hydrolysis and condensation mechanism involve the attack of an hydroxyl anion and silanolate anion, respectively, on the neutral silicon species. Electron- withdrawing substituents that help stabilize the negatively–charged transition states should increase the reaction (hydrolysis and condensation) rates. Thus, the decreasing electron withdrawing power OSi>OH>OR will preferentially direct the condensation (as well as hydrolysis) towards the middles rather than the end of the chains (Figure 5), leading to more compact, highly-branched species. The charge distribution according to the partial-charge model within a titanium oxo polymer suggests the same catalysis effects than with silicon alkoxides (Table 1). The values of (OR) reflect the fact that the ease of protonation will be descending order in Table 1, so chain end sites will be more reactive than chain centre in acid catalysis. Conversely, for base catalysed reactions, the highly nucleophilic Ti-O- species will attack the more positively charged metal. Based on values of (Ti), the order of reactivity will be ascending order in Table 1, so chain center sites will be more reactive. Table 1. Charge distribution in a titanium oxo polymer (from ref [14])

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Species Ti(OR)3OTi(OR)2(OH)OTi(OR)2(O-)2 Ti(OR)(O-)3

δ(OR) -0.08 -0.01 +0.04 +0.22

It must be reminded that the hydrolysis and condensation rates are other than inductive effects: steric effects, solvent interactions, and transition metal alkoxide also important factors. It is inappropriate in a attempt a comprehensive review, but the next section will illustrate interesting of these factors.

N(Ti) +0.68 +0.70 +0.71 +0.76

sensitive to factors oligomerization of text of this size to some of the most

2.4. Silicon Alkoxide Vs Transition Metal Alkoxide Transition metal alkoxides (M=Ti, Zr, Al…) are much more reactive toward nucleophilic attack than silicon alkoxides (hydrolysis rate for Ti(OR)4 is at least five orders of magnitude larger than for Si(OR)4 [15])for two reasons: (i) the higher electropositivity of transition elements as compared to silicon which leads to a much higher electrophilic character of the metal and (ii) the possibility of coordination expansion for most transition metals (N>z). These two points are respectively exemplified by the positive partial charge on metal M for some metal alkoxides and the maximum coordination N of the metal(Table 1). The practical consequences of this difference in reactivity are that metal alkoxides, as such, must be handled very carefully in a dry environment. Since temperature both activates hydrolysis and the condensation processes, the temperature can be lowered to have a better control of the sol-gel transition with metal alkoxides. Another consequence is that the hydrolysis of metal alkoxides does not necessarily require the addition of external water and can readily proceed with air humidity.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

88

FX Perrin Table 2. Positive partial charge, ,on M for some group IV alkoxides (z=4), and maximum coordination N Metal alkoxide Si(OEt)4 Ti(OEt)4 Zr(OEt)4 Ce(OPri)4

δ +0.32 +0.63 +0.65 +0.75

N 4 6 7 8

On the other hand, the hydrolysis of silicon alkoxides that is far less reactive is preferably performed by adding a stoichiometric or even an excess amount of water (nH2O/nSi2). Increasing temperature can be viewed as an alternative way to activate the hydrolysis and condensation processes of poorly reactive silicon alkoxides. We‘ll see below that chemical modification or the appropriate choice of solvent is a common way to control the hydrolysis and condensation of metal alkoxides. After its stabilization by chemical modification, the hydrolysis-condensation of metal alkoxide precursor is preferably performed with addition of external water, i.e. under the same conditions than for silicon alkoxides.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.5. Nature of Metal Alkoxide – Solvent Couple The full coordination of the metal being not satisfied in monomeric metal alkoxides, the metal atom tends to increase its coordination number by oligomerization. In non-polar solvents, coordination expansion occurs via alkoxy bridging (Figure 6).The degree of oligomerization increases with the atomic size of the metal within a given alkoxy group, while it decreases with the bulkiness of the alkoxy group because of steric hindrance. As a representative example of the effect of the nature of OR group, Ti(OPri)4 is monomeric while Ti(OEt)4 is essentially as a trimer (molecular complexity=2.9) in the crude state or when it is diluted in a non polar solvent such as benzene [16]. In polar solvents such as the parent alcohol, solvate formation, ie the addition of a solvent donor molecule, is also possible. The stability of such solvates increases with ,(N-z)and the Lewis basicity of the alcohol. Thus, the adducts M2(OPri)8(PriOH)2 (M = Ce, Zr, Hf [17]) were isolated, whereas analogous adducts Ti2(OR)8(ROH)2 are only stable when the Lewis acidity of the metal was increased by more electronegative substituents [18]. Both metal atoms in the adducts are octahedrally coordinated owing to both the formation of alkoxo bridges and coordination of an alcohol molecule (Figure 7). The coordinated alcohol is stabilized by hydrogen bonding with the neighboring alkoxo ligand. From Figure 8, it is apparent that Ti(OEt)4 exhibit a reduced molecular complexity when dissolved in the parent alcohol than in a non polar (that is to say, inert) solvent. Alkoxy bridges being more stable towards hydrolysis than solvate bonds, the kinetics and resulting structure of the oxide material can be controlled by appropriate choice of solvent. For instance, partial hydrolysis of Zr(OPrn)4 dissolved in n-propanol results in a precipitate due to the presence of solvate bonds and, consequently, rapid hydrolysis and the formation of highly-condensed products. Precipitation can be avoided and homogeneous gels are obtained when Zr(OPrn)4 is dissolved in a non polar solvent, such as cyclohexane, due to the presence of more stable alkoxy bridges [19].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Protective Coatings Based on Sol-Gel Chemistry: A Review

89

R O

M

2 M-OR

M O R

Figure 6. Oligomerization by alcolation.

H L RO

R O

M RO

OR OR M

O RO

OR L

H Figure 7. Schematic structure of M2(OR)8(LH)2 (M=Ti, Zr, Hf ; LH=ROH or RNH2) (from ref.[39]).

2[Ti3(OEt)12] + 6 EtOH

3 [Ti2(OEt)8,2 EtOH]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 8. Dissociation and solvation of trimeric Ti(OEt)4 in ethanol.

Apart from the extent of oligomerization, other factors that affect the hydrolysis and condensation kinetics are the size and electron-providing or –withdrawing of the organic ligands. For a series of titanium n-alkoxides, the hydrolysis rate decreases with the alkyl chain length. This result was explained in terms of steric hindrance and charge distribution with a trend of decreasing (Ti) and (H) with alkyl chain length. However, the hydrolysis rate for isomeric titanium butoxides deceases as tertiary > secondary > normal [20]. The latter result underlines the fact that steric effects must not be overestimated to explain the hydrolysis rate of metal alkoxides. Inductive +I effects of the alkyl chains, and, even more importantly, the molecular structure of metal alkoxide(degree of association and nature of metal-ligand bonds) should also probably be taken into account.

2.6. Chemical Modification of Metal Alkoxides We have seen before that metal alkoxides are generally very reactive species. Their hydrolysis and condensation reactions can be controlled via chemical modification using chelating ligands such as glycols, organic acids,  diketones, amines, etc. The common feature of such chemical additives is that they are protic Lewis bases, LB-H, that are thus able to substitute one or more OR groups of M(OR)n with concomitant ROH elimination to give a

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

90

FX Perrin

new precursor M(OR)n-x(LB)x. These chelating ligands effectively occupy some of the coordination sites of the metal alkoxide, thereby lowering the rate and extent of hydrolysis. Besides, the replacement of one or more OR groups with LB lowers the connectivity of the molecular building blocks. This results in less highly condensed products and promotes gelation instead of precipitation. Acetylacteonate-, carboxylate-, and nitrogen-containing ligands will be considered as representative chemical modifiers of metal alkoxide in the following discussion. Acetylacetone is a strong chelating ligand that has been frequently used as a stabilizing agent for metal alkoxide precursors : W(OEt)6[21],Al(OBus)3 [22], Zr(OPri)4 [23] Ti(OPri)4 [24], Ti(OBun)4 [25]. An illustrative example is the adduct Ti(OPri)3acac prepared by addition of equimolar quantities of acetylacetone (acacH) to titanium tetraisopropoxide [24,26]. Titanium coordination turns from 5 to 6 as soon as water is added to Ti(OPri)3acac. NMR and FTIR spectroscopy indicate that OPri groups hydrolysed first and that all acac ligands can not be removed even when a large excess of water (H2 -diketones and -keto esters can be used instead of acacH. Some representative examples are given in Figure 9. O

O H2C

H2C

O

O

acetylacetone (acac-H)

2,2,6,6-tetramethylheptane-3,5-dione(tmhd-H)

O

O

HC

HC

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

O

O

3-allyl-2,4-pentanedione (apd-H)

3-acetylpentane-2-one (acp-H)

(MeO)3Si

O

O H2C

HC

O

O EtO 3-acetyl-6-trimethoxysilyl-hexane-2-one (ats-H)

ethylacetoacetate (eaa-H)

O H2C O O H2C

O O

O

O O allylacetoacetate (aaa-H)

methacryloxyethyl-acetoacetate (meaa-H)

Figure 9. diketones and -keto esters that can be used for the chemical modification of M(OR)x.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Protective Coatings Based on Sol-Gel Chemistry: A Review

91

Two types of crystal structures have been isolated from reactions of titanium alkoxides with -diketones and -keto esters : [{Ti(OR)3(dik)}2] (dik = acac, R = Me, Et, Pti; dik = tmhd, R = Me, Prn, Pri) are obtained from reactions between Ti(OR)4 and one equivalent of dikH [27]. In the solid state, these compounds adopt centrosymmetric, alkoxide-bridged binuclear structures (Figure 10). R'

RO O

R'

OR R O Ti O O R

R'

O O Ti OR OR

R' Figure 10. The structure of dimeric Ti(OR)3(-diketonate).

By contrast, reaction of Ti(OR)4 with -diketones or -keto esters in a 1:2 molar ratio resulted in the isolation of monomeric Ti(OR)2(dik) 2 with octahedrally-coordinated titanium atoms and cis OR groups [28-30] (Figure 11).

R' O

R' Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

O

O

R'

Ti O R'

OR OR

Figure 11. The structure of monomeric Ti(OR)2(diketonate)2.

It should be noted at this point that the situation is more complex in solution: [{Ti(OR)3(dik)}2] undergo ligand redistribution to give mixtures of [{Ti(OR)3(dik)}2], [Ti(OR)2 (dik)2] and Ti(OR)4. When R = Pri and dik=tmhd or acac, the equilibrium favours the mononuclear compounds, whereas when R = Me, Et or Prn (ie with primary alkoxides), the dimeric species are present in solution [26]. Another problem is related to the fact that the substitution reaction does not go to completion. The maximum complexation degree (average number of -diketonate ligands, L, per metal atom, M) was determined for the reaction of Ti(OBun)4,Zr(OBun)4 and Al(OBus)3 with a 3-fold excess of the -diketones acetylacetone (acacH), 3-allyl-2,4-pentanedione (apd-H), and the -ketoesters ethyl-acaetoacetate (eaa-H), allylacetoacetate (aaa-H) and methacryloxyacetoacetate (meaa-H) [31]. A complexation degree of 2 was found only with acacH. In all other cases, 1.5L:M 1, the substitution reaction is no longer quantitative (the di-substituted derivative Ti(OBu)2(OOCMe)2 has never been observed) and a substantial amount of free acetic acid remains which can react with the cleaved alcohol as discussed above. An ATR analysis of the reaction of Ti(OBu)4 with acetic acid showed that a maximum degree of complexation carboxylate ligands per titanium atom) of 1.85 can be reached when a large excess of acid (>8 equivalents) is employed [34]. This agrees well with the structurallycharacterized cluster Ti9O8(OPr)4(methacrylate)16, the compound with the highest average number of carboxylate ligands per Ti atom (1.78) known at present[35]. It is noteworthy that the formation of oxo-clusters has been observed in many cases even when a 1:1 molar ratio of Ti(OR)4 and carboxylic acid is employed. There are at least three explanations for this observation (i) reaction Ti(OR)4 + R‘COOH  Ti(OR)3(OOCR‘) + ROH is a true equilibrium reaction ,(ii) the low solubility of the oxo-cluster shift the substitution equilibrium and, (iii) oxo ligands are formed by non-hydrolytic routes, Ti-OR + Ti-OOCR‘  Ti-O-Ti + R‘COOR (1) and/or Ti-OR + Ti-OR  Ti-O-Ti + ROR (2). Route(1) has been proved to occur by observing structural changes of structurally-defined carboxylate-substituted oxo-clusters (see refs [36] and [37] as two representative examples). The carboxylic acid: metal alkoxide is definitely the key parameter that determine the kind of cluster that is formed in a given system: with titanium [38] and zirconium [39] alkoxides, a medium carboxylic acid: metal alkoxide ratio results in the most compact structures (lowest M:O ratio, ie, highest degree of condensation). Indeed, at low ratios, not enough water is produced to enable a high degree of condensation while at high ratios, the complexation ratio (R‘COO:M) is high and the bidentate carboxylate ligands block coordination sites necessary for a high degree of condensation. The reaction of alkoxides other than group IV alkoxides (M= Ti, Zr, Hf, etc) with carboxylic acid illustrate the difficulty to foresee the kind of cluster that is formed. Contrary to Ti and Zr alkoxides, no oxo cluster is formed when Y(OCH2CH2OMe)3 is reacted with methacrylic or acetic acid. The OR are completely substituted with carboxylate groups and crystalline Y(methacrylate)3 and Y(acetate)3.0.5H2O are obtained, respectively [40]. The interaction of nitrogen-containing ligands with alkoxides has been investigated by Schubert and coworkers [41] (and references herein). A series of amine adducts M2(OPri)8(NH2R)2 (M=Ti or Zr) with the same overall structure as the alcohol adducts M2(OR)8(ROH)2 (Figure 7) has been isolated. The proton on nitrogen undergoes hydrogen bonding with the neighbouring alkoxo ligand but it is not transferred to OR because HNR2 group is less acidic than the HOR group. The stability of the amine adducts does not correlate with the basicity of the employed amine. The hydrogen donor ability of the amine is found to decrease as the basicity increases. Thus, it is suggested that the stability of the amine adducts is a balance between the strength of the M-N donor-acceptor bond and the N-H…O hydrogen bond between the N-H group and the alkoxo ligand [40]. Steric factors also hinder the formation of the NH…O hydrogen bond. An illustrative example of the steric effects is that of Zr2(OPri)8(NH2CH2CH2NHMe)2 where only the NH2 group is coordinated to Zr, but not the NH group. When diamines with two terminal NH2 are reacted with Ti(OPri)4, coordination polymers were obtained. The general structure is the same as that shown in Figure 7 with each NH2 group of the diamine coordinated to another Ti2(OPri)8 unit.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

94

FX Perrin

The deliberate formation of amine adducts for the processing of protective sol-gel coatings has been seldom reported in the literature. However, the interaction between amino groups and metal alkoxides must be taken into account in all formulations with amine functionalities. Two particularly illustrative examples are the use of mixed alkoxides, aminopropyltriethoxysilane (-APS) and metal alkoxide or the metal alkoxide/epoxy hybrids incorporating amine compounds as a hardener. In both cases, the formation of amine adducts will, at least partly, ―neutralize‖ the amino group. This has two consequences: in the first case, amino groups of -APS will no longer exhibit any catalytic activity (base catalysis). In the second case, less amine will be available to react with epoxy which will have an impact on the network structure. Unfortunately, the consequences of the formation of amine adducts are still largely ignored.

2.7. Transesterification Reactions

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Transesterification reactions are metathesis reactions that are typically catalyzed by Brnsted acids and bases [42]. Lewis acids such as metal alkoxides were found to be efficient catalysts or reactants for these reactions, notably in the case of the transesterification of an ester with an alcohol [43,44,45] (Figure 13, rxn (1)), the transesterification of a silicon alkoxide with an alcohol (Figure 13, rxn (2)) [46], the cross-transesterification of two silicon alkoxides [46] (Figure 13, rxn (3)), the cross-transesterification of a silicon alkoxide and a transition metal alkoxide [47] (Figure 13, rxn (4)), and the cross-transesterification of an ester and a transition metal alkoxide(Figure 13, rxn (5). The latter reaction (Rxn(5)) certainly explains the formation of the polynuclear alkoxide Ti5(OPri)9(µ-OPri)(OC2H4O)5 in toluene at room temperature from Ti(OPri)4 and 2-hydroxyethylmethacrylate [48]. (OR ) x R1COOR‘ + R‖OH M    R1COOR‖ + R‘OH

(1)

(OR ) x Si(OR‘)4 + R‖OH M    Si(OR‘)4-y(OR‖)y

(2)

(OR ) x Si(OR‘)4 + Si(OR‖)4 M    Si(OR‘)4-y(OR‖)y

(3)

Si(OR‘)4 + M(OR)x   Si(OR‘)4-z(OR)z + M(OR)x-y(OR‘)y

(4)

R1COOR‘ + M(OR)x   R1COOR + M(OR)x-1(OR‘)

(5)

Figure 13. Transesterification reactions where the transition metal alkoxide M(OR)x participate as a catalyst (rxns (1), (2) and (3)) or as a reactant (rxns (4) and (5)); y,z=0,1,2,3,4.

Figure 14 depicts the two types of activated complexes that are thought to be generated during the transesterification of silicon alkoxides (Rxns (2) to (4)) in presence of a Brnsted acid or a transition metal alkoxide [46]. The latter compound behaves as a Lewis acid species due to the empty d orbitals on the metal. The differences in reactivity between The transition metal alkoxides Ti(OPri)4 and OV(OPri)3 are shown to catalyze the transesterification of

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Protective Coatings Based on Sol-Gel Chemistry: A Review

95

tetramethyl and tetraethyl orthosilicates (Rxns(2)and (3)). The better efficiency of titanium with respect to that of vanadium was explained in terms of a mechanism involving ligand migration between the metal and the silicon (Figure 15). The presence of the short terminal oxygen on the vanadium affords a region of low steric congestion that would favour a coordination geometry of silicon alkoxide away from the alkoxides bonded to the metal and, hence, not favorable for ligand migration (Figure 15a). By contrast, the titanium complexes would not strongly favor any particular orientation and the alkoxy groups in the activated complex would have the alkoxy group in close proximity for metathesis (Figure 15b).

H+

H+ (R'O)3Si

O R'

Si(OR')4

M(OR)x M(OR)x

(R'O)3Si

O R'

Figure 14. Substitutionall- labile complexes formed by interaction of Si(OR‘)4 with a Brnsted acid or a Lewis acid (M(OR)x).

OR'

R'O

R'O

OR'

Pri

Si O Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

OR' Si

O O

V

O R'

O Pri

OR'

Ti

O O

O Pri

Pri (a)

Pri

R' O

O

Pri Pri (b)

Figure 15. Alkoxide exchange between the metal and the silicon in the activated complex in (a) titanium tetraisopropoxide and (b) oxovanadium triisopropoxide (from ref [46]).

High temperatures for several hours are required for the reaction (1) to go to completion [43-45] and Lewis acid species are less efficient catalysts for Rxns (2) and (3) than equivalent amounts of strong Brnsted acids [46]. Consequently, transesterification reactions (1), (2), and (3) would generally appear as side reactions of the sol-gel process. Despite this, the consequences of such reactions should not be discarded. By changing the alkoxide group on the silicon and/or metal center, Rxns (2) to (5) likely alter the kinetics of hydrolysis and condensation and, hence, the overall properties of the resultant oxide. In particular, the transesterification reactions are significant in mixed alkoxide systems or when the solvent alcohol differs from the alkoxy group on the silicon or metal element. This is the opinion of the authors that the possibility of transesterification reaction deserves more consideration than it has received in the past.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

96

FX Perrin

3. COATING TECHNOLOGIES 3.1. Fundamentals of Film Formation Sol-gel films are usually formed by three alternative wet coating techniques: dip-coating [49], spin-coating [50], and spraying [51]. Compared to CVD, evaporation, or sputtering methods, these methods are simple to apply and require less expensive equipment. In the Newtonian regime, the coating thickness, h, attained by dip coating can be calculated by the Landau and Levich equation (Eq1) [52] : h = 0.94

(U) 2 / 3 / 6 (g)1/ 2 1LV

(1)

where  is the sol viscosity, U the substrate speed, LV the liquid vapour surface tension, g the gravity and  the density. According to the model of spin coating by Meyerhofer [53], the final thickness of a spin coated layer is related to the processing and material parameter by the semi-empirical formula shown in Eq2.

 3  .m  h= 1 A / A 0 .   2 A 0  ² 

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1/ 3

(2)

where A is the mass of volatile solvent per unit volume, A0 is its initial value, m is the evaporation rate and  is the angular velocity. The fundamentals of sol-gel thin film formations have been thoroughly discussed by Brinker et al. [54]. The essential aspects of sol-gel film formation are (i) the overlap of the deposition and evaporation stages that result in a competition between evaporation (which compacts the structure) and continuing condensation reactions (which stiffen the structure) (ii) the time scale of aggregation, gelation and drying is considerable less than for bulk systems (seconds to minutes for films rather than days or weeks for bulk systems). Crosslinking reactions that occur during aging of a monolith (the term aging referring here to the process of change in structure after gelation) result in a greater resistance to compaction for the structure during the later stages of drying, and, consequently, to less compact structures than dried films. The relative rates of evaporation and condensation during film deposition can be adjusted by varying the rate of air flow over the surface or the pH of the coating bath, respectively. A representative example is given by Brinker et al. [55]: films prepared from strongly branched and fractal precursors at pH 3.2 have a 30% porosity while a reduction of the deposition pH by addition of 2M HCl reduces the condensation rate, causing the porosity to decrease (3% at pH 1.2). The aging of sols were found to have significant influence on the corrosion protection of the sol-gel coatings: aging would result in further condensation reaction and consequently the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Protective Coatings Based on Sol-Gel Chemistry: A Review

97

formation of a stronger network, which would result in the formation of a more porous structure [55].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

3.2. Sol-Gel Films via Electrochemical Approaches Spin-coating and dip-coating have two main drawbacks: first, they can not be applied to surfaces with complex geometries and, second, there is no selectivity between conducting and insulating parts of a surface. The formation of sol gel films by electrochemical methods is a simple way to circumvent these two limitations. The electrodeposition of metal oxides has been reviewed by Zhitomirsky [56]. In what follows, we will only remind the different deposition mechanisms that have been developed for electrodeposition of metal oxides. The reader interested in more fundamental aspects about the various types of interparticle forces that govern colloidal stability or particle coagulation are invited to read the review by Zhitomirsky. For the electrochemical deposition of metal oxides, two processes are distinguished: electrophoretic and electrolytic deposition. Electrophoretic deposition (EPD) is accomplished via motion of charged particles towards an electrode under an applied electric field. Electrophoretic deposition of particulate sol-gel sols has been reported by several authors [57-60]. The particle sols are generally prepared under basic catalysis [57-59]. The isoelectric point of silica in water typically lies at ca. pH 2. Then, the particles dispersed in neutral and basic pH conditions are negatively charged at the surface and are thus prone to migrate towards the anode. The limiting utility of anodic EPD is due to the possible concomitant oxidation of the substrate to be coated. However, it is reported that crack-free coatings as thick as 12 µm can be produced by anodic EPD on stainless steel after sintering in air at 500°C with short deposition times (5 min) [57]. No iron deposition was detected from the substrate to the coating. That was related to the fast EPD kinetics that inhibits the substrate corrosion for the applied current densities. Cathodic EPD is usually preferred to anodic deposition since, in these conditions, the metal corrosion is avoided. However, H2 bubbling can affect the homogeneity of the cathodic EPD coating [59]. Cathodic EPD was performed from suspensions of colloidal silica in acid catalysed sols [59] or by surface modification of silica particles with 3aminopropytriethoxysilane (APS) [60]. In the case of APS modified silica particles, the isoelectric point of silica particles is shifted to pH 6.5 [60]. Therefore, electrophoresis towards the cathode is effective in the range of pH smaller than 6.5. Electrolytic deposition (ELD) leads to a direct precipitation of metal oxides from solutions of metal salts as a result of Faradaic process occurring on the electrode surface. Various ELD strategies have been developed. Cathodic ELD is based on the reduction of the metal cation which cause their deposition [61] (Eq.3) or by driving a proton-dependent reducing process, such as water (Eq.4) or nitrate (Eq.5) reduction (see the deposition of niobium oxide [62] and Ni(OH)2 [63] as two representative examples of water and nitrate reduction, respectively), leading to an increase of the pH on the electrode surface and the subsequent metal hydroxide deposition (Eq.6). Mn+(aq) + (n-2x)e- + x H2O(l)  MOx(s) + 2x H+(aq)

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(3)

FX Perrin H2O(l) + e-  ½ H2(g) + OH-(aq)

(4)

NO3- + 6 H2O + 8 e-  NH3 + 9 OH-

(5)

Mn+(aq) + n OH-(aq)  M(OH)(s)

(6)

2 H2O  O2 + 4 H+ + 4e-

(7)

Anodic ELD has been less frequently used than cathodic ELD. A typical strategy for anodic ELD is based on the oxidation of metal ions at a higher oxidation state, which usually results in the precipitation of the corresponding hydroxide salt due to its lower solubility [64]. The approach developed by Shacham et al [65] is very near from the ELD method based a proton-dependent reducing process except that organometallic species (metal alkoxides) are involved rather than inorganic species: in the electrodeposition method by Shacham, the hydroxyl ions that are generated near the electrode surface act as a catalyst in the condensation process and results in a controllable sol-gel film deposition. This is illustrated in Figure 16 which shows the typical pH-dependence of hydrolysis and condensation rates for a silicon alkoxide precursor. The bulk solution is at an acidic pH (3-4) where the hydrolysis rate is high and the condensation rate is low. The generation of OH- ions at the surface of the electrode increases the pH at the surface (> 8.2) which enhance the condensation rate by a factor of about 105 as compared to the acidic medium. The thickness of the film depends on the applied potential, the time of applying the potential, and the nature of the electrode, i.e., its over-potential for hydrogen reduction. For example, by applying –1.3 V for 30 min in a water-ethanol-KNO3 solution of methyltrimethoxysilane (phthalate buffer, pH=3.5), the deposited film was 3-fold thicker on gold (~10 µm) than on indium-tin oxide (~3.5 µm). This result is in line with an electrogenerated-base mechanism which predicts that a thicker film is expected from electrodes possessing a lower overvoltage for hydrogen evolution. bulk-solution pH

pH at the electrode-solution interface

8

Log k

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

98

condensation

6

hydrolysis

4 2

4

8

12

16

pH

Figure 16. A generalized representation of the relation between the rate of hydrolysis and condensation of an alkoxysilane precursor and pH. The typical bulk-solution pH and pH at the electrode-solution interface during electrodeposition of a sol-gel film are shown. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Protective Coatings Based on Sol-Gel Chemistry: A Review

99

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

It is to be noted that the electrochemical generation of hydronium ions under anodic polarisation (Eq 7) results in a loss of pH near the electrode surface. These hydronium ions are said to act as a catalyst (acid catalyst) for condensation reaction [66]. However, there is no report to date about the electrodeposition of silica-based films via the application of positive potentials. The only exception concerns the electrodeposition of Ludox colloidal silica on the surface of glassy carbon electrodes [67]. Ludox colloidal silica consists of uniform discrete particles of silica with diameters ranging from 7 to 22 nm. These discrete particles are stabilized due to the high pH (9-10) of the medium. Upon application of a sufficiently positive potential, hydronium ions are generated at the electrode surface that destabilizes the sol (by neutralizing the surface SiO- on the particles) producing a porous silicate film on the surface. Therefore, the anodic electrodeposition process is not related to an enhancement of condensation rate by acid catalyst but to the destabilization of the particulate sol. The porous silicate film can be used as ion-exchange cations. Few studies concern the electrodeposition of titania [68] and zirconia [69] films, respectively, from titanium and zirconium alkoxide sols. Contrary to silica based films, ZrO2 and TiO2 films (50-600 nm and 20-1000 nm thick, respectively) can be obtained under both oxidative (acidic) and reductive (basic) conditions. ZrO2 films were formed in the presence of minute quantities of water (water/monomer molar ratios in the range of 10-5 to 10-1). Although protons (Eq.(8)) and hydroxide (Eqs (9-10)) ions are generated electrochemically even in the absence of water, water is crucial for the deposition of zirconia film. It means that water is involved not only in the electrochemical reaction but also in the condensation itself. The role played by water is primarily to provide the nucleophile, which attacks zirconium under acidic (Eq.(11)), and basic (Eq.(12)) conditions. Me2CHOH  Me2CO + 2H+ + 2e-

(8)

Me2CHOH + e-  Me2CHO- + ½ H2

(9)

Me2CHO- + H2O  Me2CHOH + OH-

(10)

[Zr(OPr)3(PrOH)]+ + H2O  [Zr(OPr)3(OH2)]+ + PrOH

(11)

[Zr(OPr)4] + OH-  [Zr(OPr)3OH] + PrO-

(12)

Despite the high reactivity of the precursor (Zr(OPr)4) with water, a small fraction water remains as a free reagent in the system because the large excess of the 2-propanol solvent quenches the hydrolysis equilibrium (Eq. 13) according to Le Châtelier‘s principle. Zr(OPr)4 + x H2O  Zr(OPr)4-x(OH)x + x PrOH

(13)

It is important to note that the thickness of ZrO2 and TiO2 films obtained by the electrochemical approach is limited to the sub-micrometer range. The authors suggest that the process is self-controlled and ceases as the pH at the film/electrolyte interface becomes insufficient to drive film deposition. Moreover, the fact that the amount of water must remain low (in the 0.1 mM range) for having a successful deposition underlines the complex effect of water concentration on film thickness. These effects are still unresolved.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

100

FX Perrin

To date, the studies that focus the corrosion protection of sol-gel silica films based on the electrochemical generation of the condensation catalyst are scarce. Sheffer et al. [66] prepared sol-gel films from tetraethoxysilane, methyltrimethoxysilane and phenyltrimethoxysilane on aluminium electrodes via the electrogenerated-base mechanism. The more hydrophobic film, which was based on phenyltrimethoxysilane showed the best corrosion protection in a 0.15 M NaCl solution. The advantages of the electrochemical method compared to the spin coating and dip coating methods are the followings: (1) the film only forms on conducting surfaces and not adjacent insulating surfaces, (2) the electrochemical approach allows depositing films onto complex geometries, (3) the concentration of the metal alkoxide in the batch can be as low as 1 mM (4) the thickness and the rate of formation can be controlled by the applied potential, the time of applying the potential and the nature of the electrode, and (5) specific reagents (potentially, corrosion inhibitors) can be entrapped into the film by adding them to the sol prior to electrodeposition [70]. In the electrochemical approach, gelation and evaporation are separated in time: the films are potentially porous and rough which can be a serious drawback when applied as corrosion protectors. Stirring the sol during electrodeposition appeared necessary in order to maintain film homogeneity. Without stirring, the surface of the films becomes significantly rougher and thicker [65].

4. SOL-GEL COATINGS FOR CORROSION PROTECTION

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.1. Inorganic Sol-Gel Coatings The development of inorganic sol-gel coatings for corrosion protection has been driven by environmental and technical aspects: sol-gel coatings are non-carcinogenic, environmentally safe, stable, and strongly adherent on metal surfaces. The strong adherence to metal substrates has been often related to the formation of covalent M-O-M‘ bonds (M = substrate and M‘=Si, Ti, Zr…, Eq.14) that are produced during the drying stage. -M-OH + HO-M‘  -M-O-M‘ + H2O

(14)

However, very few experimental evidence for the formation of M-O-M‘ bonds is presented in the scientific literature. In this respect, one of the most relevant studies was conducted by Gettings and Kinloch [71]. These authors detected an ion of mass 100, assigned as FeSiO+, in the SIMS spectrum of a steel surface treated with a 1% aqueous solution of glycidyloxypropyltrimethoxysilane that is a strong evidence of the formation of a chemical bond, probably –Fe-O-Si- between the metal oxide and polysiloxane. Table 3 gives few representative examples of inorganic sol-gel coatings. For inorganic sol-gel coatings, the corrosion behaviour is largely controlled by the degree of cracking and porosity in the coatings. This is exemplified by the inverse drying temperature- dependence of the corrosion rate of inorganic sol-gel coatings [72-74,76]. Sintering at high temperatures induce cracks that probably result from coating-substrate thermal expansion difference induced tensile stresses on heating beyond the sol gellation

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Protective Coatings Based on Sol-Gel Chemistry: A Review

101

stage. For example, phosphate-bonded sol-gel composite alumina coatings on stainless steal substrates were prepared using three different processing temperatures: 300, 400 and 500°C [72]. Although no correlation could be found between processing temperature and coating hardness, micro-cracking and corrosion were minimum when the lowest processing temperature was used. Due to the thermal expansion mismatch effect, inorganic sol-gel coatings are free of micro-cracks if they are kept below a certain critical thickness (typically < 1 µm). The thermal expansion of aluminium is twice as high as that of steel. Therefore, the critical coating thickness is much lower for aluminium than for steel substrates [77]. Due to their high crack-forming ability, inorganic sol-gel coatings can not provide adequate corrosion protection. The most promising applications of such coatings are their uses as thin adherence promoter layers for a subsequent organic coating or paint.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table 3. Examples of inorganic sol-gel coatings for corrosion protection

Substrate

Molecular precursor

Catalyst/ Chelating agent

steel

Zr(OBu)4

Acetylaceto ne or acetic acid

steel

Zr(OPr)4

acetylaceton e

stainless steel

Zr(OPri)4 Ti(OEt)4+Si (OEt)4

acetic acid

nH2O/nM

Thickness/ µm

Drying conditions (time in min)

air dried (70% RH)

0.3-0.6

150-200°C (30)

2

0.2

400°C (15)+600°C (60) (argon atm.)

>4

0.4-0.6

800 °C (120)

Corrosive environment Salt spray Na2SO4 (0.3%) thermal oxidation (450-550 °C) H2SO4 (15%) NaCl(3%)

Ref.

[73]

[74]

[75]

4.2. Organic-Inorganic Hybrids The brittleness of inorganic sol-gel coatings can be mitigated by the incorporation of an organic component into the dried film structure. The organic component makes the gel network more flexible and, thus, less prone to cracking during heat treatment of the film. The chemical compatibility between the organic polymer and the inorganic component is an essential requirement to obtain a nanometric dispersion of the inorganic particles into the polymer. If this condition is satisfied, organic-inorganic hybrid materials with tailored properties can be obtained [78-80]. The morphology and properties of O-I hybrid-materials can be tuned by changing the experimental framework related either to the building of the inorganic network (nature of the alkoxide precursor [1], chelating stabilizer [1], alkoxides bearing reactive vinylic functionalities released during the hydrolysis step resulting in lower shrinkage during drying [81], catalysis [82], amount of water [83]) or to the building of the organic polymer chains or network (preformed or in situ formed polymer, nature of the grafted coupling agent [84], and grafting ratio [85]). Silica-based hybrids have been studied much more frequently than titania-based or zirconia-based hybrids. This is due to the fact that silicon can readily be covalently linked to the organic component through Si C stable bonds. By using organoalkoxy silanes, R1xSi(OR2)4-x with x=1 or 2, a silsesquioxane network built from R1SiO1.5 units(x=1, Eq.15)

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

102

FX Perrin

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

or polysiloxane linear chains (x=2, Eq.16) are obtained. For example, crack-free films with thickness higher than 20 µm can be obtained from a silicon precursor with two nonhydrolyzing phenyl substituents [86]. n R1Si(OR2)3 + 1.5 n H2O  [R1SiO1.5]n+ 3n R2OH

(15)

n (R1)2Si(OR2)2 + n H2O  [(R1)2SiO]n+ 2n R2OH

(16)

The organic group R1 can be non-reactive (for example methyl [87-92] or phenyl [93,94]) or reactive (for example amino [95-97], oxirane [98-103] or methacryloyl groups [104-108] in -aminopropyltriethoxysilane, -glycidyloxypropyltrimethoxysilane and methacryloyloxypropyltriethoxysilane, respectively). The reactive group R1 can attach to the paint polymer, which is applied on the silane-treated metal. Functional groups R1 can also be selected to strengthen the coating/substrate interface. For example, phosphonate functionalities in diethylphosphonatoethyl-triethoxysilane are capable of strong chemical bonding with the surface metal oxide layer on magnesium alloy substrates [109,110]. This results in an enhancement in the corrosion protection of magnesium AZ31B in dilute Harrison‘s solution (0.35wt% (NH4)2SO4 and 0.05 wt% NaCl) [110]. Bis-functional silanes (Figure 17), having the structure (R2O)3SiR1Si(OR2)3 result in highly crosslinked structures that have demonstrated their corrosion protectiveness for many metals in various performance tests [111-119]. Among the potential candidates, the bis-sulphur silane, bis-[triethoxysilylpropyl]tetrasulphide (C2H5O)3Si-(CH2)3-Sx-(CH2)3-Si(OC2H5)3 with an average x value of 3.8 has a promising future as chromate replacer [112-114,118,119]. The presence of the sulphur on the silane molecule certainly plays an important role on the deposition process by increasing film hydrophobicity and by its interaction with the metal substrate [114,118]. Recently, bis[(ureapropyl)triethoxysilane]-end-capped bis(propyl)-terminated polydimethylsiloxane diluted in either EtOH or a mixture of EtOH-PrOH(Figure 17) was used in thin film form (Vcr the melting begins amorphously.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

118

Igor V. Shishkovsky

The volume-time growth of the new phase is defined by equation of Kolmogorov Avraami:

V t   1  e kt

n (2)

above it is Avraami interpretation, and below the Kolmogorov expression is present

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

t  4 3 V t   1  exp  I r   * vr3 t    d  3 0 

(2a)

where the I(r) – is the nucleation velocity for the new phase. It is characterizing a nuclear value, which had been formed per the unit of time per the unit of volume. And vr – is the growing velocity (ref. to equation (1)). Under the great overheating and supercooling, I(r) leads to the great nuclear quantity per the unit of volume. Hence, the unique character of the laser treatment is a structure refinement. Under the small overheating - T, the nucleation velocity I(r) will small too. But if the time will enough for their growing (eq. (1), as under a furnace heating), so a great grain will grow. The grain size determines the future mechanical properties of a material. In any case the metal structure after crystallization will strongly depend on the velocity of crystallization process and the number of the crystallization centers, which appear per the unit of time and per the unit of volume (1/(mm3 s)), and the velocities of their linear growing (mm/s). Under very greater supercooling degree T, the atom mobility is insufficient already to realize the rearrangement them from the chaotic location in liquids in the ordering crystal. So, the structure of metal becomes amorphous. Below the analytical solution of kinetic equation for the DFPN for the crystalline phase Z(n,t) in laser coating is presented after supercooling. The existing approaches to the description of the cooling kinetics process under the LI are presented in [10-12,17]. We considered, that the Z(n,t) satisfies the normalization requirement 

 Z (n, t )dn  N

(3)

0

where N – is the full number of atoms in the volume. The knowledge of the distribution function Z(n,t) allows to determine an average size of a nuclear crystallite under cooling, as well as a relative fraction of the new phase volume. Due to the conditions of the continuous temperature variation T(t) at the achievement of crystallization temperature - T0, the thermodynamic condition of the PT of the first order will be carried out [13, 14]:

 S ( T0 )   l ( T ) , T  T0 ,

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(4)

High Speed Laser Assisted Surface Modification

119

here the Фs and Фl – are the specific free energy of a solid (s) and a liquid (l) phases, accordingly. According to a classical theory the PT of the I-st order occurs via the thermal fluctuation of a new atom arrangement fabrication. The new phase nuclei can grow only when the thermodynamic stimulus of the PT is presented. Such stimulus is determined by the sum of the energy change and is stipulated by the nucleus‘ appearance. A change in the specific free energy under the new phase of nucleus‘ appearance – the general thermodynamic system potential (P,T) can be written in the form: (P,T) = V(P,T) + S(P,T) + Ch(P,T) + El(P,T)

(5)

here V(P,T) – is the volume part of the thermodynamic potential; S(P,T) – is the surface part; Ch(P,T) – is a chemical part and El(P,T) – an elastic part. Each part could be written explicitly:

T  T t  4 r 3     f  f * n , f  h 0 V 3  T0

(6)

where r – is the nucleus radius of the new phase, n – the particle number of the new phase;  = a3 – is a voluentary unit of the new phase (~ a3, and а – the lattice parameter); h – is the latent heat, required to the new phase volume creation.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.





S

 4  r 2      n 2 / 3 ;

C     n  nk T  ln  inf  ; C  Ch  fin 



El

   n .

(7)

Here  - is the surface energy of the created phase interface,  - shape factor for this surface (it‘s equal 2/3 for the sphere),  - the chemical potential of system, which depends of concentration in the initial – Сinf and final – Сfin phases, at last  - change of the elastic energy under creation and movement of the phase interface. On the different stages of the nucleation process the any part of the thermodynamic potential (5) could be determinative. At present time, the comprehensive approach of all parts (5) was not realized. Traditionally, during the PT of the first order, the volume and surface parts of the thermodynamic potential are analyzed only. However, with the value other components, it cannot be underestimated. Under the load in the strained body with defects, without which is impossible, it‘s the thermodynamic balance; the hetero-phase fluctuations of strains, stresses, and density appeared. The microcavities begin to generate at some value of the local strain in the solid

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

120

Igor V. Shishkovsky

state. The diffusion of vacancies is a motive force for the pore growth by one approach. A difference vacancy condition under the loading and without them determined the thermodynamic potential changing. More consequently, by our opinion, is a second approach, according in which the microcavity growth is realized for the account of a pore nucleation directly near the surface and transitioning them by solid state into pores. Thereby, process of the pore generation under the loading is possible to present like a new phase generation from melt [19]. Herewith in the solution case, the potential difference is an effect of the concentration difference and the accordingly energy conditions of atoms in the new and old phases. In the case of the pure melt, this potential difference is a consequence of a different temperature stability of the new and old phases. The quantitative measure of degree of the system‘s metastability is a difference of chemical potentials in the old and new phases under the first order PTs. The temporary evolution of this value allows completely for one to describe the kinetics of critical phenomena. At the description of the PT in the solution or the liquid this value is identified as a supersaturation or an undercooling. So, for the building of the consequent kinetic theory of the PT in the free metastable condensing system (for instance, at the ferroelectric repolarization, the ferromagnetic magnetization, the martensite transformations, etc.), it is necessary to determine the appropriate analogue for the supersaturation or undercooling. In particular, in the kinetic theory of the one-axis ferroelectric switching [19], where the phase transition is described by the monpropellant order parameter, it is proposed to introduce the values, called as re-polarization, relative re-polarization, and re-deformation. [19]. Finally, we should be saying that in the process of the new phase grain nucleation under the decay of oversaturated solid solution, it is accepted to distinguish between two stages: the fluctuation regime of the nuclear growth and the coalescence stage. The coalescence is a growing of the large grains to the account of the dissolution smaller. On the relatively late stage of the PT, the coalescence process is dominated and named Oswald aging. The kinetics of coalescence was actively explored in classical paper Liphshitz - Slezov [20], in which a universal function of distribution by sizes was found for a new phase. At the asymptotic limit, any initial distribution has been evolved, then (ref. monograph [21] reviewed [22]). It was determined that nonlinear kinetics stipulates the extremely special system of behavior on the late stage of the decay. After some time, any system asymptotically "forgets" about its initial condition and transforms to single in the corresponding variable stable asymptotic condition, depending from acting in the system of the mass transfer mechanism only. Physically, the interfacial surface tension is formed universally (in corresponding variables) and uncombined from the initial stage, the distribution functions by the sizes. Such distribution function is one for all existing phases. The conservation law and active mass law are determined the phases, which asymptotically survive in the process of competitive growing. The boundary volumes of coexistence phases are estimated as stochiometric and non-stochiometric composition that enables buildings of the decay phase diagrams. The evolution of an average size of the microscopic defect in the dispersive system on late decay stage depending on the mass transfer mechanism is described by the diffusion law. Such kinetics is asymptotically stable under the mass transfer controlled dynamics. The kinetic equation could be presented:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High Speed Laser Assisted Surface Modification

 Z ( n,t ) U   2 Z 1   ( n )   2   n 2 3 exp  Z ( n , t )  t kT( t ) n  n   kT( t )  n

121

(8)

where U – is the activation energy of the atom transition through the phase boundary section, ν – is the characteristic Debye frequency, k – the Boltzmann constant.

T  T( t )    h 0 n T0

(9)

Δh – is a latent heat of the PT. As the initial condition we used a condition of the system completeness, which indicates the conservation of the particle number in their: 

Z( n,0 )  2N( n  1 ) ,  ( n  1 )dn  1

1 . 2

(10)

Boundary conditions were selected in the next form:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Z ( n , t ) n  1  N , Z ( n   ,t )  0 .

(11)

The study, conducted in [23-26], was shown that the crystalline grain under the supercooling has a needle dendrite shape. So we have been considered an existence of the crystalline phase, having mainly lamellar form. In this case, preexponential factors (8) will be proportional by ~ n, rather then ~ n2/3, peculiar to the spherical shape of nuclei. Thereby, we obtain:

 Z U   2 Z h T0  T ( t ) Z     2   n exp   t kT( t ) T0 n   kT( t )   n

(12)

We produce a changing of the variables

( t ) 

T  Z  Z   h T0  T ( t )  , ( t )  , , kT0 T0  t t  t

(13)

here υ – is the cooling velocity from the melt. So the equation (12) can be presented in the following form:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

122

Igor V. Shishkovsky 2 Z Z   U    Z ( , n )  nD0 (  ) exp       2  n    h   n 2  U  h  h   D0 (  )     exp   (  )k  kT0   kT0 

(14)

The equation (14) we solved with aid of the operator method, applicable for solving of the Schrödinger equation [12,17]. The operator of the system evolution could be presented:

Z f  SˆZ in , Z in  Z ( n,  0 ) , Z f  Z ( n,   )

(15)

where Zin and Zf – are initial and final distribution function. The operator evolution determination is reduced to deciding the differential equation system for α and β parameters [17]: t  U  ( t )   exp ( t ) exp( t )  d k T(  )   0 t  h T0  T (  ) U  d ( t )   exp   k T0 0 T (  ) k T (  )  

(16)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

For the explicit definition of parameters (t) and β(t), it is necessary to know the temporary dependence of temperature. For example, under the exponential temperature dependence

T ( t )  T0 exp( pt ),

0  t  ,

(17)

the cooling speed will vcool = T(-p), and the integrals (16) were taken explicitly:

( t   )  

h 1  k T0 U U exp( )  Ei(   k T0 p  U k T0 k T0

( t   )  

  U  Ei   F (    ); p  k T0 

( 1 ) k F (  )  exp( ) ( ) k! k 0 

k

 ) ;  .

(18)

1  k  k ( k  1 )  k ( k  1 )( k  2 )     k ( k  1 )( k  2 ) 5 * 4 * 3   

Here is Ei(x) – the integral exponential function. The explicit solution for the operator of the system evolution allows analyzing its action on the initial function of distribution. In initial notations the DFPN can be presented obviously in such integral type:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High Speed Laser Assisted Surface Modification   4n   2 n 1 2  Z f (  , , n )  dZ in   e  exp   0 4 4  

  n   I 1        

123

(19)

where I1 – is moditied Bessel function of the first-order. The explicit type of DFPN is determined among the other the cooling speed. For instance, if (20) Z in  Z 0 ( n  n0 ) so for linear law of cooling

 n  Z f   2 e    

34

 exp 





n  n0 e  2    

(21)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Fig. 1 presents results of the numerical simulation of the DFPN - Z(n,t). It is seen, that if at the initial time moments of the high-speed cooling the DFPN is «drifting» by the particle number axis greatly, but the cooling finish the "driftage" is stopping practically, then the surface structure of the coating is fixed.

Figure 1. Dependence of the distribution function by particle number as ln(n) for the different moment of time.

Average size of the crystalline nucleus or the average number of atoms in the nucleus of new phase is defined as:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

124

Igor V. Shishkovsky n

1 n   Z ( n,t )ndn N0

(22)

Take into consideration that under a large cooling a transition is possible within the limit of integration from n to ∞, average number of particles in crystalline; also it is possible to find in the explicit type, substituting the solution (19) in the (22):

n 

3  1      1  ln 1    2    3 

Here follows the next very important principle conclusion that

(23)

n

is completely

determined by α value, i.e. the nature of the diffusion of the initial distribution of the particle numbers. Then the relative volume new crystalline phases can to define as

V  1  exp( ( t )), V

(24)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Hence the volume of new phase is completely defined by the "driftage" velocity for the distribution functions in the space of the particle number. Fig. 2 is presented he results for the average particle number in the crystalline, depending on the cooling speed for aluminum, iron, and gold.

Figure 2. Dependence of average particle number in crystalline from cooling speed - v. 1 – aluminum, 2 – iron, 3 – gold.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High Speed Laser Assisted Surface Modification

125

In the paper [17], it was explored an asymptotic nature α and β under t   for the simplest cooling laws (linear, power, and exponential types) only. Knowledge of temperature distributions T (t and xi) in the liquid and solid phases allows a mathematically adequate description of the nucleation process.

CRITERIA OF AMORPHOUS STATE The obtained solution for Z(n,t), when the final function of the distribution is determined through the initial by means of the parameters  and , permits to a natural image "to trace" behavior of the volume of the new phase depending on cooling speed and to reveal some principle particularities of nano structure and amorphous phase generation after the LI. It is possible confirm that between the thermodynamic and kinetic parameters of the cooling process, the optimum relationship must exist [28]. The kinetic condition of the amorphous state is an achievement of such cooling velocities, when crystallization process has not a time. On the other hand, the velocity of heat abstraction from the melt in the volume of material is defined by its thermophysical properties and can limit the cooling velocity. We shall consider that the amorphous phase will be realized if the volume of new phase in accordance with (24)

exp(( t   )  P,

(25)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where P – is some value, closed unit, for example Р = 0,99(9). Hence, for the impulse LI, the maximum specific cooling speed will be on the surface and it is equal to [27]

vmax 

T0

AQ  * c

(26)

where А – is the absorption coefficient of material surface, Q – the laser power density, - the heat conductivity, density and heat capacity,  - the time of the LI. Taken  from (18) in the type

 h 1  k T0 U U exp exp( )  Ei(   k T0 k T0  k T0 vmax  U

 )   P , 

(27)

it is physically reasonable to assume that for the atomic-smooth surface U/(kT)>>5 [19], and then the function integral exponent - Ei could be simplified as

Ei ( 

kT U U )  C0 0 exp(  ),C  0.95. kT0 U kT0

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(28)

126

Igor V. Shishkovsky

So, the (28) is rewritten.

 h( 1  C0 ) U exp exp( kT0  Uvmax

 )  P 

(29)

The explicit type for the specific critical cooling speed under which the amorphization will be possible:

vmax 

h( 1  C0 ) U AQ exp(  ) . U ln P kT0 T0  * c

(30)

Hence, in the equation (30), we have established the relationship between the thermophysical (,  and c) and kinetic (U, ) material parameters for the laser source with the given parameters (A, Q, ). The weakness of the proposed criteria of the amorphous state lies in the fact that the parameter (1-С0)/ln(P) is an infinitesimal value and its further refinement is not physically obvious.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

CONCLUSION The next chemical factors are influencing the PT – the multi- componentry, the chemical reaction rate during LI on the surface, a medium acidity, etc. It is necessary to note that the technique diversity of these factors is situated a general approach by the I-th order PT analysis indeed, which was already presented above. So, high-constituent nature of powdered compositions is possible to take into account of writing the Fokker –Plank equation, in which the diffusion factor by the particle sizes must be recorded for each component. Accordingly, the nucleation and growing velocities, in which the diffusion multiplier is attended to also, must be presented in the form of the equation system by each of the components. If the complex chemical transformation had been gone during the PT and the new phase had appeared over a catalyst on the intermediate reaction, so it is possible a self-organization and self-oscillation processes in the system [29]. According to the I. Prigogine determination, there is a necessary existence of nonlinearity on the rate of the catalyst formation. The precipitation from the ion solution is one method for the nanostructure receiving. The solution acidity plays the decisive role, effecting the DFPN on the future composition and structure. Since the saturation and supersaturation depends of the acidity, the nucleation velocity on the phase boundary (joining rate of atoms at the nucleus surface) will be changed [19,30]. The Oswald coalescence in the multicomponent chemical mixtures is much more varied, since redistribution goes as by the particle sizes as in the concentration space. Hence, the general equation (8) must be complemented by the diffusion equation by each concentration plus the relationship between the separate parts of mixtures (boundaries and the domain of phase co - existence by the equilibrium phase diagram).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High Speed Laser Assisted Surface Modification

127

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Now there is a nano-technology development that has been initiated in particular interest to a quantum dot growing under the PT of first order. The growing mechanism of the ensemble coherent of nano-particles is satisfying the Stransky – Krastanov approach [31]. In the classical type it occurs at the substantial difference in the lattice parameter of the substrate and the nanostructure. The coherent relationship between the substrate and growing film implies a template mechanism of the growth. The optimal coating depth could be estimated under the balance condition between mechanical stresses caused by the lattice parameter discrepancy and the surface tension. Such a growing mechanism is called an island effect. Then, under the transition from the 2D to the 3D growing, obviously, the process must be managing so that the film thicknesses were less critical. The conditions of the defect (dislocation and pore) development on the micro and nano level significantly influence the nano-structural catalysts. The vacancy approach of crack and pore generation was proposed in the paper [32]. The generation and growing vacancy processes can be described in the frame work of the PT of first order also. The vacancy ensemble behaves similarly to the gas, which under the specific conditions can be condensed, which leads to the pore generation. At this time, the analytical solution of the kinetic equation of the DFPN was obtained under the high-speed laser cooling from the melt. Kinetic equation is solved with using the physical prerequisites of the crystallization process, i.e. the existence of sufficiently large number of nucleus of supercritical size and the lamellar shape of nucleus instead of spherical. The integral type of the DFPN allows one to determine the final results of the nanocrystallization, knowing an initial distribution function. The kinetic equation solution can be used for the experiment description by the nano-structure fabrication on the metal surface under the femtosecond pulse LI.

REFERENCES [1]

Waheed Ul Haq Syed, Andrew J. Pinkerton, Zhu Liu, Lin Li. Applied Surface Science, 2007, v. 253, 7926–7931. [2] E. Yarrapareddy, R. Kovacevic. Surface & Coatings Technology, 2008, v 202, 1951– 1965. [3] A. Piqu´e, D.B. Chrisey, R.C.Y. Auyeung, etc. Appl. Phys., 1999, v. A 69 S279–S284. [4] M.J. Aziz. Film Growth Mechanisms in Pulsed Laser Deposition. Harvard School of Engineering and Applied Sciences, Cambridge MA, October 13, 2007. [5] H.U. Krebs, O. Bremert. Appl.Phys. lett.1993, v. 62, 2341. [6] J. Dechant. Utrarotspektroskopische Untersuchungen an Polymeren. Akademie Verlag. Berlin. 1972. [7] J.A. Martin, L.Vazquez, P.Bernard and etc. Appl. Phys. Lett., 1990, v. 57, 1742. [8] M. Okoshi, K.Higashikawa, M. Hanabusa. The Japan Society of Applied Physics, Toyohashi University of Technology, June 2000. [9] S. Nishio, T.Chiba, A.Matsuzaki and etc. J. Appl. Phys. , 1996, v. 79, 7198. [10] Lubov B.Y. Theory crystallization in the big volumes. М.: Science, 1975. 256 p.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

128

Igor V. Shishkovsky

[11] Christian J. Theory of transformation in the metals and alloys. М. World. 1987. v. 1, 780 p. [12] Frenkel Y. I. Kinetic theory of liquid. L.: Science, 1975. 591 p. [13] Shklovskiy V.A Surface. Physics, chemistry, mechanics. 1986. 6 issue, p. 91-98. [14] Kudinov G.M., Shmakov V.A Reports of AS USSR, 1982, Vol. 264, issue 6, p. 610-614. [15] Glytenko A.L., Shmakov V.A. Reports of AS USSR 1984. Vol. 276. issue 6, p. 13921396. [16] Zeldivich Y.B. Sov. Phys - JETF 1942, v.12, 4(10), 525. [17] Zavestovskaya I.N. and etc. Proceedings of LPI of RAS, 1993. V.217. p. 3- 12. [18] Gureev D.M., Mednikov S.I. Proceedings of LPI of RAS, 1993, v. 217. p. 37-41. [19] Kukushkin S.A., Osipov A.V. UFN, 1998, V 68, 1083. [20] Lifshits I M and Slyozov V.V. Sov. Phys.—JETP 1959, V. 8, 331. [21] Kashchiev D. Nucleation: Basic Theory with Applications 2000, (Oxford: Butterworth Heinemann) [22] Dubrovskii V G, Cirlin G E and Ustinov V.M. Phys. Rev. B2000,v. 68 p. 409. [23] Koch J., Korte F., Bauer T., Fallnich C., Ostendorf A., Chichkov B. Appl. Phys. A. 2005. V. 81. P. 325-328. [24] Vorobyev A.Y., Guo Chunlei, Optics express. 2006. V.14, № 6, P. 2164-2169. [25] Nolte S., Chichkov B., Welling H., Shani Y., Lieberman K., Terkel H. Optics letters. 1999. V.24. №13. P.914-916. [26] Koch J., Korte F., Fallnich C., Ostendorf A., Chichkov B. Optical Engineering. 2005. V.44(5), P. 051103. [27] Rukalin N and etc. Laser treatment of materials. М.: World, 1988. 496 p. [28] Shishkosky I. Proceedings of All-Russian Conf. by physics-chemical properties of amorphous alloys, 1989, p. 12. [29] Nikolis G., Prigojin I. Selforganization in heterogenous systems, 1979, World Publ, М. 512 p. [30] Osipov A V, Schmitt F, Kukushkin S A and Hess P. Appl. Surf. Sci.,2002, v. 188, p.156. [31] Stoyanov S., Kashchiev D. Curr. Top. Mater. Sci, 1981, V. 7, p 69. [32] Bimberg D., Grundmann M. and Ledentsov N.N. Quantum Dot Heterostructures 1999 (Chichester: Wiley).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: High Performance Coatings for Automotive and Aerospace… ISBN: 978-1-60876-579-9 Editor: Abdel Salam Hamdy Makhlouf, pp. 129-140 ©2010 Nova Science Publishers, Inc.

Chapter 6

AN ATTEMPT FOR DESIGNING ECONOMICALLY ATTRACTIVE CHROME-FREE CONVERSION COATINGS FOR MAGNESIUM ALLOYS Abdel Salam Hamdy*and Mahmoud Farahat Department of Surface Technology & Corrosion, Central Metallurgical Research and Development Institute, CMRDI, PO Box: 87, Helwan, Cairo, Egypt.

ABSTRACT

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Magnesium alloys are advanced, light structural and functional materials being increasingly used in the automotive, aerospace, electronic, and energy industries. However, their corrosion protection performance at the current stage of development is still not good enough for increasingly diverse practical applications. Chromate has been reported as the most efficient widespread conversion coatings for the corrosion protection of magnesium alloys prior to application of adhesives or paint coatings. Because of the high toxicity and carcinogenicity of hexavalent chromates, alternative conversion coatings are being urgently sought. This search has been intensified by the restrictions placed on automobile manufacturers, for whom the use of hexavalent chromium has been severely limited since 2003. Corrosion and prevention of light alloys such as aluminum and magnesium are an important parts of the CMRDI research program and worldwide as well. This chapter discusses the recent research activities by our research group at CMRDI and relevant research work in this area in the world. The aim of this chapter is to deepen the current understanding of corrosion and protection of magnesium alloys and to provide a base for future research work in this field. It reports the recent development in designing ecofriendly conversion coatings based on cerate or zirconate surface treatments for magnesium alloys as alternatives to toxic chromate-based systems.

*

Present address: Max Planck Institute of Colloids and Interfaces, Am Mühlenberg 1, 14476 Potsdam-Golm , Germany . http://www.mpikg.mpg.de/english/04-interfaces/researchGroups/nonPlanarInterfaces/activeInterfaces/members/Makhlouf/index.html

E-mail: [email protected] High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

130

Abdel Salam Hamdy and Mahmoud Farahat

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION The demand for environmental protection and energy saving have been promoting researchers to develop lightweight materials for automotive and aerospace industries. Magnesium alloys are considered to be an excellent material for reducing vehicle weight, lowering fuel consumption and, thereby, reducing CO2 emissions. Several automobile manufacturers (Ford, General Motors, Chrysler, Volkswagen, Opel, and Fiat) have cooperated to develop new magnesium alloys for manufacturing less energy-consuming cars and hence, a less polluted environment [1–4]. According to the International Magnesium Association's website, ―world production of magnesium exceeds 429,000 tons per annum and the figure is increasing annually as the lightweight properties of magnesium alloys are used increasingly in the automotive industry as a means of reducing weight, increasing fuel efficiency and reducing greenhouse gas emissions‖. Competition from lower cost Chinese magnesium (about 1840 $/t in 2006) has resulted in a supply by China of almost 80% of the world‘s primary magnesium [1–4]. Rather than compete in magnesium metal production, the strategy of several companies is based on developing new magnesium alloys for use in new products. Magnesium alloys have a variety of excellent properties, including a high strength-toweight ratio, low density, dimensional stability, and castability. Therefore, they found widespread applications in many industrial sectors. Although the corrosion resistance of magnesium has been significantly improved by alloying with aluminum, manganese, or zinc and reducing the content of impurities, its corrosion resistance is still limited. Therefore, several surface modification treatments such as chemical conversion coatings [5–13], anodizing [14–18], electroless nickel deposition [19], nickel/polymer electro-deposition [20], PVD of high-purity magnesium [21], and nitrogen ion implantation [22,23] are widely used to improve the corrosion resistance of Mg. However, the existing methods are frequently either expensive such as PVD and nitrogen ion implantation or unable to produce the surface properties desired for many applications where magnesium alloys would otherwise be highly competitive. Chromate conversion coatings (CCC) have been widely applied for corrosion protection of aluminum and magnesium alloys. Major reasons for the widespread use of CCC are their self-healing nature, the ease of application, high electric conductivity, and their high efficiency/cost ratio. These advantages have made them a standard method of corrosion protection. Moreover, they provide the greatest level of under-film corrosion resistance and they facilitate the application of further finishing treatment [24]. However, hexavalent chromate is ranked by EPA among the most 168 top toxic substances due to the carcinogenic effect and environmentally hazardous as a waste product [25]. Because current environmental legislation and tightening regulatory pressures are moving towards total exclusion of the process involving toxic hexavalent chromate, many attempts are being made to develop chrome-free surface treatments for corrosion protection [26]. This chapter discusses our research efforts for developing environmentally-acceptable surface treatments based on cerate or zirconate conversion coatings that can provide covalent bonding for strong coating adhesion and act as a barrier coating to limit the transport of water to the surface of magnesium alloys [27-44]. The effect of surface modification prior to

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

An Attempt for Designing Economically Attractive Chrome-Free Conversion …

131

applying cerate- [27-35], or zirconate-based [42-44] conversion coatings on the corrosion protection of magnesium alloys in NaCl solution were studied.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2. CERIUM CONVERSION COATINGS FOR AZ91D Cerium conversion coatings have been proposed as alternative to toxic chromate-based conversion coatings especially for light weight materials such as aluminum and magnesium and their alloys and composites materials. The effect of surface modification prior to cerate conversion coatings on the corrosion behavior of AZ91D was investigated. A series of specimens was prepared under the following conditions a) as polished, b) directly treated, c) alkaline etching, d) acidic etching, e) oxide thickened and, f) alkaline etching followed by oxide thickening. After surface modification, the specimens were immersed in concentrated ceria solutions [more details about the experiments can be found in Ref. 27-35]. A novel surface modification method consisted of pre-etching the surface of AZ91D with alkaline pH solution prior to applying ceria-based ceramic coating was proposed [32-35]. Results showed that ceria treatments improve the pitting and crevice corrosion resistance due to the formation of cerium rich protective oxide films which act as a barrier to oxygen diffusion to the metal surface, and hence, shift the pitting potential to nobler one [32-35]. Although the surface resistance of the as-polished samples is 6.6 X 103 Ω.cm2 as measured in previous work [32-35], the localized corrosion resistance of the ceria treated samples was generally improved which was confirmed by the relaxation of the impedance spectra. According to the electrochemical impedance measurements [32-35], the surface resistance of the acidic-pickled samples and the oxide-thickened samples is almost with the same order of magnitude (0.5X103 Ω.cm2), while the surface resistance of the alkaline-treated samples and the samples alkaline-treated and oxide-thickened is almost three times higher than the other surface treatments (1.5X103 Ω.cm2). These results indicate that the alkaline etching step in KOH inhibited many active surface sites, whereby the number of pitting areas decreased from about eight pits for the as-polished specimens to 3 and 4 tiny pits for both the alkaline treated samples and the samples alkaline-treated and oxide-thickened respectively. According to EIS, optical microscope and SEM, alkaline etching prior to ceria treatments plays an important role in inhibiting the active surface sites by blocking and repairing the pits (Fig. 1) and rejecting the chloride ions from the surface [32-35]. Moreover, it seems that the alkaline etching enhances the formation of cerium-rich oxide at the surface of Mg substrate. Such film acts as a barrier to oxygen diffusion to the metal surface thereby impeding but not preventing corrosion. Because this film is relatively thin, general corrosion increased due to diffusion of Mg ions through this film. Localized corrosion is the predominant corrosion species in the as-polished samples. Conversely, the most predominant corrosion species in the samples that subjected to alkaline etching and oxide thickening is the general corrosion. The surface resistance is the summation of pitting and general corrosion resistances.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

132

Abdel Salam Hamdy and Mahmoud Farahat

Pitting corrosion

Pitting Corrosion

Directly Treated

Oxide thickening

Self Healing

Micro-cracks

Alkaline etching

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Acidic pickling

Pitting corrosion

Micro-cracks

Alkaline etching followed by oxide thickening

Figure 1. SEM of cerate-coated samples after one week of immersion in NaCl solution. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

An Attempt for Designing Economically Attractive Chrome-Free Conversion …

133

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

3. ZIRCONATE CONVERSION COATINGS FOR AZ31D Zirconium oxide thin films and coatings have wide applications in optics, fuel cells, capacitors, sensors, catalysts, and protective coatings due to its unique properties such as excellent corrosion resistance, superior refractoriness, good strength, high ionic conductivity, low thermal conductivity, and high thermal expansion coefficient which is very similar to those of many metal substrates [45–47]. Zirconium oxide coatings can be prepared by different methods such as sputtering [48,49], plasma spraying [50], laser ablation [51], metal organic chemical vapor deposition [52], and sol–gel [53,54]. However, these methods have several shortages such as costly equipment, complicated process, and relatively low interface bonding strength; which hindered their further developments and applications. Recently, a new plasma electrolytic oxidation technique was proposed for aluminum and titanium alloys to provide a solution for these problems [55–57]. However, the work reported on magnesium alloys are very scarce and the corrosion resistances due to the proposed technique are not attractive [58,59]. In a recent study [4], a new Ce-Zr-Nb- based conversion coating was designed for some magnesium alloys. Electrochemical measurements showed that Mg alloys treated during 24 h in the Ce–Zr–Nb conversion bath exhibit improved corrosion resistance, and excellent paint adhesion. It was reported that such coating offers an alternative to the chromate conversion coating for magnesium alloys. However, the treatment time is too long and several treatment steps should be used. To date, the proposed processes for designing zirconate coatings for the corrosion protection of magnesium alloys are industrially inapplicable where multi-steps should be used and several salt solutions such as Ce, Nb, or Mo in addition to zirconate should be mixed together which is economically unattractive. Moreover, the overall corrosion protection resulted from such types of conversion coatings is limited. Very recently, the effect of zirconate solution concentration on the corrosion resistance of magnesium AZ31D was studied by our research group at CMRDI [42-44]. Linear polarization, cyclic voltammetry, electrochemical impedance spectroscopy, and SEM-EDS were performed to assess the performance of the zirconate conversion coatings in 3.5% NaCl solution. Generally, zirconium conversion coatings of concentration lower than 50 g/l improve the pitting and crevice corrosion resistance due to the formation of ceramic coating of magnesium hydroxide doped with zirconium oxide that act as a barrier to oxygen diffusion to the metal surface, and hence, decrease the chance for pitting corrosion. Our results showed that increasing the zirconate concentration from 5-20 g/l improves the corrosion resistances [42-44]. The highest surface resistance obtained from the samples that treated with 20 g/l zirconate coatings. Increasing the zirconate concentration from 20 to 50 g/l has an adverse effect of the corrosion resistance [more details can be found in Ref. 42]. SEM showed that Mg surface treatment in 20 g/l zirconate solution plays an important role in inhibiting the active surface sites, rejecting the chloride ions from the surface, and forming uniformly-distributed magnesium hydroxide layer enriched with tin oxide. Moreover, treatment of Mg in zirconate solution has a strong ability to repair the surface defects and pitting corrosion (self-healing nature) as shown in Fig. 2 [43].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

134

Abdel Salam Hamdy and Mahmoud Farahat

Figure 2. SEM image for the zirconate-coated samples (20 g/l) after one week in NaCl solution.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

On the other hand, the promising results we previously obtained after the surface modification of Mg AZ91D (using alkaline etching) prior to cerate conversion coatings that resulted in improving the localized corrosion resistance in 3.5%NaCl solution [32-35], encouraged us to try it with AZ31D prior to zirconate coatings. Therefore, the effect of alkaline etching of AZ31D followed by treatment in zirconate solutions of different concentrations was investigated in details.

Figure 3. Optical photos for Mg AZ31D samples etched in KOH solution then treated in different zirconate solutions after seven days in 3.5% NaCl solution.

Fig. 3 is the optical photos for as-polished sample and the samples etched in alkaline solution of KOH followed by zirconate treatment at different concentrations (5, 10 and 20 g/l). All samples revealed severe localized and general corrosion after seven days of immersion in 3.5% NaCl solution. The photos show the adverse effect of alkaline etching step prior to zirconate conversion coating application although the samples treated in 10 g/l zirconate showed a relative improvement in the localized corrosion (pitting and crevice) resistance compared with other samples. However, this relative improvement in localized

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

An Attempt for Designing Economically Attractive Chrome-Free Conversion …

135

corrosion can not be compared economically or technically with that one obtained in the previous studies [42-44] after one-step direct treatment in 20 g/l zirconate.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 4. Nyquist plots of different zirconate coated AZ31D after one week of immersion in NaCl solution.

The EIS spectra in Nyquist plots in Figure 4 are similar in all curves except for the difference in the diameter of the loops. This means that the corrosion mechanisms in all samples are the same, but their corrosion rates are different [37]. The best surface resistance was obtained from the samples that etched followed by treatment in 10g/l zirconate solution which is 2.75 X 103 Ω.cm2, while the surface resistance of the as-polished sample is approximately 2.10 X 103 Ω.cm2. Although the difference between the total values of the surface resistances is small, a significant improvement in the pitting and the crevice corrosion resistances is observed (Fig. 3). Samples treated in 5 and 20 g/l zirconate solutions revealed a surface resistance even less than the as-polished. These results are in agreement with the macro examination (Fig. 3) and SEM images (below) which revealed a loosely adhered oxide film on the surface of the sample treated in 20 g/l zirconate. This loosely adhered film is easily to be penetrated by the electrolyte through the bare metal allows chloride ions and oxygen to reach and attack the bare metal. Bode plots are quite informative in interpretation of diffusion process as it provides data related to the variation of impedance and phase angle with respect to frequency. As presented in (Fig 5) one can see the presence of three different frequency regions with respect to impedance │Z│ and phase angle (θ). In general, the high frequency region determines the properties of reference electrode and solution resistance. The medium frequency region describes capacitive properties of film and the low frequency region explains the charge transfer process occurring at solution / coating interfaces [39]. In figure 5 at the low frequency region in which a relaxation in the spectra was observed which is a sign for the pitting corrosion occurrence in magnesium alloys. However, it can be observed that the sample treated in 10g/l zirconate solution has a limited fluctuation in the phase angle

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

136

Abdel Salam Hamdy and Mahmoud Farahat

compared with the other samples which is an indication for the relatively limited diffusion of chloride ions through solution/coating interfaces and localized corrosion occurrence [39].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 5. Bode plots of different zirconate coated AZ31D after one week of immersion in NaCl solution.

SEM images of the as-polished sample (Fig. 6), revealed some surface defects before corrosion. These surface defects are usually the preferred sites for pits initiation in the presence of chloride ions and oxygen, promoting pitting corrosion. A network of mud-like cracks was formed on the surface of the alkaline etched samples treated with zirconate even before immersion in 3.5 % NaCl solution (Fig. 6). These cracks may be attributed to the evolution of hydrogen during the treatment process or due to the hydration after coating application. The samples treated in 10 g/l zirconate revealed a needle-like zirconium-oxide network partially covered the pits and micro-cracks. It seems that the relative improvement in the corrosion resistance of such samples is due to the partial self-healing behavior due to formation of the needle zirconium-oxide network after immersion in 3.5% NaCl solution.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

An Attempt for Designing Economically Attractive Chrome-Free Conversion …

137

Figure 6. SEM of alkaline etched samples followed by zirconate treatment with different concentrations before (left) and after (right) corrosion in 3.5 % NaCl solution.

5. CONCLUSION This chapter discusses our research achievements for preparing new eco-friendly surface treatments based on salt solution of ceria or zirconia conversion coatings for the corrosion protection of magnesium alloys AZ91D and AZ31D in NaCl solution. The effect of surface modification prior to ceria coatings on the corrosion behavior of magnesium was investigated. Results showed that alkaline etching is more effective in reducing the pitting corrosion than acidic pickling. Alkaline etching plays an important role in inhibiting the active surface sites, rejecting the chloride ions from the surface and forming uniformly-distributed oxide film.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

138

Abdel Salam Hamdy and Mahmoud Farahat

On the other hand, the effect of zirconate solution concentration on the corrosion resistance of magnesium AZ31D was also studied. A one-step surface treatment in diluted zirconate solution (20 g/l) showed the highest surface resistance. Increasing or decreasing the zirconate concentration more or less than 20 g/l negatively affects the corrosion resistance. Surface modification (using alkaline etching) prior to applying zirconate conversion coatings was found to have a negative effect in the corrosion protection performance of magnesium. The optimum conditions under which cerate and zirconate treatments can provide good corrosion protection to the magnesium substrate were determined.

REFERENCES [1] [2] [3] [4] [5] [6] [7] [8]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]

Hydro Magnesium, Corrosion and finishing of magnesium alloys. http://www.hydro.com/magnesium/ Kramer D.A., Mineral industry survey, Magnesium in the First Quarter, USGS Minerals, 2006, p. 1. Pure Magnesium from China, Inv. No. 731-TA-895 (Review), USITC Publication 3908, 2007. Ardelean H., Frateur I., Marcus P., Corrosion Science 50 (2008) 1907–1918 Lin C.S., Lin H.C., Lin K.M. and Lai W.C., Corrosion Science 48 (2006) 93–109 Simaranov A.Y., Marshakov A.I. and Mikhailovskii Y.N., Protect. Met. 25 (1990) 611. Hawake D. and Albright D.L., Met. Finish. 34 (October) (1995). Gonzalez-Nunez M.A., Skeldon P., Thompson G.E., Karimzadeh H., Lyon P. and Wilks T.E., Corros. Sci. 37 (1995) 1763. Gonzalez-Nunez M.A., Skeldon P., Thompson G.E. and Karimzadeh H., Corrosion 55 (1999) 1136. Rudd A.L., Breslin C.B. and Mansfeld F., Corros. Sci. 42 (2000) 275. Ono S., Asami K. and Masuko N., Mater. Trans. JIM 42 (2001) 1225. Umehara H., Takaya M. and Kojima Y., Mater. Trans. JIM 42 (2001) 1691. Dabala M., Brunelli K., Napolitani E. and Magrini M., Surf. Coat. Technol. 172 (2003) 227. Biestek T. andWeber J., Electrolytic and Chemical Conversion Coatings, 1st ed., Portcullis Press Ltd.,Poland, 1976, pp. 208–210. Ono S., Asami K., Osaka T. and Masuko N., J. Electrochem. Soc. 143 (1996) L62. Khaselev O. and Yahalom J., J. Electrochem. Soc. 145 (1998) 190. Khaselev O., Weiss D. and Yahalom J., J. Electrochem. Soc. 146 (1999) 1757. Bonilla F.A., Berkani A., Liu Y., Skeldon P., Thompson G.E., Habazaki H., Shimizu K., John C. and Stevens K., J. Electrochem. Soc. 149 (2002) B4. Xiang Y., Hu W., Liu X., Zhao C. and Ding W., Trans. IMF 79 (2001) 30. Mori K., Hirahara H., Oishi Y. and Sasaki H., Mater. Trans. JIM 42 (2001) 1219. Yamamoto A., Watanabe A., Sugahara K., Fukumoto S. and Tsubakino H., Mater. Trans. JIM 42(2001) 1237. Fukumoto S., Yamamoto A., Terasawa M., Mitamura T. and Tsubakino H., Mater. Trans. JIM 42(2001) 1232. Nakatsugawa I., Martin R. and Knystautas E.J., Corrosion 52 (1996) 921.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

An Attempt for Designing Economically Attractive Chrome-Free Conversion …

139

[24] Bethencourt M., Botana F.J., Calvino J.J., Marcosand M.M. and Rodrìguez-Chacòn M.A., Corrosion Sci., Vol.40, No.11, P.1803, 1998. [25] McCoy C., Patented at the second AESF/EPA Chromium Colloquium, Miami, FL, Feb. 1990. [26] Mansfeld F. and Wang Y., Material Science and Engineering A 198, P.51, 1995. [27] Hamdy A. S., "The correlation between Electrochemical Impedance Spectroscopy (EIS) and other polarization techniques for the corrosion evaluation of coated and bare metals in aqueous solutions‖. Invited chapter in: T. Kalniņš and V. Gulbis (Eds), ―Corrosion Protection: Processes, Management and Technologies‖, Nova Science Publishers, USA, Chapter 7, PP. 161-173, 2009, ISBN: 978-1-61668-226-2. [28] Hamdy A.S., ―Tailor-making of eco-friendly ceramic and nano-ceramic coatings for industrial applications‖, Invited Speaker at Nano Cement, Steel and Construction Industries Conference, 16-17 May 2009, Cairo, Egypt. [29] Hamdy A.S., "Novel eco-friendly surface coatings for automotive and aerospace materials". Invited Speaker at The Arabian-European Symposium on Environmental Protection, 27-28 September, 2005, Tunisia. [30] Hamdy A. S., J. Pitture e Vernici-European Coatings, Vol. 86, No. 3, PP 43-50, 2008. [31] Hamdy A. S., "A novel approach in designing high performance eco-friendly coatings for aerospace and automotive industries.", Invited distinguished lecturer at the Summer Program on "Advanced Materials and Structures -AMS'08", School of Engineering, Universitatea 'Politehnica" din Timisoara, Romania, May 28-30, 2008. [32] Hamdy A.S., and Farahat M., "Novel approaches in designing protective coatings for magnesium alloys", Electrochemical Society Symposium on Coatings and Corrosion Protection, October 4-9, 2009, Vienna. [33] Hamdy A.S., J. Anti-Corrosion Methods and Materials, Vol. 53, No. 6 (2006) 367–373 [34] Hamdy A.S., "Green cerate based surface treatment for improving the corrosion resistance of magnesium alloy AZ91D in marine environments", Symposium ―Green Engineering for Materials Processing‖, Materials Science & Tech. Proceeding MS&T‘06, Oct. 15-19, 2006, Cinergy Center, Cincinnati, USA, PP. 141-150. [35] Hamdy A. S., J. Electrochemical and Solid-State Letters, Vol.10, No.3, C21-C25, 2007. [36] Hamdy A. S. and Butt D., "Eco-friendly conversion coatings for automotive and aerospace materials‖. Invited talk at the European Coatings Conference "New Concepts for Anti-Corrosive Coatings", 14-15 June 2007, Berlin, Germany. [37] Wenbin H., Meifeng H., Lei L., Yating W., Zhixin T., Corrosion Science, 50 (2008) 3267–3273 [38] Thomsen D., Hamdy A.S., Marx B. and Butt D. P., ―Corrosion behavior of some newly developed Mg alloys in NaCl solution‖, Symposium "Corrosion and Coatings Challenges in Industry", 88th Annual Meeting, Boise, ID, June 19, 2007, co-located with the 62nd Annual Meeting of the American Chemical Society. [39] Metikos-Hukovic M., Tkalcec E., Kwokal A., Piljac J., Surf. Coat. Technol. 165 (2003) 40. [40] Marx B., Hamdy A.S., Butt D. P. and Thomsen D., ―Corrosion of newly developed magnesium alloys in chloride containing solutions‖, Automotive Symposium, Proceeding MS&T '07– Sept. 16-20, 2007, Cobo Hall, Michigan.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

140

Abdel Salam Hamdy and Mahmoud Farahat

[41] Hamdy A.S., Marx B. and Butt D. P., ―A new approach in designing eco-friendly low cost chemical conversion coatings for magnesium alloys‖, Keynote Speaker at 6th International Symposium on Surface Protective Coatings, Goa, India, Feb. 25-28, 2009. [42] Hamdy A.S., and Farahat M., Surface and Coatings Technology, 204 (2010) 2834–2840 [43] Hamdy A.S., and Farahat M., "Zirconia based ceramic coatings for magnesium alloys", The European Corrosion Congress, paper #7816, Session ―Coatings WP 14‖, EUROCORR 2009, 6-10 September 2009, Nice, France. [44] Hamdy A.S., ―Over 10-years of muti-international cooperation in designing high performance environmentally friendly coatings technologies for automotive and aerospace materials‖, Invited Speaker at conference ―Cooperation with Germany – experience, new forms and perspectives‖, in memoriam of 150 years of Alexander von Humboldt, 17-20 September 2009, Kishinev, Moldova, sponsored by the Alexander von Humboldt Foundation, Germany. [45] Nishizawa K., Miki T., Fukaya H., Masuda Y., Suzuki K., Kato K., Thin Solid Films 516 (2008) 2635–2638. [46] Shi J.Y., Verweij H., Thin Solid Films 516 (2008) 3919–3923. [47] Guo G.Y., Chen Y.L., J. Solid State Chem. 178 (2005) 1675–1682. [48] Tsai T., Barnett S.A., J. Electrochem. Soc. 142 (1995) 3084–3087. [49] Ruddell D.E., Stoner B.R., Thompson J.Y., Thin Solid Films 445 (2003) 14–19. [50] Schwingel D., Taylor R., Haubold T., Wigren J. and Gualco C., Surf. Coat. Technol. 108–109 (1998) 99–106. [51] Caricato A.P., Cristoforo A.D., Fernandez M., Leggieri G., Luches A., Majni G., Martino M. and Mengucci P., Appl. Surf. Sci. 208–209 (2003) 615–619. [52] Carta G., Habra N.E., Rossetto G., Zanella P., Casarin M., Barreca D., Maragno C. and Tondello E., Surf. Coat. Technol. 201 (2007) 9289–9293. [53] Lee Y.H., Kuo C.W., Shih C.J., Hung I.M., Fung K.Z., Wen S.B. and Wang M.C., Mater. Sci. Eng. A 445–446 (2007) 347–354. [54] Viazzi C., Bonino J.P., Ansart F. and Barnabé A., J. Alloys Compd. 452 (2008) 377– 383. [55] Wu Z.D., Jiang Z.H., Yao Z.P., J. Inorg. Mater. 22 (3) (2007) 555–559 (in Chinese). [56] Yao Z.P., Jiang Z.H., Zhang X.L., J. Am. Ceram. Soc. 89 (9) (2006) 2929–2932. [57] Yao Z.P., Jiang Y.L., Jia F.Z., Jiang Z.H., Wang F.P., Appl. Surf. Sci. 254 (2008) 4084– 4091. [58] Yao Z.P., Gao H.H., Jiang Z.H., Wang F.P., J. Am. Ceram. Soc. 91 (2) (2008) 555–558. [59] Mu W.Y., Han Y., Surf. Coat. Technol. 202 (2008) 4278–4284.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: High Performance Coatings for Automotive and Aerospace… ISBN: 978-1-60876-579-9 Editor: Abdel Salam Hamdy Makhlouf, pp. 141-227 ©2010 Nova Science Publishers, Inc.

Chapter 7

HIGH-PERFORMANCE NANOSTRUCTURED SEMICONDUCTOR AND METALLO-DIELECTRIC LAYERS FOR SPACE APPLICATIONS I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu Academy of Sciences of Moldova, Republic of Moldova

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Nanostructured semiconductor and functionalized dielectric layers demonstrate highabsorbance which is extremely important for the development of new generation solar cells. We demonstrate that nanostructuring of semiconductor materials significantly increases their radiation hardness, a very important issue for space applications. Nanostructured semiconductor and dielectric templates represent an important basis for the production of novel nanocomposites, including smart and negative-refractive index materials for optoelectronic and photonic applications. We demonstrate that electrochemistry is one of the most efficient and cost-effective approach for controlling the architecture of III-V and II-VI semiconductor compounds on the nanometer scale by introducing porosity. A variety of architectures are demonstrated, including periodic spatial distribution of pores achieved by anodic etching governed by self-organization phenomena. A neutral electrolyte based on aqueous solution of NaCl is proposed instead of commonly used aggressive acids or alkaline electrolytes for the purpose of producing nanostructured semiconductor films. This technology in combination with electrochemical deposition of nanowire, nanodot, and nanotubular metallic structures in semiconductor and dielectric nanotemplates possessing ordered two-dimensional hexagonal arrays of pores is increasingly used for the fabrication of metalo-dielectric nanocomposites for optoelectronic, photonic and plasmonic applications. Special attention is paid to the development of methods for the fabrication of nanostructured layers of wide-band gap compounds such as GaN, ZnO, and TiO2. These methods include metalorganic chemical vapor deposition for growth of ZnO layers of various morphologies, photoelectrochemical treatment for nanostructuring GaN layers, and electrochemical oxidation of Ti foils for the fabrication of nanotubular TiO2 structures. A new technology, the so-called ―surface charge lithography‖, for micro- and

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

142

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu nanofabrication based on GaN layers is presented. Applications of nanostructured ZnO layers in micro- and nano-lasers, GaN layers in electronic sensors, and TiO2 layers in photonic crystals are demonstrated.

1. INTRODUCTION The discovery of luminescent microporous Si in 1991 [1,2] triggered considerable interest in porous semiconductors. Products relying on porous Si are reality in the ELTRAN process used by Canon to produce ―SOI‖ (Si on insulator) wafers on an industrial scale [3]. Porous semiconductors other than Si are receiving more and more interest, the most noteworthy being III-V and II-VI compounds (and SiC [4,5] which will not be discussed here). Porous compounds exhibit new properties with a large potential for applications in high-efficiency solar cells, nonlinear optical devices and THz emitters, sensors, waveguides and photonic crystals [6]. Compound semiconductors offer much more possibilities as compared with Si due to the fact that the shift from elementary to compound semiconductors entails a major crystallographic modification that influences the anisotropy of electrochemical etching. Some important properties of porous compound structures have been reported. In particular, the following findings deserve attention: 

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.



 







A sharp increase in the intensity of the near-band-edge photoluminescence (PL) in anodically etched GaP along with the emergence of blue and ultraviolet luminescence [7, 8, 9]; A strongly enhanced photoresponse was observed during pore formation in n-GaP electrodes by anodic etching in sulphuric acid solution [10,11]; and enormous light scattering in the porous medium was reported [12,13,14]; Porosity-induced modification of the phonon spectrum was observed in GaP, GaAs, InP [15-18], InAs [19], ZnSe [20,21]; An efficient optical second harmonic generation (SHG) was observed in porous GaP and InP membranes [22,23,24]. Enhanced THz emission was reported in nanostructured InP layers under excitation by ultrashort laser pulses [24,25]. Evidence has been found for artificially introduced birefringence in porous InP and GaP [23,26,27]. It is well known that bulk III-V compounds possess second order nonlinear optical coefficients several orders of magnitude higher than those of KDP, ADP and other materials used in frequency upconversion. However, the utilization of large nonlinear susceptibilities of III-V compounds has not been possible due to high dispersion and lack of birefringence necessary for phase matching [28]. Enhanced optical nonlinearities accompanied by artificial anisotropy make porous III-V compounds promising for advanced nonlinear optical applications. InP has been shown to be important material for integrated optoelectronic systems due the possibility to integrate active and passive devices on the same substrate [29,30]. Photoelectrochemical (PEC) etching has been shown to be capable of rapid, dislocation-selective, dopant selective, or band-gap selective etchings of GaN and

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 143

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

related ternary compounds [31-34]. In addition, different kinds of nanostructures, such as nanowires and nanocolumns, have been fabricated by PEC etching of GaN thin films [35]. GaN and related ternary alloys are important wide band gap semiconductors for a broad range of applications in high-frequency/high-temperature electronics and visible and ultraviolet emitters and detectors [36,37,38]. Further improvement in device performance hinges, particularly, on the development of micro- and nanoscale patterning techniques. This research is fuelled by the possibility to produce porous coatings by a versatile and cost effective technology of electrochemical etching. Porous semiconductors attract additional attention as a basis for the preparation of a variety of composite materials. Filling pores with different types of materials, e.g. with metals, polymers, or other semiconductors etc., is an emerging new aspect of porous semiconductors with yet more possible applications. Interweaving the porous substrate with some other material filled into the pores is a novel kind of ―art‖ in the field of producing novel (nano) composite materials [39,40,41] and opens the way to a new branch of (nano) technology and composite materials. Semiconductor-metal nanocomposites show fascinating nonlinear electrical and optical properties and are considered as promising materials for various applications in modern micro- and optoelectronics [42,43]. They may also find applications outside traditional electronics and optoelectronics in areas such as applied electrochemistry and analytical chemistry. Note that the properties of nanocomposites are often dramatically different from those of bulk constituent materials with identical chemical composition and, in this connection, represent considerable interest for basic research. In recent years, considerable progress was achieved in filling in pores in porous Al2O3 with different metals [44,45,46]. Insulating porous alumina film is electrochemically grown on a conductive Al substrate. Because the bottom of the pores touches the conductive substrate, it is relatively easy to fill in such pores by metals using traditional electroplating techniques. The electrochemical deposition of metal starts at the pore bottom, and the process assures continuous growth of metal wires inside the pores, as there are no alternative ways for the current to flow. In contrast to a porous alumina matrix, the semiconductor constituent may have a specific contribution to the formation of the nanocomposite properties. In particular, the properties of the semiconductor matrix may be modified in a controlled way by proper illumination. Apart from that, nanotubular structures can de produced by electrochemical deposition of metals into porous semiconductor templates, in addition to metal nanowires.

2. POROUS III-V AND II-VI COMPOUNDS 2.1. 2D and 3D Porous InP Structures One of the most versatile ways of introducing porosity in semiconductors is electrochemical anodic etching. The generation of holes in n-type semiconductors is a necessary condition for the pore formation process. In general, for n-type semiconductors the holes necessary for anodic etching can be generated by avalanche breakdown mechanism,

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

144

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

i.e. applying a sufficiently high positive potential to the electrode, or, alternatively, by illuminating the semiconductor with photons possessing energies larger than the semiconductor electronic band gap. The scheme of the technological set-up usually used for introducing porosity in III-V and II-VI semiconductors is shown in Figure 1. The anodization is carried out in an electrochemical double-cell. A four electrode configuration is used: a Pt reference electrode in the electrolyte, a Pt reference electrode on the sample, a Pt counter electrode, and a Pt working electrode. The electrodes are connected to a Source Measure Unit (i. e. Keithley 236). The temperature is kept constant with the help of a thermostat (i. e. Julabo F25). The electrolyte is pumped continuously through both parts of the double cell with the help of a peristaltic pump. All equipment involved in the experiments is computer controlled.

Figure 1. Schematic illustration of the double-room electrochemical cell.

The produced pores are classified according to the characteristic size and direction of propagation. According to the International Union of Pure and Applied Chemistry (IUPAC), three categories of pores (not only electrochemically obtained) have to be distinguished if taking into account the average pore diameter and average distance between pores: (i) nanopores, with pore diameters and pore distances smaller than 2 nm; (ii) mesopores, with diameters in the 2 to 50 nm range; (iii) macropores have geometries larger than 50 nm. According to the direction of propagation, two types of pores can be introduced in III-V semiconductor compounds, namely crystallographically oriented or ‗crysto‘ pores, and current-line oriented or ‗curro‘ pores. Crysto pores are usually generated at low anodic current densities due to direct dissolution of the material. They grow along B directions of the zincblende structure and are enveloped by pore walls defined by three {112} crystallographic planes. Exhibiting a triangular-prism like shape, the pores involved reflect the anisotropy of anodic etching inherent to semiconductor compounds. Crysto pores can cross each other and this phenomenon is usually observed in the nucleation layer of anodically etched (100)-oriented substrates [47]. As to curro pores, they are generated at relatively high anodic current densities, their growth being mediated by oxide formation and its dissolution at the pore tip [47]. Curro pores are oriented along current lines and they show

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 145 usually circular shape, independent on the crystallographic orientation of the substrate. A fascinating property of the curro pores is their long-range interaction in the process of formation, leading under certain conditions to self-arrangement of pores in two-dimensional hexagonal closed packed lattice [47,48]. One of the most fundamental questions about pore formation is the mechanisms of pore growth. Different models, most of them developed only for Si, have been proposed, trying to answer these questions. The current burst model (CBM) is one of them. In contrast to other models, CBM assumes that the processes which take place at the nanometer scale during the dissolution, i.e. at nearly atomic level, are similar for all semiconductors, whereas the macroscopic behaviour of different electrodes is determined by the interaction (in space and time) of these nanometer events. This allows the current burst model to have a high prediction power for different semiconductors. The general assumptions of the CBM are the following: (i) a mechanism for a local oscillator is required, the local oscillators being called ‘current bursts‘; (ii) a synchronization mechanism between the current bursts must exist in order to see macroscopic electrode oscillations; (iii) a desynchronization mechanism is required as well, otherwise the electrodes would either oscillate strongly or not at all. A current burst is assumed to be composed of four main steps: direct dissolution, oxide formation, oxide dissolution and surface passivation. The repetition of these steps in time will result in a local ON/OFF oscillatory behaviour of the current. These implies that charges are mainly flowing through a current burst during direct dissolution and oxide formation (ON), whereas no or little charge is flowing (OFF) during oxide dissolution and passivation. The synchronization between the current burst is due to oxide overlapping of neighbouring current bursts, i.e. the surrounding current bursts will contribute with their oxide bumps to the oxide bump of a newly nucleated current burst. This way, its stopping point is already much closer to the stopping point of the surrounding current bursts, because it has to produce less oxide as if it would have been alone, i.e., without neighbours. For a realistic model, a desynchronization mechanism is necessary as well. Once a current burst starts, the current density is locally increased, leading to increased ohmic and diffusion losses, which locally reduces the potential across the oxide layer. This reduces the electric field strength in the neighbourhood of an active current burst, and therefore reduces also the probability for nucleation of a new current burst next to an active current burst [49]. This process can be regarded as desynchronization. Due to the fact that the nucleation of current burst is a stochastic process, it may happen that the oxide of an old current burst is completely dissolved but there are no new current bursts nucleated meanwhile on the same place. In such a case the passivation process of the surface will start. The nucleation probability will depend on how strongly the new surface was passivated by different species from the solution, i.e. the higher the passivation the lower the probability that a current burst will nucleate on this surface. The passivation of the surface in time is called "aging". Since the nucleation probability is larger for smaller degrees of passivation, and passivation in turn reaches minimum just after the oxide is removed, there is an intrinsic coupling of new current bursts to old ones. This will result in clusters of current bursts, and one possible consequence will be the formation of pores. Passivation in this context means the removal of mid-gap states. In III-V compounds the dissolution process is divided in several reaction steps. After the first reaction step, a high

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

146

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

density of mid-gap states is generated. If these mid-gap states are easily passivated by the species in solution, then the next reaction steps cannot occur and dissolution will stop. In other words, the probability for current burst nucleation will decrease if passivation increases. The shape and growth velocity of the pores depend on the processes which dominate the current of a burst: direct dissolution or dissolution via oxide formation. A porous structure containing crystallographically oriented pores produced by anodization of a (100) oriented n-InP sample with the electron concentration of n = 1017 cm-3 at low current densities in a 5 % HCl electrolyte is illustrated in Figure 2a. A big number of triangles is observed on the (011) cleavage plane of the sample, which demonstrates the formation of triangular-prism like pores.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 2. a) Crystallographically oriented pores in (100) n-InP, n = 1017 cm-3, anodized in 5 % HCl aqueous electrolyte; anodization has been performed at low current densities (4 mA/cm2). b) The switch from curro to crysto pores upon lowering the current density. The n-InP sample was anodized in turn at high and low voltages (7 and 1 V respectively). The upper part of the porous layer shows curro pores, whereas the lower part shows crysto pores.

At high current densities the system is trying to minimize the ohmic losses by minimizing the path for the current [50] and thus "forces" the pores to grow perpendicularly to the surface of the sample or in more scientific terms perpendicularly to the equipotential lines of the electric field in the substrate. Therefore, ‘current-line oriented‘ pores are produced as illustrated in the upper part of Figure 2b. A change in the current density from low to high values leads to a switch in the pore growth mechanism. The same (100) n-InP sample was subjected to anodization first at high and then at low current densities. As can be seen in the cross section of the anodized sample, a switch from current-line oriented to crystallographically oriented pores occurs. The upper part of the porous layer exhibits a current-line oriented morphology and corresponds to high current density, while the lower part of the layer contains crystallographically oriented pores and corresponds to low current densities. Taking into account that porosity of the layer with crystallographically oriented pores is lower than for the layer with current-line oriented pores, it can be suggested that successive layers with different porosities can be formed by alternating the voltage from high to low values. According to the current burst model, the dissolution during the formation of crystallographically oriented pores is dominated by direct dissolution and therefore they expose very strong anisotropic features. On the other hand, the strong anisotropic features should partially disappear if the dissolution via oxide formation is dominating the whole

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 147 dissolution process. Thus, looking at the main characteristics of the curro pores, i.e. they have no preferential crystallographic direction for dissolution, it can be concluded that in this case the main dissolution mechanism is the dissolution via oxide formation. Due to the fact that anodic oxides are normally amorphous, the dissolution at the pore tips tends to be more isotropic, i.e. from the electrochemical point of view there is less difference between different crystallographic planes. The uniformity of porous semiconductor structures is "a must" if these structures are intended to be used as photonic band gap materials. The standard way of obtaining highly ordered porous structures in semiconductors is by means of lithography, i.e. by defining a pre-patterned surface and then anodizing it. Pre-patterning in this context means that the wafer is covered by a thin layer of photoresist and by a suitable mask a periodic arrays of small windows are developed in the resist, which will serve as nucleation points for the pores in a subsequent anodization step. This method is extensively used in Si for obtaining uniform 2D porous structures [51]. However, there are particular cases when uniform porous structures can be obtained without pre-structuring, e.g. by self-organization. Taking into account that pore ordering in a self-organization process is strongly dependent on the anodization conditions, for the purpose of optimization it is important to study the degree of ordering in a self-arranged porous structure. It is well known that diffraction patterns (DP), e.g. X-ray, TEM etc., provide a way for studying the structure and uniformity of atomic crystals. Investigating the DP of different materials it is possible to distinguish between perfect, amorphous, textured etc. structures. In fact, a diffraction pattern is nothing else but the image of a crystal in its reciprocal space, i.e. the Direct Fourier Transform (DFT) of the crystal. Using the Inverse Fourier Transform (IFT) it is possible to reconstruct the image of the crystal from its DP. It is not always necessary, however, to use hardware methods like TEM or X-ray in order to generate DPs. DPs of porous structures can be generated numerically, for example by taking DFTs from top-view SEM images. Thus, a DP can be obtained by calculating the direct 2D Fourier Transform from the pixel-arrays of SEM pictures. The way how the DP will look like, i.e. rings, spots etc., will characterize self-organized arrangement of pores. An important feature of the curro pores is that they self-arrange locally in a hexagonal closed packed lattice. Local self arrangement means that domains of nearly perfectly arranged pores can easily be distinguished. However, the curro pores do not start growing in a hexagonal lattice immediately on the surface of the sample. The hexagonal arrangement will be always preceded (at least in HCl aqueous electrolytes) by a nucleation layer (NL). The NL layer is also porous and is made of crysto pores. The width of the nucleation layer is normally up to several microns and depends on the electrolyte concentration and voltage/current applied to the sample. It increases slightly by increasing the electrolyte concentration and by decreasing the voltage/current. As shown in [47,48], multiple branching of a primary pore in the nucleation layer results in a whole set of secondary pores oriented along crystallographic directions B. The end points of the set of pores originating from the same root nucleus form a linear domain and serve as the nuclei for a corresponding domain of curro pores (Figure 3). Thus, in case the nucleation layer is well developed, there is a general tendency of current line oriented pores to form rows oriented along direction. This tendency accompanied by the repulsive pore-pore

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

148

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

interaction due to overlapping space charge regions surrounding neighboring pores leads to the observed ordered close packed 2D distribution of curro pores.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 3. Schematic outline of the development of the porous structure in n-InP during anodic etching (see text for details).

In order to study the current-line oriented pores and their arrangement into the depth of the sample, the NL should be removed. The NL layer can be removed chemically or mechanically. Figure 4a presents a SEM picture taken immediately from the surface of the anodized sample, whereas the inset shows the frequency domain pattern, i.e. the diffraction pattern, generated by 2D FFT analysis. It is obvious that the frequency domain pattern is a highly diffuse spot, which undoubtedly is a characteristic of a highly random arrangement of nucleated pores. However, the DP changes if the FFT is taken from the top of a current-line oriented layer (see for example Figure 4b) after the nucleation layer has been removed. In this case, the frequency domain pattern is a diffuse ring. Interestingly, the ring exposes slight variations in intensity along its perimeter. Namely, six regions with a slightly higher intensity can be distinguished. This means that there are six directions along which the pores are arranged more preferentially, i.e. the pores tend to expose a long-range six fold symmetry. Thus, from Figure 4b follows that the current-line oriented pores have a significant improvement in their arrangement as compared to crystallographically oriented pores nucleated immediately on the surface of the sample. The self-arrangement of the current-line oriented pores can be further improved by optimizing the etching conditions. Figure 4c shows a top view SEM picture taken from a sample anodized under optimized conditions. The FFT pattern, taken from a larger area of the porous sample, shows evident spots exhibiting clearly a six-fold symmetry of a closed packed structure. The presence of spots instead of rings in the frequency domain pattern undoubtedly proves the long-range order, or in other words proves that the structure tends to be "single crystalline". In spite of the fact that the monocrystalline porous structure is still not perfect, i.e. it possesses a large number of defects which make the spots to be more or less diffuse, this is a self-arranged long-range ordered 2D porous structure. The anodization conditions must be carefully optimized in order to reach a long-range order. The optimized conditions mainly depend on the electrolyte concentration, substrate doping level and the anodization voltage. For example, for (100) InP, n = 1018 cm-3 and 5 %

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 149

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

HCl aqueous electrolytes, the long-range order is obtained at approximately U =6 V. Amazingly, at anodization voltages higher that the optimum the long-range order disappears (Fig. 4d). More than that, the intensity of the ring in the DP is uniformly distributed along its perimeter. Thus, at voltages higher than the optimum the tendency to long-range order disappears completely.

Figure 4. Pore formation in (100) n-InP, n = 1018 cm-3 anodized in 5 % HCl at different voltages. The insets show Fourier transforms. a) Nucleation layer exhibiting crysto pores; the structure is amorphous; b) sample anodized at U = 5 V, i.e. lower than the optimized voltage. The corresponding 2D FFT image which is composed of a diffuse ring exposing six slightly higher in intensity spots, i.e. a tendency to a long-range six fold symmetry; c) sample anodized at an optimized voltage of U = 7 V. The six fold symmetry is easily observed in the FFT image, i.e. instead of diffuse rings this time spots along definite directions are present. This is a clear indication that the structure has a long range order, i.e. is a single crystalline structure but with a high density of defects; d) sample anodized at U = 8 V, i.e. above the optimized voltage U = 7 V. The long range order is destroyed again, i.e. the FFT image is a diffuse ring which is a characteristic of amorphous structures.

One of the ‘forces‘ leading to a single crystalline self-arrangement of pores is the nucleation layer. The nucleation layer has a strong influence on the whole porous structure if it is taken into account that the nucleation layer consists of crystallographically oriented pores, i.e. the crystallographic nature of the NL can determine the global ordering of curro pores. The 2D long-range ordering of current-line oriented pores in InP results from an interplay of two independent phenomena: (i) the overlap of the space charge regions of neighbouring pores induces a next neighbour repulsing force, leading to a medium range ordering and locally closed packed pore array; (ii) the nucleation layer of crystallographically oriented pores induces a global orientation for all domains of current-line oriented pores. If one of these two factors is not sufficiently strong, then no long-range order can be obtained. For example, if the anodization voltage is too low, the space charge region of the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

150

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

pores will be also smaller, i.e. the pores will not interact sufficiently strong with each other in order to generate a local closed packed structure. This insufficient interaction at lower voltages is proved also by the triangular shapes of the curro pores (see Figure 4b). On the other hand, if the voltage is too high the thickness of the nucleation layer decreases, i.e. the domain size decreases, and as a result the disorder in the structure increases (see Figure 4d). This is actually what was observed experimentally: there is an optimum voltage assuring long-range ordering of pores. The current-line oriented pores can form a two dimensional (X, Y directions) periodical structure, however the Z direction is still non-periodic. The sharp transition from current-line oriented pores to crystallographically oriented pores can be used to induce periodicity in the third dimension as well. In order to change the degree of porosity as a function of depth and thus to make the structure periodic in the Z direction, the anodic etching can be periodically switched on and off, e.g. from j = 600 mA/cm2 (duration t1 = 0.1 min) to j = 0 mA/cm2 (duration t2 = 0.5 min). It is obvious that etching occurs only during the first 0.1 min of the cycle, while in the last 0.5 min no dissolution is possible since no current is flowing. Thus, periodically pulsing, the current leads to concomitant switches from crysto pores to curro pores and vice versa. The result is a stack of porous layers with different porosities and consequently different effective refractive indices. Such structures are called Bragg-like structures (Figure 5).

Figure 5. Cross-sectional SEM micrograph of a porous n-InP Bragg-like structure with spatially modulated degree of porosity. The current was periodically switched from j = 600 mA/cm2 (duration t1 = 0,1 min) to j = 0 mA/cm2 (duration t1 = 0,5 min).

A completely different morphology can be produced if a pulsed current/voltage is applied instead of sinusoidal one. The pulse is applied in two phases: the first phase with the duration t1 and the second phase with the duration t2 (see Figure 6a). If both phases are of low current/voltage, then no essential changes in the morphology occur, i. e. crystallographycally oriented pores grow in both phases. On the other hand, if the first phase is of high voltage, and the second one is of low voltage, a structure with the morphology illustrated in Figure 6b is formed. As one can see, in this case a switch from crysto pores to curro pores occurs. However, this commutation does not result in a flat transition between the layers, but a wavy interface is produced.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 151

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 6. a) Two phases of the pulse. Each phase has an own duration and amplitude; b) wavy Bragg structures, phase 1 U = 7 V, t1= 0.04 min, phase 2, U = 1 V, t2= 1.5 min; c) A schematic representation of the mechanism of wave front formation.

A schematic representation of the mechanism of wave front formation is presented in Fig. 6c. The crysto pores propagate in the B directions forming an angle of 109o between them. Since the crysto pores grow during the low voltage phase, the density and the branching rate of pores is low. Respectively, the number of pores and the porosity is very low. In the second phase, the curro pores are initiated only at the tip of crysto pores and tends to occupy the space, growing in radial directions from the points of nucleation. As a result, a domain type structure is produced. Anodization of n-InP in aqueous solution of NaCl proves to be an effective and environmentally-friendly approach for the purpose of uniform nanostructuring and fabrication of semiconductor nanotemplates with ordered distribution of pores [52]. Figure 7 illustrates the optimization of technological process for crystalline (100)oriented substrates of n-InP with 500 µm thickness and free electron concentration of 1.3 x 1018 cm-3. The anodic etching was carried out in NaCl aqueous solutions of different concentrations in potentiostatic regime with various applied voltages at different temperatures. To optimize the electrolyte concentration, the anodization was performed at 20 o C under the applied voltage of 5 V in NaCl electrolytes with the concentration ranging from 0.25M to 4M. It was found that under these technological conditions cylindrical current-line oriented pores are formed perpendicularly to the sample surface. The diameter of pores decreases with increasing the electrolyte concentration. For instance, the pore diameter decreases from 200 to 100 nm with increasing the electrolyte concentration from 1M to 3.5M. Sections (a) and (b) in Figure 7 demonstrate the possibility to control the diameter of pores by the variation of the applied voltage. One can see from the section (a) that the anodization under the applied voltage of 5 V results in self-arrangement of pores with average diameter of 100 nm in two-dimensional hexagonal closed packed lattice. The increase of the applied voltage to 7 V leads to increase of the diameter of pores to 200 nm, disordering in the pore arrangement, and increase of the fluctuation of the pore diameter. The morphology of the prepared porous template is also controlled by the temperature of the electrolyte. The decrease of the electrolyte temperature from 30 to 5 oC leads to the increase of the pore diameter up to 150 – 200 nm and the wall thickness up to 70 – 100 nm, accompanied by deviation of the pore shape from a cylindrical one and total disorder in the pore arrangement, as evidenced by comparison of sections (c) and (d) in Figure 7.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

152

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Figure 7. (100)-oriented n-InP anodized in 3.5M NaCl electrolyte at 20 oC and U = 5 V (a) and 7 V (b), and at 30 oC (c) and 5 oC (d) at U = 5 V.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Apart from manufacturing of 2D structures, anodization in salty water proves to be a powerful tool for manufacturing 3D ordered porous structures. To fabricate a 3D periodic structure, the anodic etching was carried out under quasi-potentiostatic conditions, the value of the applied voltage being periodically modulated with time in three steps, as illustrated in the insert of Fig. 8a. At the beginning of anodization, periodically pulsing current gives rise to concomitant switches from crysto to curro pores and vice versa, see Figure 8a. With the time, however, the morphology exhibits more and more distinct wavy entities. As a result a new architecture of pore distribution is reached. As one can see from Fig. 8b, the porous structure consists of periodically distributed domains exhibiting crysto pores, these domains being surrounded by submicrometer regions possessing curro pores. Unfortunately, the structure is rather fragile that impedes the fabrication of a high quality cleavage in the porous region.

Figure 8. (a) SEM image taken from a porous InP sample anodized under periodic modulation of the applied voltage with time as illustrated in the insert; (b) enlarged view of the bottom part of the porous layer. Reproduced with permission from [52], I.M. Tiginyanu et al., Phys. Stat. Sol. RRL 1, 98 (2007). © 2007, Wiley-VCH Verlag GmbH & Co.KGaA.

2.2. Peculiarities of Producing Porous Coatings in GaP A variety of porous morphologies can be produced also in GaP substrates, ranging from triangular structures with characteristic size of 100 nm, as shown in Figure 9a, to complex porous networks with characteristic size of 1 – 2 µm, as presented in Figure 9b. The obtained

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 153 morphologies are reproducible and controlled by the technological conditions of anodization. However, the technological parameters must be adjusted individually for substrates with different parameters.

Figure 9. GaP porous templates with different degrees of porosity.

In n-GaP substrates both crystallographically oriented and current-line oriented pores can be produced. Figure 10 presents the anodization results obtained at low (a) and high (b) current densities for (100)-oriented n-GaP samples. As it can be observed from Figure 10, at

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

low current density the angle between the pores equals 109 o , which is a confirmation of the fact that also in the case of (100)-oriented n-GaP the pores are oriented along B directions.

Figure 10. a) Crystallographically oriented pores in (100), n-GaP, n = 1017 cm-3, anodized in 5 % H2SO4 aqueous electrolyte. Anodization has been performed at low current densities (0.5 mA/cm 2). b) Current pores in n-GaP ( j = 80 mA/cm2, 5 % H2SO4, t = 30 min).

At high current density, as in the case of n-InP, the pores do not show any preferential crystallographic directions of growth and simply grow perpendicularly to the initial surface of the sample (Figure 10b). Thus, the pores obtained in n-GaP at high current densities can be also catalogued as current-line oriented ones. Nevertheless, there are some major differences between the current-line oriented pores in n-InP and n-GaP. First of all the current-line oriented pores in n-GaP are much more unstable (Figur 10b). Secondly, the pore diameters are much bigger in GaP, in the range of 3-4 µm, whereas for the same doping level the diameter of the current-line oriented pores in n-InP are in the submicron range. Apart from that, the ratio pore-diameter/pore-wall-width in GaP is much bigger than in InP. These differences can be explained assuming stronger passivation efficiency for the pore walls and probably also weaker oxidation and less qualitative native oxide in InP as compared to GaP.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

154

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

According to the current burst model, less oxidation in InP means smaller pores. Additionally, in order to break the more effective passivation and consequently to dissolve InP, a stronger electric field at the pore tips is necessary, i.e. a smaller radius of curvature, which is directly related to smaller pore diameters. On the other hand, in GaP the passivation at high current densities is not so efficient, therefore the electric field inside the space charge region can be lower, i.e. the radius of curvature and consequently the pore diameters can be bigger. Also, low pore wall passivation leads to smaller space charge region in the semiconductor and consequently to thinner pore walls. In order to obtain a more uniform nucleation of pores and to grow pores of cylindrical shape oriented perpendicular to the sample surface it is necessary to add a small amount of HCl to the aqueous H2SO4 solution (5 ml of 35 % HCl to 200 ml of 5 % H2SO4). This leads to the increase of the pore wall passivation in GaP and to the growth of much more stable pores. Apart from that, the addition of HCl to the aqueous H2SO4 solution leads to the decrease of the pore diameter. The increase of the electrolyte temperature during anodization was found to improve the porous sample morphology and to result in a more uniform arrangement of pores. In this conditions GaP templates with quasi-ordered arrays of pores can be produced, comparable to those obtained on n-InP (see Figure 11).

Figure 11. Templates with 2D arrangement of pores in GaP (a) and InP (b).

2.3. Etching Pores in GaAs Substrates in Acid and Neutral Electrolytes Like in InP substrates, etching of GaAs at low current densities leads to the growth of crystallographycally oriented pores. Figure 12a shows a cross-sectional SEM image taken from a (100)-oriented GaAs substrate cut from the liquid encapsulated Czochralsky-grown Si-doped ingot with a free electron concentration n = 5x1018 cm3 anodized at j = 54 mA/cm2 in 5% HCl aqueous electrolyte.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 155

Figure 12. (a) SEM cross-sectional view of anodically etched n-GaAs in 5% HCl aqueous electrolyte at j = 54 mA/cm2. Directions of pore growth make an angle of 54° with the direction perpendicular to the surface. (b) Intersection of [111] and [111] oriented pores in anodically etched (100) n-GaAs. Reprinted with permission from [61], S. Langa et al., Appl. Phys. Lett. 78, 1074 (2001).© 2001, American Institute of Physics.

As can be observed, the angle between the directions of pore growth (marked by arrows) and the direction perpendicular to the surface ([100] direction in this case) is 54°. Thus, the two directions of pore growth can be assigned to [111] and [111] ones. The intersection of

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

these [111] and [111] oriented pores is most likely the reason for the formation of the visible terraces as a result of the cleavage. Apart from that, the analysis of the SEM micrograph presented in Figure 12a shows the existence of sets of pores oriented along [111] and [111] crystallographic directions. To control the architecture of 3D porous structures, etching conditions which provide crossing of pores and their continued growth along the initial directions must be maintained. Figure 12b illustrates an example of the intersection of two pores oriented along [111] and

[111] crystallographic directions. Somewhat surprisingly, the intersection has no influence on the pore shape, the size, and the direction of subsequent propagation. This important finding demonstrates that anodic etching may be a suitable and unique tool for the production of 3D micro- and nanostructured III–V compounds, e.g., for photonic crystals applications. Nevertheless, in addition to pore interconnection a photonic crystal requires a very high level of uniformity [53]. Since the dissolution is stimulated by defects [54,55] and/or illumination [56], it is expected that a predefined nucleation and thus a controlled intersection of the pores can be provided by conventional lithographic means followed by a conventional chemical treatment to generate small pits (defects) or by proper front side illumination of the sample at the beginning of the anodization. The crossing of the pores is somewhat unexpected because according to the existing models [56–58] the formation of pores in n-type semiconductors is a self-adjusted process controlled by the distribution of the electric field at the semiconductorelectrolyte interface. The pores may branch and form porous domains, but both individual pores and domains should be separated by walls with characteristic dimensions of twice the thickness of the surface depletion layer [10, 55]. These observations have been argued supposing that when a wall becomes too thin, it can no longer support a field perpendicular to

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

156

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

the surface which is sufficiently high for anodic hole generation, so that etching stops [10,59]. On the other hand, crossing suggests that, independent of the distance separating two neighboring pores, there are sufficient holes at least to initiate the rate-limiting step of the dissolution reaction. The oxidation intermediates produced by this initial reaction are usually considered as surface states with energy levels above the valence band edge [60] and can be further easily oxidized by injection of an electron in the conduction band by tunneling and do not need anodically generated holes. Thus, in spite of the fact that the field strength decreases when the space charge region of two pores overlaps, it is still high enough to support the dissolution leading to intersection. The observation of crossing pores in single crystalline n-GaAs suggests that the depletion layers may be overcome also in other materials [61]. Our experiments were carried out under conditions of strongly anisotropic etching along guidelines adopted from the ‗‗current burst model‘‘ proposed for pore formation in Si [62]. The strong influence of the crystal anisotropy under those conditions probably plays also a major role in the crossing of pores and subsequent continuation of the pore growth along the same crystallographic directions. In contrast to InP and GaP, no current-line oriented pores have been observed in GaAs even at high current densities. Similarly to n-InP, anodization of n-GaAs in aqueous solution of NaCl proves to be suitable for uniform nanostructuring [63]. Crystalline 500 µm thick (111)-oriented substrates of Si-doped n-GaAs with free electron concentration of 2 x 1018 cm-3 were subjected to anodic etching at T = 23 oC in 3.5 M NaCl aqueous solution. Figure 13a presents a cleavage of a porous sample produced by anodic etching of n-GaAs substrate in three dissolution steps with the duration of 5 minutes each, with the current densities of 75, 50 and 25 mA/cm2, respectively. One can see from the image the formation of three porous layers with the thickness of about 40 µm, the degree of porosity being dependent upon the anodization current. Triangular-prism like pores are clearly seen in the top view of the sample shown in Figure 13b, therefore suggesting the formation of ‗crysto‘ pores.

Figure 13. (a) SEM image taken in cross section from a porous GaAs sample produced by anodic etching in 3.5 M NaCl aqueous solution in three dissolution steps with the current densities of 75, 50 and 25 mA/cm2. (b) Top view of the sample.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 157

2.4. Electrochemical Preparation of InAs Nanostrucures

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

InAs is a direct band gap semiconductor with zinc-blende structure and a narrowest energy gap among the III-V binary compounds, except for InSb. This material is of great interest as compared to other III-V materials since it is characterized by the largest exciton Bohr radius among III-V compounds which assures quantum confinement effects to be observed at bigger characteristic dimensions of nanostructures. Technological conditions have been elaborated for a controllable introduction of porosity in InAs substrates. One of the key problems with electrochemical methods of introducing porosity in semiconductor materials is the appropriate choice of the electrolyte composition. This problem is solved individually for each material. As shown in previous paragraphs, aqueous acid solutions such as HF, HCl, or H2SO4 are usually used for electrochemical nanostructuring of semiconductors. Experiments with anodization of InAs substrates in these electrolytes demonstrated dissolution of the material without the formation of homogeneous nanostructures at any concentration of the electrolyte and electrical parameters applied, but the rate of dissolution increased with the increase of the electrolyte concentration and the voltage applied. Two approaches have been applied to solve this problem. The first one is the search of an appropriate electrolyte. The second one is the anodization under pulsed applied voltage. The application of pulsed anodization with traditional HF, HCl, or H2SO4 electrolytes was not successful, it resulting in just the decrease of the dissolution rate. The dissolution rate decreased with decreasing the width of the pulse. It was found that anodization in aqueous solution of H3PO4 or HNO3 acids leads to the nanostructuring of InAs surfaces. However, the obtained structures are inhomogeneous. Using a mixture of these electrolytes improves the homogeneity. The optimum solution was found to be a 1:1 mixture of H3PO4 and HNO3 aqueous solution.

Figure 14. (a,b) SEM image of InAs samples etched under the applied voltage pulses of 15 V amplitude in a solution comprising 10 ml H3PO4, 10 ml HNO3, and 300 ml water with the addition of 0.2 g (a) and 3 g (b) of potassium dichromate. (c,d) SEM image of InAs samples etched with applying voltage pulses of 7 V (c) and 5 V (d) amplitude in a solution comprising 10 ml H 3PO4, 10 ml HNO3, 0.2 g of potassium dichromate, and 300 ml water.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

158

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Pulsed anodization with this electrolyte demonstrated the formation of sharp structures in the form of micro- and nano-pencils as illustrated in Figure 14. The structures were produced on a (111) oriented n-InAs substrate with carrier concentration n = 3x1017 cm-3 under pulsed anodization with the pulse width of 10 s and frequency of 10 Hz. Two possibilities were found for the control of the dimensions and the density of the obtained structures. The first one is the addition of potassium dichromate to the electrolyte composition (as illustrated in Figure 14a,b), while the second one is the change of the electrical parameters applied (as illustrated in Figure 14c,d). The research demonstrated that apart from the above mentioned acid electrolytes a neutral aqueous NaCl solution is also suitable for the porosification of InAs substrates. Figure 15 compares the InAs morphologies produced under pulsed anodization with the voltage amplitude of 15 V in a H3PO4 + HNO3 electrolyte (a) and a NaCl (b) electrolyte. One can observe the formation of similar triangular structures. These structures demonstrate sharp tips when the image is taken at a certain angle, as illustrated in Figure 14.

Figure 15. Top view SEM image of InAs samples anodized in acid (a) and NaCl (b) electrolytes under pulsed conditions.

In contrast to structures obtained under pulsed electrochemical treatment, anodization under DC conditions results in the formation of columnar structures without sharp tips at the surface, as shown in Figure 16.

Figure 16. Cross sectional SEM image of InAs samples anodized in acid (a) and NaCl (b) electrolytes under DC conditions. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 159

2.5. Electron Field Emission from InAs Nanostructured Surfaces Narrow band gap semiconductors such as InAs, InSb, -Sn as well as carbon nanotubes and a set of graphite-enriched carbon and diamond-like carbon (DLC) materials are of great interest from both fundamental and applicative points of view. Some of these materials are widely used in optoelectronic devices for IR spectral range, resistive gas sensors, laser media etc. Nowadays, in connection with the growing role of vacuum micro- and nanoelectronics, it becomes important to search for narrow band gap semiconductors as promising materials for high efficiency electron field emission (EFE) cathodes. It is known that a well conductive carbon-mixture of nanotubes and nanoplates demonstrates very good EFE properties [64-67], and it is a promising material for the fabrication of large areas flat displays [67]. In this case the micro- or even nanosize tip cathodes have been used. Recently we proposed to use InAs, a narrow band gap semiconductor compound with high electron mobility, as a stable and rather perfect material with a well-developed technology of crystal growth and nanotexturing for the fabrication of effective electron field emitters [68]. III-V semiconductors are characterized by multi-valley structure of the conduction bands (C-bands) with rather different energy separation of the valleys in the energy scale. Such complicated band structure can be evidenced in the field emission phenomenon. The energy barriers for electron field emission from the main and satellites valleys are different. In spite of low energy barriers of the upper valley, its input in emission current can be remarkable only after hot electron filling. The barriers for electron emission can be estimated from the relation [69]:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

0 + Eg  5.5 -  Ev,

(1)

where electron affinity 0 is the energy interval (barrier height) between bottom of the conduction band and vacuum level, Eg is the band gap and Ev is a small correction of ~ -0.5 eV to the valance band variation [69]. In case of quantum size structures, the increase of Eg takes place, and this negatively influences the electron field emission due to the decrease in the free carrier concentration. On the other hand, a positive impact of quantization appears due to the decrease in the electron affinity 0 and intervalley distance Ec1-Ec2. These results were obtained theoretically on the base of the model of spherical quantum dots. InAs with highly textured surface was used to study the peculiarities of the electron field emission from multi-valley narrow band gap semiconductors. Figure 17 presents the results of calculations of the lowest direct conduction band bottom Ec1 (-valley) and the indirect ―satellites‖ conduction band bottom Ec2 (L-valley) energy positions in relation to bulk material as a function of the InAs QD diameter d. The energy separation between the Ec2 and Ec1 conductance band bottoms is about 0.73 eV for the bulk InAs material (insert in Figure 17) and continuously decreases with the decrease in the QD size d down to 0.4 eV for smallest QDs (Fig. 17). For comparison, the ∆E12 intervalley distance for small GaAs quantum size (QS) structures is less than 0.2 eV, while in nanostructured GaN layers it reaches much higher values of 0.7 – 1.0 eV [70,71,72]. The energy of the Ec1 band sharply increases with the decrease in the QD diameter due to the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

160

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

small effective mass of electrons in the lowest conductance band valley (m*=0.026 m0), while the higher value of the electron effective mass in the upper L-valley (m*= 0.64 m0) leads to a smaller increase in the Ec2 band energy (Figure 17).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 17. Calculated dependences of the conductance band bottoms taking into account the barrier height variation (1 is the main valley in the  point; 2 is the satellite valley in the L point) and intervalley distances (3) as a function of the QD diameter. Inserted is the energy band structure of bulk InAs. Reprinted with permission from [68], V. Litovchenko et al., Semicond. Sci. Technol. 22 1092 (2007). © 2007, Institute of Physics.

Electron field emission characteristics were measured in vacuum of 210-7 mbar. The InAs samples with nanostructured surface illustrated in Figure 18 were used as cathodes. A highly doped n-type Si wafer was used as an anode electrode. A 7.5 m kapton spacer defined the distance (L) between the InAs emitter and the Si anode. The macroscopic electric field (E) can be determined as E=V/L, where V is the applied voltage. The hole in the spacer defined the emission cathode area with A= (500 m)2. The set-up allows one to reliably measure emission currents at the level of 510-11 A. The voltage was increased in steps of 0.2 V when measuring I(V) characteristics. The high value of the current was limited at 10-6 A in order to protect the emitting nanotips against overheating and melting.

Figure 18. SEM image of the InAs nanostructured surface used in filed emission experiments. Reprinted with permission from [68], V. Litovchenko et al., Semicond. Sci. Technol. 22 1092 (2007). © 2007, Institute of Physics. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 161 The position of both valleys of the conduction band can be increased by a factor of two in relation to bulk material by nanostructuring as shown above (Figure 17). An even larger increase in Ec is predicted for the infinite barrier heights (dotted lines) or for large values of the dielectric constant of the surrounding media (the increase can reach factors from two to five). Another effect induced by nanostructuring is the decrease in the inter-valley distances by a factor of two or even more in comparison with those inherent to the bulk material (see curve 3 in Figure 17). Crossing of the - and L-valley energies can also take place at the characteristic size of the nanostructure around 8-10 nm in the case of infinite energy barriers as deduced from Figure 17. These predictions may pave the way for the realisation of new types of HF-generator devices such as nano-Gunn oscillators, photo-nano Gunn oscillators, IR-nano-sensors with the variation of the energy band edge. As follows from the above presented relation (1), the performed calculations allow one to deduce the dependence of the barrier heights () for both valleys upon the characteristic size of the nanostructure. The calculations indicate a substantial decrease of the barriers for both valleys (main and satellite ones). This is particularly important for a remarkable enhancement of the field emission current and photo-emission in various types of solid state barrier devices. The electron field emission is described according to the Fowler-Nordheim (F-N) equation

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

I  AV 2 exp[ 

B 3 / 2 ] V

(2)

where I is the emission current, V is the applied voltage,  is the work function,  is the electric field enhancement coefficient, A and B are constants. As one can see from the equation (2), the decrease of the barrier leads to an exponential increase of the field emission current. The values of initial barriers in InAs are equal to 1 = 4.7 eV and 2 = 4.0 eV [70] (where 1 and 2 are the work functions at emission from -valley and L-valley correspondingly). The decrease of these values by 1 eV is realized at characteristic sizes of the nanostructure around 1-3 nm. While this decrease results in a significant increase of the emission current, the change of the slope of the F-N curves is not so significant, it being of the order of (1/2‘)3/2 ~ (4.7/3.7) 3/2, i. e. the change is by a factor of less than 1.5. Another factor influencing the F-N equation is the enhancement of the electric field on the tip of the field emitting cathode. In the approximation of floating sphere it can be presented as [73,74]



h 3 r

(3)

where h is the tip height and r is the curvature radius of the tip. The scanning electron micrograph of the nanostructured InAs surface (Figure 18) demonstrates a homogeneous cone-like texture with the height of the cones h ~ 0.5-1 m and High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

162

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

with sharp tops of ~ 100-200 nm. The thorough analysis of the micrograph suggests the presence of two characteristic sizes of the nanostructures. About 85-90 % of the nanostructures consist of cones with the radius of the top r ~ 100-200 nm, while a small part of the area (~ 10-15 %) is covered by ultra-small tips with the radius of the top r ~ 30-50 nm. Due to such relatively large size of the top diameter we don‘t expect the influence of the quantum size restriction effect on energy band reconstruction and, correspondingly, on peculiarities of electron field emission. Really, as can be seen in Figure 17, the significant energy band reconstruction is observed only at d < 10 nm. The explanation of complex shape of the emission I-V characteristics (Figure 19) is the following. At relatively small applied voltages, the field emission comes from tips with larger radius of the tops, which dominate the surface structure. However, in spite of the high efficiency of emission, the integral value of the current is low at these voltages (< 510-9 A). Three regions with different slopes of the I(V) dependence are clearly seen in Figure 19. The straight line with a large slope of 1.46104 Vdec at low voltages is followed by a line with a smaller slope of 4.0103 Vdec at higher voltages, i.e. the slope decreases by a factor of ~ 3.5. A clear tendency to saturation accompanied by current fluctuations is also observed at high voltages. -12

10

-12

10

-13

10

-14

10

-13

10

-15

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

10

-14

10

-16

10

2.1

-3

2.4

1/V x 10 , 1/V

2.7

2.1

2.4 -3

1/V x 10 , 1/V

Figure 19. Fowler-Nordheim dependence of the electron field emission from nanostructured InAs surface (a), and a series of repeated measurements of the current (1-3) in the high voltage region (b). Reprinted with permission from [68], V. Litovchenko et al., Semicond. Sci. Technol. 22 1092 (2007). © 2007, Institute of Physics.

The observed large change of the I(V) characteristics slopes in case of InAs cannot be explained only in the frame of the model of hot electron redistribution due to electron transitions from the low valley (in the  point) to the upper L-valley, as was previously proposed for the electron field emission from GaN nanotextured surfaces [74]. Such intervalley electron redistribution, as was mentioned above, can lead to a change of the slope by a factor equal to about 1.3. We suggest that the large change of the slope is also related to the presence of two characteristic sizes of the nanostructure as estimated from SEM micrograph (Figure 18). The two types of cones with different radii of the tip are characterized by two different values of  parameters. The values of the electric field enhancement coefficients are 2  42, 1  15, and 2/1  2.8. One can conclude that ~ 1/3 of the contribution to the slope changing comes from the inter-valley electron redistribution, while another ~ 2/3 of the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 163 contribution is related to the electric field enhancement coefficient. The estimation of the values of the work function for both valleys using the slope of the FowlerNordheim and 1, 2 gives: 1  4.7eV ,

 2  3.9eV , and the difference of

1   2  Eg1  Eg 2  0.8eV . These results are in good agreement with the literature data for bulk material [70]. The parallel shift of I(V) curves in the high-voltage region with some current lowering observed at repeated measurements can be caused by possible tip destruction and removing the surface native oxide due to bombardment of residual gas ions or heating.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.6. Nonlinear Optical Effects in InP and GaP Nanostructured Surfaces It is well known that bulk III-V compounds possess second order nonlinear optical coefficients several orders of magnitude higher than those of KDP, ADP and other materials used in frequency upconversion. However, the utilization of large nonlinear susceptibilities of III-V compounds has not been possible due to high dispersion and lack of birefringence necessary for phase matching [28]. As mentioned above, electrochemistry proved to be a powerful tool for introducing the necessary optical anisotropy in semiconductor materials. Anodically etched Si, for instance, was found to exhibit anisotropy in the infrared and visible regions [75,76,77]. The measured birefringence, defined as the difference in the effective refractive indices of the electric fields polarized parallel and perpendicular to the pore axis, reaches a maximum value of 0.366 at the wavelength  = 6.52 µm [75] which exceeds the birefringence of quartz by a factor of 43. Golovan el al [78] used the anodization-induced birefringence to achieve phase matching of SHG in porous Si films. On the other hand, crystallographically oriented pores were used to fabricate semiconductor sieves of gallium phosphide, i.e., two dimensionally nanostructured membranes exhibiting a strongly enhanced optical second harmonic generation in comparison with the bulk material [22]. In this chapter it will be shown that enhanced optical nonlinearities accompanied by artificial anisotropy make electrochemically etched III-V compounds promising for advanced nonlinear optical applications. In the framework of the effective medium theory a cubic semiconductor with randomly distributed air-filled pores all aligned in z-direction, is considered as a homogeneous uniaxial material with an effective dielectric tensor semic. air e (e ,e ,porosity) ik

(4)

and an effective non-linear optical tensor semic.  ikl  123 ,  air  0, porosity 

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(5)

164

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu In the case of a (111)-surface the cubic point group T is changed to C , in which case d 3V

 ikl

has three independent components

[28]. Following Hui and

 111 e.g. is given by

Stroud [79]

1 3

semic .  111   123 6 a 22 a1  a13

where ...

 111 ,  113 , and  333

is the average

SV

(6)

SV

1 dV... over the volume occupied by the semiconducting Vs VS

material.

a1, 2 (r )

The

a1, 2 (r ) 

where

E (r )

are given by

E x , y (r )

is the electric field inside the sample (calculated in the approximation of the

linear medium) and

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

E0  E

(7)

E x0

V



E0

is the average over

E (r )

taken over the total sample volume

1 dV E (r ) V V

(8)

The other non-vanishing components of

 ikl have to be calculated from similar

formulae, all containing combinations of third order fluctuation terms of the kind

Ei Ek El

SV

divided by products of the kind Ei0 Ek0 El0 .

An enhanced non-linearity, i.e.

 ikl

semic . components are much bigger compared to 123 ,

requires strong third order field fluctuations. By common sense this seems to be unlikely, because fields in a semiconductor material with high dielectric constant are screened,

E

SV

 E0

Likewise it can be proved exactly that

(9)

E

2

2

 E 0 . However, using the special SV

isotropic model of Bruggeman [80] for a non-linear metal-insulator composite, Bergman [81] was able to prove that fourth order fluctuations E

4 metal

/ E04 even can diverge. Repeating

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 165

this analysis for the dielectric case

I  0 1 , with  M    2 ,

2  10 , the 1

fourth order fluctuations diverge, too. These divergences are connected with and due to a so-called percolation threshold, a relative concentration 0  f 0  1 at which the

 M - respectively the  2

- material forms

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

connected paths through the sample. All simple structure models without such a percolation threshold do not result in large fourth order fluctuations. The Bruggeman model is based on spherical inclusions of both materials in a self-consistently calculated effective matrix. On the other hand, it is well known that the fields near sharp edges are much larger than fields near spherical surfaces, therefore it is believed that non-spherical inclusions (non-cylindrical pores in our case) can result in large third order fluctuations, leading to strong nonlinear effects in such structures. Due to their non-spherical shape the pores growing perpendicular to the surface of (111)B oriented III-V sample in a good approximation can satisfy the conditions implied by the theoretical prediction discussed above. In order to perform polarized "second harmonic generation" (SHG) measurements in transmission geometries the porous layers have to be detached from the bulk substrate, i.e. free standing membranes have to be fabricated. Scanning electron microscope images taken from such a porous InP(111) membrane are illustrated in Figure 20. One can see that most of the pores possess triangular-prism shapes and the lateral size of pores is between 50 and 100 nm. As a fundamental excitation beam for optical measurements, the 1064 nm output of a Q-switched Nd-YAG laser was used. To minimize the influence of the laser output fluctuations, the measured SHG intensity has to be normalized by monitoring the laser intensity in a reference channel.

Figure 20. SEM images taken from a porous InP membrane: a) top view; b) cross-section view. Inset is the enlarged view demonstration of the triangular shape of pores.

Figure 21a curve 1, shows the transmission spectrum of a porous InP(111) membrane exhibiting parallel pores with triangular-prism like shape and transverse dimensions less than 100 nm. As one can see from Figure 21a, the optical transmission spectrum shows pronounced interference fringes in the spectral interval corresponding to quantum energies lower than the band gap of bulk InP (h < Eg = 1.3 eV). The observation of interference fringes is an indication to the optical homogeneity of the porous medium. Thus, due to the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

166

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

relatively small dimensions of both pore and skeleton entities, the porous medium proves to be optically homogeneous and therefore the light propagates through it without pronounced scattering. In case of a non-uniform distribution of pores the membranes become optically inhomogeneous which leads to pronounced light scattering and decrease in transparency (Figure 21a, curve 2). Similar results have been obtained for porous GaP membranes (Figure 21b).

Figure 21. Transmission spectra of optically homogeneous (curve 1) and inhomogeneous (curve 2) porous InP (a) and GaP (b) membranes. Reproduced with permission from [23], I.M. Tiginyanu et al., Phys. Stat. Sol. A 197, 549 (2003).© 2003, Wiley-VCH Verlag GmbH & Co.KGaA.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

In optically homogeneous porous membranes the degree of porosity defines the optical anisotropy caused by the preferential orientation of pores along the crystallographic direction. According to the effective medium theory described above, in the case of pores stretching perpendicular to the initial surface, the components of the dielectric tensor of the porous membrane are determined by the equations (10) and (11).

 // ( )  (1  c)1  c ( )        

(10)

1   2  c   c     1  c       2  c 

(11)

where c is the concentration of semiconductor material, () is the dielectric function of the III-V compound, and

1

is the dielectric constant of air. Due to

  ()   II () for all c,

the porous semiconductor represents a positive uniaxial material. Figure 22a shows the transmission of light with = 1064 nm by a porous membrane with the thickness 8.2 μm as a function of the incident angle of the laser beam. The position of the maxima displayed by the interference patterns depends upon the direction of light polarization. For the ordinary beam, the maxima occur at incidence angles of 17 and 43 degrees, while for the extraordinary beam the maxima occur at 21 and 49 degrees. The analysis of the interference conditions for the two beams taking into account Eqs. 10 and 11 allows one to calculate the refractive indices for ordinary and extraordinary beams: no = 2.43 and ne = 2.67. Thus, it is obvious that the porous membranes exhibit pronounced birefringence necessary for phase matching in optical second harmonic generation.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 167 Figure 22b illustrates the transmitted s-polarized second harmonic signals (

λ2 w  532 nm) from both bulk (111)-oriented GaP and a porous membrane as a function of the incident angle of the s-polarized fundamental beam. Despite of the short coherence length ( Lcoh  1 μm ) it is not possible to see Maker fringes in bulk GaP because of a sufficiently high absorption at the SHG frequency [82]. The pronounced absorption at 2 is caused by the fact that the corresponding energy is higher than the indirect band gap of GaP 2ω  Eg  2.24 eV .

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 22. (a) Transmission of light with =1064 nm by a porous GaP membrane as a function of the incident angle of the laser beam measured in q-s and q-p polarization geometries. (b) Measured spolarized SH intensity as a function of the incident angle of the s-polarized fundamental beam for bulk and porous GaP. Reproduced with permission from [23], I.M. Tiginyanu et al., Phys. Stat. Sol. A 197, 549 (2003).© 2003, Wiley-VCH Verlag GmbH & Co.KGaA.

As one can see from Figure 22b, under identical conditions the porous membranes exhibit a SHG intensity of at least two orders of magnitude higher than that inherent to bulk GaP. This enhancement can be attributed to giant electric field fluctuations expected, according to theoretical estimations, for some porous structures. Another interesting feature is that the fundamental incident angle dependence of the SHG intensity for porous membranes measured in s-s polarization geometry shows pronounced shoulders at -35 and +35o. Figure 23 (solid squares) illustrates the rotational dependence of the second harmonic intensity for an optically homogeneous porous GaP(111) membrane possessing triangularprism like pores. It reflects perfectly the crystallographic features of (111)-oriented GaP demonstrating the high crystalline quality of the porous skeleton. On the contrary, optically inhomogeneous GaP membranes reflect no crystallographic features of the semiconductor compound (Figure 23, solid triangles) since in the case of strong diffuse scattering any dependence of the SHG signal upon the rotation angle of the porous membrane about the surface normal is removed. In optical SHG type I phase matching is achieved if the condition n0  2   ne 0   is

satisfied, where

n0        , ne 0     e 0  

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(12)

168

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

with

1 cos2  sin2     e0 ()   ()  // ()

(13)

where  is the angle between the optical axis and the exciting laser beam inside the membrane. Taking into account that 0=1, ε(ω)  3.1192 2 and ε(2ω)  3.45952 [28], one can solve the equation n0 (2ω)  ne0 (ω) in order to determine  as a function of c. A

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

solution exists for all c < 0.696, it means for the degree of porosity (1 - c)  30 %. Note that membranes with 30 % porosity fulfil the phase matching conditions provided that the fundamental and SHG beams propagate in directions that are nearly perpendicular to the pores.

Figure 23. SH intensity induced by a 1064 nm polarized pump beam at normal incidence as a function of the azimuthal rotation angle of the optically homogeneous (solid squares) and inhomogeneous (solid triangles) porous GaP membranes measured in parallel polarization. The solid line is a fit. Reproduced with permission from [23], I.M. Tiginyanu et al., Phys. Stat. Sol. A 197, 549 (2003).© 2003, WileyVCH Verlag GmbH & Co.KGaA.

Therefore, porosity-based technological approaches prove to be important for elaborating new nonlinear optical elements ready to be integrated in optoelectronic circuits. First of all, the formation of pores leads to symmetry breaking. In particular, pores parallel to the direction in III-V compounds change the cubic crystal symmetry (point group Td) to the uniaxial trigonal one (point group C3v ). The porosity-induced artificial birefringence opens the possibility to meet the phase matching conditions for the second harmonic generation in III-V materials. For gallium phosphide, in particular, the phase matching conditions can be fulfilled for degrees of porosity higher than 30 %. Secondly, porous structures represent heterogeneous media where the electric field undergoes large spatial variations. Especially strong local field fluctuations are expected in structures containing pores with sharp edges, e.g., triangular-prism like pores. Porous GaP membranes with triangular-prism like pores exhibit a SHG efficiency two orders of magnitude higher than that of bulk material. Thus, in spite of the electric field screening in the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 169

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

semiconductor, the third order field fluctuations responsible for the SHG enhancement seem to be giant in such kind of structures. Taking into account existing theories, an important role in the SHG enhancement may be attributed to the material percolation which is responsible also for the mechanical stability and good thermal conductivity of porous membranes. It is interesting to note that new nonlinear optical media can be created just by filling in the pores in porous III-V compounds with other materials. In this case the semiconductor skeleton can be designed to provide phase matching while the material filling the pores will contribute mainly to the SHG. Note that porous III-V compounds as phase-matching matrices are much more promising than elementary semiconductors due to their larger band gap and their more pronounced anisotropy when subjected to electrochemical etching. Optical second-harmonic generation (SHG) was also investigated from bulk and porous InP. The measured azimuthal dependence of the SHG intensity is shown in Figure 24. It is seen from Figure 24 that the second harmonic intensity measured from porous sample is approximately three times smaller than that measured from the bulk InP sample. Given the fact that the skeleton feature size in the porous network is comparable with the generated SHG wavelength, significant scattering of the SHG radiation in the porous network is possible. It is therefore expected that only a fraction of the total SHG radiation generated in reflection, in the specular direction from the porous sample, is detected by the photomultiplier tube used in experiments. To test this assertion, the reflectance of SHG radiation from the porous membrane relative to the bulk sample was examined using a variable aperture technique, where it was determined that 90 % of the reflected radiation was diffusely reflected.

Figure 24. Azimuthal dependence of the p-polarized second-harmonic intensity measured in reflection from bulk (squares) and porous (triangles) InP (100) under p-polarized excitation. Solid lines are qualitative fits to the data reflecting azimuthal dependence. Reprinted with permission from [24], M. Reid et al., Phys. Rev. B 71 081306(R) (2005). © 2005, American Physical Society.

Assuming that the porous InP surface is a Lambertian surface, it was estimated by the authors that only 0.86 % of the total reflected SHG radiation has been collected and detected in their experiments. Correcting the measurement for scattering fraction gives a measured power ratio of SHG radiation from the porous membrane relative to the bulk sample of 33. Concerning the s-polarized SHG, it was observed that the ratio of power in the SHG beam

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

170

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

from porous, relative to bulk, was 0.48. Again, correcting for the scattering losses, an estimated enhancement factor for the p-s geometry of 56 was calculated. This qualitatively agrees with the observation that a higher conversion efficiency results from the porous network in the p-s as compared to the p-p geometry for THz emission. As in the case of GaP, the observed enhancement is believed to be a result of local field enhancement within the porous network. The power in the SHG radiation scales as the input pump intensity squared, such that the output scales as the input electric field strength to the fourth power. Therefore the volumetric averaged field strength in the porous network would only have to be approximately 1/4;30=2.3 times larger than in the bulk in order to explain the experimental results. Local field enhancement is conceptually similar to focussing a pump beam to achieve higher conversion efficiencies. At this point it is worth noting that an increased effective interaction length of the fundamental and SH radiation may exist due to scattering, which would also lead to an enhancement [83,84]. However, it is expected that this effect is minor in comparison to local field enhancement for the following reasons. First, the absorption in InP of the second harmonic beam is strong enough (escape depth at 400 nm in InP is approximately 18 nm), that even if there were substantial scattering of the pump beam, the effective interaction length with the SH radiation is limited by the escape depth of the 400 nm light, which is much less than the optical absorption depth of the fundamental beam (800 nm).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.7 THz Emission from Porous InP Terahertz radiation (1 THz = 1012 Hz) is the part of the electromagentic spectrum between microwaves and infrared. It encompasses frequencies invisible to our eyes in the range from 100 GHz up to roughly 10 THz. The interest to terahertz-rays is due to the fact that they can shine through matter and are therefore of interest for many applications from luggage scanners at airports to biological imaging and study of superconductors. Semiconductor surfaces have been considered as suitable candidates for THz emission as well. Generally, THz generation mechanisms from semiconductor surfaces can be classified into two distinct categories: (i) Nonlinear-optical response of the material, i.e. optical rectification - the generation of an electric polarization by the applied optical electromagnetic wave in a nonlinear medium; (ii) Transient photocarrier related effects. Nonlinear contributions may come from bulk [85] or surface second-order nonlinearity of the semiconductor [86], or through higher-order nonlinear effects [87]. Photocarrier related effects arise as a consequence of transient photocurrents, resulting from either acceleration of carriers in the surface depletion field [88] or from diffusion of carriers into the sample away from the surface [89]. It is important to note that at high excitation fluence, the contribution from the nonlinear response of the material is quite large [85]. As discussed in the previous chapter, electrochemical formation of pores can introduce a large birefringence into the semiconductor, allowing phase matching and consequently an enhancement in optical second-harmonic generation as compared with bulk [23]. In this connection one may expect that porosity will have an impact upon THz emission, too. More than that, changes in the sample surface architecture could affect transient currents generated in semiconductors, which may lead to changes in the THz emission characteristics.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 171 The enhancement of THz emission from a porous InP membrane was for the first time reported by Reid et al [25]. The experimental setup is presented in Figure 25. A regeneratively amplified Ti:Sapphire laser system is used as a source (center wavelength of 800 nm). The probe pulse is delayed with respect to the pump using a scanning optical delay line. A variable attenuator (/2 plate and polarizer) is used in the pump beam to vary the fluence.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 25. Experimental setup for investigating THz emission from InP. The InP samples are oriented at an angle of incidence of 45o. BS is a beam splitter, HWP is a half-wave plate, P is a polarizer, M are mirrors, ODL is an optical delay line, L1 and L2 are lenses, PM are parabolic mirrors, QWP is a quarter-wave plate, WP is a Wollaston prism, PD are photodiodes, and LIA is a lock-in amplifier. Reprinted with permission from [25], M. Reid et al., Appl. Phys. Lett. 86, 021904 (2005). © 2005, American Institute of Physics.

The THz radiation from the surface of the sample, oriented at 45o angle of incidence, is collected in the specular direction and imaged onto a detector using four F/2 parabolic mirrors. A ZnTe (110) electro-optic crystal is used as detector, oriented for sensitivity to ppolarized THz emission [90] and can be reoriented to attain sensitivity to the s-polarized THz emission. A typical wave form and spectrum from bulk and porous InP is presented in Figure 26. The wave forms are similar, with the bulk semiconductor exhibiting slightly higher field amplitudes at higher frequencies. In order to determine if there is a measurable difference between the radiated THz fields from the bulk and porous InP, the authors varied the fluence of the laser beam. Figure 27 shows the peak detected THz field as a function of the pump fluence for bulk and porous InP samples. The peak THz field is seen to saturate for both bulk and porous InP. At low excitation fluences, however, the peak radiated field from the porous sample proves to be one order of magnitude larger than that inherent to bulk InP.

Figure 26. Measured THz wave forms in the time domain from bulk (a) and porous (b) InP samples. Insets show the frequency spectrum, taken as the Fourier transform of the time domain signals. Reprinted with permission from [25], M. Reid et al., Appl. Phys. Lett. 86, 021904 (2005). © 2005, American Institute of Physics.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

172

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 27. Peak detected THz field as a function of fluence for porous (filled circles) and bulk (filled squares) InP samples. Reprinted with permission from [25], M. Reid et al., Appl. Phys. Lett. 86, 021904 (2005). © 2005, American Institute of Physics.

It is difficult to determine the origin of the increase in conversion efficiency from optical to far infrared between porous and bulk InP without knowing the exact contributions from the various processes to the radiated THz field. However, for InP it is known that at low excitation fluence the photo-carrier acceleration in the surface depletion field dominates (at room temperature) [91], whereas bulk optical rectification and photocarrier diffusion dominate at higher fluences, with the crossover in mechanisms occurring at fluences 0.1 - 10 J/cm2 [92]. In order to find out which process is being actually enhanced, Reid et al measured the THz emission from the bulk and porous samples as a function of the azimuthal angle. The samples were irradiated with an incident flux of approximately 1 mJ/cm2 in a p-pol in p-pol out polarization geometry (p-p geometry). The azimuthal dependence of the THz field is shown in Figure 28a. A change in polarity in the emitted electric field is observed from both samples as a function of the azimuthal angle. As one can see from Fig. 28a, the peak detected field from the bulk sample is approximately 1.6 V/cm, while the peak detected field from the porous membrane is in excess of 7.0 V/cm. Reid et al repeated the experiments in the p-s geometry. For emission resulting from photo-carrier effects in bulk InP, the generated transient current is oriented perpendicular to the surface, and cannot radiate an s-polarized wave [93], and will therefore not contribute to the THz radiation for a p-s geometry. In the porous sample, where the possibility of lateral photo-currents exists, s-polarized THz radiation may be generated, however, the emission would be expected to be angularly independent. The results for s-polarized THz radiation from the porous sample are plotted, along with the data for the p-polarized THz emission, in Figure 28b. One can see that the s-polarized THz field is non-zero, and has a two-fold rotational symmetry associated with a second-order nonlinear response. Also, there does not exist an

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 173 angularly-independent contribution to the s-polarized THz field at levels greater than 10 %, indicating that the s-polarized THz emission is primarily due to optical rectification. Taking the ratio of the square of the peak detected s-polarized THz signals from porous relative to bulk InP gives a power ratio of nearly 100. The power enhancement factor is therefore attributed to the portion of the THz wave radiated by the process of optical rectification.

Figure 28. (a) Azimuthal dependence of the p-polarized THz field amplitude in reflection from the porous (squares) and bulk (circles) InP (100) samples under p-polarized excitation. Solid lines are qualitative fits to the data reflecting azimuthal dependence. (b) Azimuthal dependence of the ppolarized (triangles) and s-polarized (circles) THz field amplitude in reflection from the porous InP (100) sample under p-polarized excitation. Solid lines are qualitative fits to the data reflecting azimuthal dependence. Reprinted with permission from [24], M. Reid et al., Phys. Rev. B 71 081306(R) (2005). © 2005, American Physical Society.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.8. Integrated Waveguide Structures Based on Porous InP There are many key elements on which the broadband information networks must rely on. Wavelength Division Multiplexing (WDM) technology is one of them. The choice of material and technological solutions for WDM depends on many factors: performance, costs, reliability etc. InP is the most promising material in this regard because of its integrability with other optoelectronic devices. In what follows we will discuss the possibility to obtain waveguide-like structures using porous InP. (100) and (111) oriented S-doped, n-InP, n = 1018 cm-3 wafers have been used. The samples were covered with a layer of photoresist in which by standard lithography parallel 10 m broad stripes have been opened (Figure 29a). Consequently, the samples were anodized in 5 % HCl aqueous solutions at potentiostatic conditions (U = const.). The etching conditions were specially chosen in order to obtain current-line oriented pores, i.e. U = 3-9 V. The etching time was between 0.1 and 1 min. After the etching the samples have been investigated in cross section and top view using a Philips XL series Scanning Electron Microscope working at 10 and 15 kV. Note that the polymer mask was intentionally not removed from the sample after the etching, in order not to affect the structure. Figure 29 shows the results of the experiments done with samples patterned with a polymer photoresist (U = 6 V for t = 0.2 min). Figure 29a and b shows a general top overview and a magnified top view under the photoresist respectively. In order to see the structure

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

174

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

under the photoresist the accelerating voltage of the SEM was increased from 10 to 15 kV. Figure 29c and d show the cross section overview (c) and a magnified cross section situated directly under the photoresist (d). Figure 29e shows the cross section view of a region between two stripes of photoresist. From Figure 29e one can observe that the pores nucleate only on the surfaces not covered by photoresist, i.e. between two stripes, and grow radially, also under the photoresist, away from the nucleation region. It is evident that the pores do not expose any crystallographic characteristic of the single crystalline substrate. The direction of pores changes gradually from perpendicular to the substrate surface (in the center of the nucleation region), to parallel to the surface (at the edges of the nucleation region, i.e. near the photoresist). The pores growing parallel to the surface are clearly visible in Figure 29b and d. The radial growth of current-line pores is an additional demonstration of the fact that such pores grow perpendicularly to the equipotential lines of the electric field in the anodized substrate. Making an analogy between a light wave passing thought a small aperture and the current flow through a region of uncovered InP surface surrounded by two stripes of photoresist, the equipotential lines of the electric field will behave similar to the wave front of the light passing thought the aperture, i.e. will move radially outward of the slit, exposing a semispherical shape. The bulk wall visible between the two porous regions is a hint that the current-line oriented pores cannot intersect, therefore they will stop to grow or will change their direction of growth when the pore wall becomes equal to the double width of the space charge region. Thus, in contrast with crystallographically oriented pores, the current line oriented pores cannot intersect. A careful investigation of Figure 29d and e reveals that the diameter of the pores increases slightly as the pores grow deeper into the substrate (or under the photoresist). In this way the porosity of the porous layer increases as well. Considering the porous layer as an effective medium with a porosity dependent refractive index, it is straightforward that the (effective) refractive index of the porous structure will decrease as the porosity increases. As a result, a gradient in the refractive index in the depth of the structure is obtained. Recently, layers exposing a porous gradient have been proposed as waveguide-like structures in Si [94]. In InP such structures are of interest as well, taking into account the integration possibility of passive (e.g. waveguides) and active elements (e.g. LEDs) on the same chip. The main idea for waveguiding as well as for optical fibers is based on the reflections occurring at the interface between a high refractive index core and a low index cladding. The light is kept inside the core as a result of the total internal reflection (TIR) effect, a consequence of Snell‘s law, which occurs for angles larger than a critical angle when the light passes from a high to a low refractive index material. A more pronounced difference in the degrees of porosity and thus in refractive indices is shown in Figure 30. In this case a layer of crystallographically oriented pores is first formed with a low degree of porosity and then the radial growth of the current-line oriented pores is allowed (higher porosity). The low porosity layer can be considered to be the core whereas the current-line oriented pores supply the cladding layer of a waveguide-like structure. Taking into account that the degree of porosity of porous layer made of crystallographically oriented pores is less than 15 %, while that of current-line-pore layers is at least two times higher, in a very rough approximation the refractive indices of the core and cladding layers shown in Figure 30 differ by 15 % and 30 % respectively, from the refractive index of the bulk InP (n = 3.1). The simulations made on the structures shown in Figure 30 show that such a waveguide

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 175

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

will be multimode, and in order to make it single-mode it is necessary to decrease the dimensions of the core to nearly 1 μm.

Figure 29. Curro pores suitable for integrated waveguide structures. U = 5 V, t = 10 sec, 5 % HCl; a) Top view showing the overview of the patterning; b) High magnification top view between two waveguide structures; The pores growing parallel to the surface of the sample as well as the wall between the neighbouring waveguides are indicated by arrows; c) Cross section view of two neighbouring wave-guide-like structures; d) Cross section view; Higher magnification between two waveguide structures; The pores growing parallel to the surface are also easily visible; e) Cross section view; An overview showing clearly the radial distribution of the pores. Reproduced with permission from [29], S. Langa et al., Phys. Stat. Sol. C 2, 3253 (2005). © 2005, Wiley-VCH Verlag GmbH & Co.KGaA.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

176

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Figure 30. Curro pores suitable for integrated waveguide structures: higher core-shell contrast. U = 4 V, t = 1 min, 5 % HCl; a) Cross section view; b) Cross section - higher magnification. n3 1 as well as for the particularly interesting case of nPC < 0. Simple geometric optics (which should be applicable for a well-defined nPC) predicts that a concave lens then should focus light for the case when the refractive index of the lens is less than that of surrounding material. The simulations show that this is indeed the case.

Figure 32. Photonic band structure (a) and transmittance spectrum (b) of 2D PC consisting of square lattice of pores with r=0.49a in dielectric matrix with =11.4. Reproduced with permission from [100], V.V. Sergentu et al., Phys. Stat. Sol. A 201, R31 (2004). © 2004, Wiley-VCH Verlag GmbH & Co.KGaA.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

178

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Ordered parallel pores embedded in a dielectric matrix to form a concave lens as shown in Figure 33a indeed focus light in the long wavelength limit associated with n > 1. Note that there is only one row of pores in the center of the lens. Calculations were made for a radiation point source situated left from the lens at a distance 100a where there is homogeneous dielectric material with  = 11.4. Figure 33b shows the distribution of the light intensity as a function of the distance from the middle of the lens. As one can see from Figures 33a and b, the focusing effect proves to be rather strong.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 33. Focusing effect of a porous PC concave lens: (a) the electromagnetic field distribution in vacuum for long wavelength limit and (b) the transmittance coefficient along the optical axis of the lens for =10a>>a; effective nPC=1.8. Reproduced with permission from [100], V.V. Sergentu et al., Phys. Stat. Sol. A 201, R31 (2004). © 2004, Wiley-VCH Verlag GmbH & Co.KGaA.

Figure 34 illustrates the results of calculations made for a frequency where nPC < 0 (a/ = 0.81). The radiation coming from a point source placed in homogeneous dielectric material with  = 11.4 proves to be focused by the PC concave lens although the electric field modulus exhibits a more complicated spatial distribution than in the case of nPC > 1 (compare Figures 33a and 34a). The focusing effect is seen also in Figure 34b where the distribution of the light intensity as a function of the distance from the middle of the lens is illustrated.

Figure 34. Focusing effect of a porous PC concave lens: (a) the electromagnetic field distribution in vacuum and (b) the transmittance coefficient along the optical axis of the lens for wavelength =1/0.81a; effective refractive index nPC 1 and in the spectral regions characterized by negative refractive index. The obtained results may be used for the purpose of designing and manufacturing novel micro-lenses ready to be integrated in optoelectronic circuits. Note that single crystals of nanopores can be easily introduced into semiconductor materials using electrochemical etching techniques as discussed in the previous subchapters [48].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.10 Luminescent Materials Based on GaP and GaAs Semiconductor Compound Templates for Random Laser Applications The control of material structure and morphology at the nanometer scale is of major importance for tailoring the macroscopic properties such as emission spectrum and luminous efficiency and the development of novel luminescent materials. This issue becomes especially important in connection with the growing interest in the development of random lasers (see, e.g., [103] and refs therein). The solid state random lasers developed to date are mainly based on crystal powders with the stimulated emission coming either from the near-bandgap electronic effects (exciton-exciton scattering or electron-hole plasma) as in the case of lasers based on ZnO material [103-107], or from the neodymium related electronic transitions in laser crystal powders doped with neodymium [108,109]. Random lasers have been also realized in a number of materials systems, e.g., -conjugated polymer [110,111], organic dyedoped films [112] and even biological tissues [113]. A random laser consists of two major components: a laser active gain medium which amplifies light through stimulated emission and a highly scattering medium in which the recurrent scattering and interference effects result in the formation of random laser cavities [114]. Most of the random lasers elaborated to date are not suitable for integration with other optical or electronic functions. Nanocomposite materials prepared on the basis of porous semiconductor and dielectric templates are more perspective in this regard. The elaboration of materials with controllable light scattering properties is a widely expanding research field. As demonstrated in previous subchapters, an accessible and cost-effective approach for tailoring the architecture of macroscopic objects on the nanometer scale is offered by electrochemistry. As concerns the introduction of optical gain properties, one way is doping of semiconductor and dielectric templates as well as the composite materials prepared on their basis with rare earth (RE) and transition metal (TM) elements. Another way consists in making use of intrinsic material properties such as exciton-exciton scattering or electron-hole plasma effects. The technology for doping porous semiconductor templates with rare earths includes impregnation of Eu3+ and Er3+ ions form EuCl3:C2H5OH and ErCl3:C2H5OH solutions, respectively. After impregnation the samples are annealed for time periods ranging from several minutes to several hours at temperatures in the range of 500 to 1100 oC in a nitrogen flow containing less than 1 % of oxygen. The oxidation of the GaP template skeleton walls was found to start at around 600 oC. At an annealing temperature of 900 oC the porous template is totally oxidized after 30 min of

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

180

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

annealing. The morphology of the GaP porous template shown in Fig. 35a is transformed into a columnar nanostructure (Fig. 35c). The oxidation of the GaAs template was found to starts under 500 oC, and the porous template is totally oxidized at annealing temperature of 500 oC after 30 min of annealing. The morphology of the initial GaAs template (Figure 35b) is preserved after annealing at temperatures up to 800 oC as shown in Figure 35d, in spite of the fact that the skeleton is totally oxidized in these conditions.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 35. Morphologies of oxide structures formed in GaP (a,c) and GaAs (b,d) templates.

Figure 36 (a) XRD pattern of native oxides obtained on GaAs templates (1), and GaP templates with fast cooling (2) and slow cooling (3) after annealing. (b) XRD analysis of the nanocomposites prepared in GaP templates infiltrated with EuCl3:C2H5OH (curve 1) and ErCl3:C2H5OH (curve 2) solutions with concentrations of 1g/2 ml and annealed at 900 oC for 30 min. (c) XRD analysis of the nanocomposites prepared in GaAs templates infiltrated with ErCl3:C2H5OH (curve 1) and EuCl3:C2H5OH (curve 2) solutions with concentrations of 1g/2 ml and annealed at 700 oC for 30 min.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 181

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The energy dispersive X-ray (EDX) analysis suggests that the native oxide of a porous GaP layer has a GaPO4 composition [115]. As concerns the crystallography, different GaPO4 phases can be produced depending on the rate of cooling after annealing. A hexagonal P3121  - quartz type phase is obtained with slow cooling (curve 3 in Figure 36a), while an orthorhombic C2221 low-cristobalite structure is produced with fast cooling (curve 2 in Figure 36a). The native oxide in samples prepared on GaAs templates is the monoclinic C2/m Ga2O3 phase (curve 1 in Figure 36a). As concerns the rare earth ions, they are too large to be well incorporated into the tetrahedrally coordinated sites of GaPO4 or in the zincblende GaP skeleton as well as in the tetrahedrally or octahedrally coordinated sites of the -Ga2O3 phase [116]. More favourable is the segregation of the rare earth impurity in the form of finely dispersed nanophases in the composite. The constitution of the composite is controlled by the conditions of infiltration and annealing. Annealing of infiltrated GaP template at 900 oC for 30 min results in a multiphase nanocomposite illustrated by curves 1 and 2 in Figure 36b in the case of Eu or Er doping, respectively. It represents a low crystobalite GaPO4 structure with incorporated xenotime ErPO4 and monazite EuPO4 phases, identified according to the JCPCD 09-0383 and JCPCD 46-1330 cards, respectively. Similarly, the rare earth related phases in composites prepared on GaAs templates are xenotime ErAsO4 and EuAsO4 structures incorporated into the -Ga2O3 native oxide phase illustrated by curves 1 and 2 in Figure 36c. The ErAsO4 and EuAsO4 phases were identified according to the JCPCD 15-0751 and JCPCD 15-0750 cards, respectively. The intensity of peaks related to REPO4 and REAsO4 phases in the XRD spectra increases with increasing the concentration of RE salt in the impregnating solution.

Figure 37. Photoluminescence spectra of nanocomposites prepared on GaP (a) and GaAs (b) templates doped with Er (curve 1) and Eu (curve 2) ions. The spectra are measured with continuum wave (cw) 514 nm line excitation at room temperature.

The produced nanocomposite materials are characterized by bright luminescence coming from intrashell transitions into rare earth ions incorporated in various nanophases. The High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

182

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

correspondence of the Stark splitting of the ground 4I15/2 multiplet manifold of the Er3+ ion deduced from curve 1 in Figure 37a to that previously measured in a xenotime ErPO4 [117] demonstrates that the green emission comes from the Er ions in this host. The spectrum 2 in Figure 37a coincides perfectly with that measured in a monazite EuPO4 [118]. Similarly, the 4f-4f intrashell transitions in Er3+ and Eu3+ ions assure green and red emission from ErAsO4 and EuAsO4 nanophases (Figure 37b).

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.11 Fabrication and Photoluminescence Properties of Porous CdSe Little attention has been paid to the study of porosity-induced changes in the properties of II-VI compounds as compared to III-V materials. In particular, Zenia et al. subjected p-ZnTe crystals to electrochemical etching and observed the formation of needle-like structures exhibiting a blueshift of the excitonic transition energies [119]. The effect of photoetching on photoluminescence (PL) of n-CdSe was studied many years ago by Garuthara et al., who claimed the formation of etch pits [120]. Recently, pore growth was investigated in n-CdSe single crystals subjected to anodic etching [121,122]. Wurtzite-phase n-CdSe single crystals grown by chemical transport techniques using iodine as the transport agent with the concentration of free electrons n = 3x1017 cm-3 at 300 K were electrochemically etched in 5% HCl aqueous solution at room temperature under potentiostatic conditions. To reach uniform nucleation of pores, the samples were in situ illuminated by focusing the UV radiation of a 200 W Xe lamp onto the CdSe surface (0.15 cm2) exposed to electrolyte. For anodization in dark, the etching process starts at surface imperfections. After this initial pitting of the surface, further etching proceeds in all directions radially away from the initial surface imperfection. As a result, a porous domain forms around each etching pit. The pores obviously grow perpendicular to the equipotential lines of the electric field in the anodized specimen. When neighboring domains meet, the pores near the border separating the domains change their direction of growth and no pore intersection occurs (see Figure 38a). Due to space confinement, the density of pores in the emerging triangular-like regions between neighboring domains proves to be higher than the density of pores outside the areas involved. Thus, both the direction of the growth of pores and their density show pronounced spatial fluctuation in this case. Note that the EDX analysis confirmed the stoichiometric composition of the porous CdSe skeleton in spite of its rather exotic morphology. In situ UV illumination of the anodized samples allows one to reach much more uniform nucleation of pores. In this case one can easily distinguish a near-surface nucleation layer with a thickness of about 3 µm, followed by a porous layer with pores stretching perpendicular to the initial surface of the sample (Figure 38b). The chemical composition of the nucleation layer is not stoichiometric, the content of Se (about 70 at. %) being more than two times higher than that of Cd. This layer can be easily removed mechanically or by isotropic wet etching. Figure 38c shows the top surface of the porous layers after removal of the nucleation layer. Note a rather uniform distribution of pores, the lateral dimensions of both pores and walls being about 200 nm. According to the results of the EDX analysis, the composition of the porous skeleton is stoichiometric.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 183

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 38. (a) SEM images taken in cross section from a CdSe sample anodized in dark illustrating the development of the porous structure after neighboring domains meet. (b) General view of a sample fabricated under in situ UV illumination. (c) Top view of the sample fabricated under in situ UV illumination after removal of the nucleation layer.

In contrast to CdSe samples with electron concentration n = 3x1017 cm-3, a uniform pitting of the top surface occurs in samples with electron concentration n = 2x1018 cm-3 even without use of in situ UV illumination. The characteristic size of the obtained porous structure can be changed by the applied voltage. This parameter deduced from the statistical analysis of the SEM image of the porous structure formed at 6.5 V is in a sample with electron concentration n = 1x1018 cm-3 is around 10–20 nm. The characteristic sizes of the porous skeleton increase up to around 50 nm with the increase of the applied voltage up to 15 V. The decrease of the skeleton size down to 10–20 nm leads to quantum-size effects in the nanocrystalline CdSe porous skeleton. The PL spectrum of the as-grown CdSe single crystals with the free electrons concentration of 2x1018 cm-3 measured at 10 K (see Figure 39a) is dominated by the PL bands related to the recombination of neutral donor bound excitons (D0X) at 1.823 eV and free XA excitons at 1.827 eV. On the high-photon-energy side of these bands, luminescence related to the excited state of the XA exciton at 1.837 eV and of the XB exciton at 1.851 eV are observed. An effective Rydberg constant of the XA exciton equal to 13 meV is deduced from these data. This exciton binding energy along with the XA exciton series limit of 1.840 eV and the 9–7 valence band crystal field splitting of 24 meV corroborate the previously reported data [123]. On the low photon-energy side of the (D0X) peak, one can find two-electron replica (D0X)2e at 1.804 eV. The (D0X) peak is due to the radiative decay of the excitons bound to a neutral donor with the liberation of free excitons and leaving the donor in the ground state. The separation of 4 meV between the (D0X) and XA peaks is the binding energy of the exciton to the neutral donor. The two-electron replica results from the recombination of the excitons bound to the neutral donor, the donor being left in an excited state. The energy of the (D0X)2e is therefore lower than that of the principal (D0X) peak by the difference in ground and excited state energies. The binding energy of the donor involved can be estimated using a simple hydrogen model, where the binding energy is just 4/3 of the separation between the (D0X) and (D0X)2e peaks. According to the Haynes rule, the excitonic localization energy scales linearly with the donor binding energy [Ebind(D0X)=bEdonor], with the constant b in the range of 0.1–0.3 for different semiconductors. A value of b equal to 0.17 is deduced from our experiments. Apart from excitonic luminescence, a much weaker PL band associated with donor-acceptor (DA) recombination [122,124,125] along with phonon replica is inherent to the as-grown material. With the temperature increase, the luminescence related to donor bound excitons sharply decreases, and at temperatures higher than 50 K the luminescence is free excitonic.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

184

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 39. (a) PL spectrum of the n-CdSe crystalline substrate with a free electron concentration of 2x1018 cm-3. Temperatures (K) of (1) 10, (2) 100, (3) 150, (4) 200, and (5) 300. (b) PL spectrum of nCdSe sample anodized in the dark under an applied voltage of 6.5 V. Temperatures (K) of (1) 10, (2) 70, (3) 100, (4) 150, and (5) 200. Reprinted with permission from [122], E. Monaico et al., J. Appl. Phys. 100 053517 (2006).© 2006, American Institute of Physics.

In contrast to the as-grown material, the luminescence of the nanostructured material obtained at low applied voltage during the dissolution process is dominated by free exciton recombination starting from temperatures as low as 10 K (see Figure 39b). The other two features of the PL from the porous material are the blueshift of the excitonic emission by 10 meV and a higher intensity of the LO phonon replica of the XA exciton. Apart from excitonic luminescence, the DA related luminescence is observed in PL spectra at temperatures up to 100 K. At temperatures above 100 K, the DA luminescence sharply decreases, since the impurity with smaller binding energy involved in DA transitions (most probably the donor impurity) is ionized. At the same time, a high-energy shoulder of the DA band associated with free-to-bound transition becomes resolved in the spectra at temperatures above 50 K. The temperature dependence of the XA exciton energy in the as-grown material (Figure 40a) is well fitted with the phenomenological Varshni formula [126] E(T) = E0-T2/(T+)

(14)

with the parameters E0 = 1.837 eV eV,  = 4x10-4 eV K-1, and  = 140 K.

Figure 40. Temperature dependence of the XA exciton line energy (a) and FWHM of the XA exciton line (b) in CdSe crystalline substrate (full symbols) and porous layer anodized at 6.5 V (open symbols). Lines represent the fit of the experimental data to the theoretical equations. Reprinted with permission from [122], E. Monaico et al., J. Appl. Phys. 100 053517 (2006).© 2006, American Institute of Physics.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 185 The deduced values of the  and  parameters are in reasonable coincidence with the previously reported data [127,128]. The temperature dependence of the XA exciton energy in the nanostructured material is well fitted with the same parameters but with the value of E0 = 1.837 eV which is 10 meV higher than that of the as-grown material. This blue shift is attributed to the quantum-size effects in the nanocrystalline porous skeleton. It is well known that the optical properties of nanocrystals strongly depend on the ratio of the nanocrystal size a to the Bohr radius of the bulk exciton aB. In the analysis of experimental data, one needs to consider three different regimes: a  aB, a  aB, and a  aB. The exciton Bohr radius in CdSe is around 5 nm. In the case a  aB, the binding energy of an exciton is larger than the quantization energy of both the electron and holes, and the optical spectra of these nanocrystals are determined by the quantum confinement of the exciton center of mass [129]. In this case the quantum confinement blue shift of the exciton energy is given by h2/(8Ma2), where M = me+mh is the exciton translation mass. This case is known as the weak confinement regime. The strong confinement regime is realized in small nanocrystals, where a  aB. In this case, the exciton translation mass is replaced by the exciton reduced mass -1 = me-1 + mh-1 in the above presented formula. By using the values of the free carrier effective masses of me = 0.13m0 and mh = 0.45m0, one can calculate a blue shift of around 6 meV in the case of weak confinement regime, and a shift of 30 meV in the case of strong confinement regime in CdSe nanocrystals with the average size of 10 nm. The experimentally observed shift of 10 meV in our nanostructured layer suggests that we deal with a case of weak to intermediate confinement regime. The higher intensity of the LO phonon replica of the XA exciton in nanostructured layer as compared with that inherent to the as-grown material is indicative of a stronger exciton – LO phonon interaction. The strength of the exciton – LO phonon interaction is an important parameter since it affects optical and electrical properties of semiconductors (emission line broadening, hot carrier cooling, carrier mobility). The influence of nanostructuring upon the exciton – LO phonon interaction was investigated via the analysis of the XA exciton linewidth (Figure 40b). The measured luminescence linewidth is the sum of an inhomogeneous part (i) that is due to intrinsic effects (dislocations, interface roughness, composition fluctuations, and electron-electron interactions), and a temperature-dependent homogeneous part (h). At low temperatures, the homogeneous component is dominated by the scattering of acoustical phonons. As the temperature increases, the LO phonon scattering becomes dominant due to the increase of the phonon population. The broadening due to the impurity scattering also becomes important at higher temperatures due to the ionization of impurities. The full width at half maximum (FWHM) of the emission line can be expressed through the following equations [130,131]:

(T) = i + h

(15)

h = ACT + LO/[exp(ELO/kBT) – 1] + imp/[exp(Eimp/kBT)],

(16)

where AC, LO, and imp are interaction constants; ELO and Eimp are the average LO phonon energy and the impurity ionization energy. By using the LO phonon energy of 26 meV and the impurity activation energy of 24 meV deduced from the (D0X)2e peak position, the best fit

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

186

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

in the as-grown material was reached with the following parameters: i = 3.5 meV; AC = 1x10-2 meV/K; LO = 60 meV; and imp = 20 meV. The previously reported values of the AC lie in the interval 1x10-3 – 3x102, the value of LO is between 20 – 60 meV, and the value of the imp most commonly used is 15 – 20 meV [127,128]. The smallness of the inhomogeneous broadening along with the PL spectra presented above is indicative of the high quality of the as-grown CdSe single crystals. In contrast with the as-grown samples, the best fit of the experimental data in the nanostructured material is reached with the LO value of 90 meV and the i = 8 meV. The higher value of the i is due mainly to a statistical distribution of the porous skeleton sizes, while the increased value of the LO by a factor of 1.5 is related to a stronger exciton – LO phonon interaction in the nanostructured layer. Therefore, preparation of nanoporous CdSe templates with the characteristic size of the porous skeleton entities down to 10 to 20 nm is possible by means of an accessible and costeffective approach based on electrochemical etching of bulk crystalline substrates. The decrease of the skeleton size down to 10 – 20 nm leads to a blue shift of the excitonic emission lines by around 10 meV due to quantum-size effects in a weak-to-intermediate confinement regime. The exciton – LO phonon interaction in the nanocrystalline CdSe porous skeleton is increased by a factor of 1.5 in comparison with that inherent to bulk crystals. Similarly to III-V compounds, anodization of n-CdSe in aqueous solution of NaCl proves to be an effective and environmentally-friendly approach for the purpose of uniform nanostructuring and fabrication of CdSe nanotemplates.

Figure 41. Top view SEM images taken from CdSe samples with electron concentration n = 3x1017cm-3 (a,b) and n = 1x1018cm-3 (c,d) anodized at 18 V in 1.75M (a,c) and 3.5 (b.d) NaCl aqueous solution.

The morphology and porosity of anodized CdSe samples depend on the electron concentration in the initial wafer, and are controlled also by the concentration of the electrolyte. For instance, the anodization at 18 V of a sample with electron concentration n = 3x1017cm-3 in a 1.75M NaCl aqueous solution (Figure 41a) results in the formation of a

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 187 porous layer with the diameter of pores around 200 nm and the thickness of the pore walls around 220 nm. The treatment of a sample with electron concentration n = 1x1018cm-3 in similar conditions (Figure 41c) leads to the production of a porous layer with the diameter of pores around 250 nm and the thickness of the pore walls around 80 nm, i. e. the increase of the electron concentration in the initial wafer leads to a significant increase in the degree of porosity. The increase of the concentration of the electrolyte leads to an overall decrease of the diameter of pores. For instance, anodization at 18 V in a 3.5M NaCl aqueous solution results in the formation of pores with the diameter around 50 nm in the sample with electron concentration n = 3x107cm-3 (Figure 41b), and the pores with the diameter around 100 nm in the sample with electron concentration n = 1x1018cm-3 (Figure 41d), while the thickness of the pore walls is around 40 – 50 nm in both samples.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.12 Preparation of Porous ZnSe Templates While the electrochemical etching is a versatile tool for use in nanostructuring of narrow and medium band-gap materials such as Si, InP, GaAs, GaP, ZnTe, CdSe, use of this method for wide bandgap II-VI semiconductors is still a challenge. The related difficulties are linked with controlling the conductivity of the source material, as high carrier concentration is needed to apply anodic etching for nanostructuring. It is difficult to obtain wide bandgap semiconductors with high conductivity due to self-compensation phenomena inherent to these materials [132]. Development of a reliable technology for controllable doping of the semiconductor material is a first imperative step towards controlled electrochemical nanostruturing. Doping crystals with III-group elements proved to be an effective way of solving this problem with ZnSe [133,134]. The possibility to control the conductivity of ZnSe crystals by doping the samples with Al from a Zn+Al melt was demonstrated [135,136,137]. This procedure allows producing suitable samples for controllable nanostructuring using electrochemical etching techniques. The Al doping of as-grown high resistivity (  108 cm) n-ZnSe single crystals grown from a melt was carried out by means of high temperature (950 oC) annealing in a Zn+Al melt with different doping impurity contents during 100 hours. The doping level was controlled by the variation of the Al content X in the melt [(100 – X) at.% Zn + X at.% Al], with X varied in the range from 0.1 to 40 at.%. The influence of the annealing in Zn+Al melt upon electrical parameters of ZnSe crystals is summarized in Table 1. The analysis of data in Table 1 is indicative of a rather complicated dependence of the electrical parameters of ZnSe crystals upon the concentration of Al in the Zn+Al melt. The mechanisms responsible for the variation of electrical parameters as a function of annealing conditions were discussed elsewhere [135]. According to the results of electrical characterization [135], the increase of the Al concentration can be divided in four intervals as follows: (i) from 0 to 0.5 at. %; (ii) from 0.5 to 5 at. %; (iii) from 5 to 20 at. %; (iv) more than 20 at. %. The enhancement of Al content in the melt within the first and third intervals leads to the increase of the electron concentration in crystals, while the increase of the Al concentration within the second and fourth intervals results in the decrease of the electron concentration. This non-monotonous behavior is explained by complex processes of interaction between the doping impurity and the intrinsic defects or residual impurities. The dynamics of the formation and dissociation of associative

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

188

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

centers determines the complex character of the dependence of electrical parameters upon the density of Al doping impurity. At low impurity concentration (first interval) Al atoms form donor centers giving rise to the increase of electron concentration. The increase of Al content in the melt above 0.5 at. % (second interval) results in the formation of (VZnAlZn) associative acceptor centers leading to self compensation of the shallow Al donor impurity and to the decrease in the electron concentration. The variation of the Al content in the melt between 5 and 20 at. % (third interval) leads to the dissociation of the acceptor complexes and to a recurrent donor doping effect. Finally, the sharp decrease of the electron concentration in the crystals with increasing the Al content in the melt above 20 at. % is explained by the increase of the concentration of compensating intrinsic VZn defects due to the weakening of the diffusion process of Zn atoms in the highly Al enriched melt. The anodic etching was carried out in K2Cr2O7:H2SO4:H2O electrolyte with the ratio of 5:100:10 at 25 C in potentiostatic regime with applied voltage varied. The morphology parameters of the anodized ZnSe samples (the pore mean diameter and the thickness of the skeleton walls) follow the variation of electrical parameters in crystals as summarized in table 1. The anodization of ZnSe samples obtained from a pure Zn melt, i. e. with the electron concentration of 7.2 x 1016 cm-3, requires voltages over 30 V for the formation of porous structure and the resulted pores have the mean diameter as high as 400 - 500 nm, as illustrated in Figure 42a for a sample anodized at 40 V. The increase of the electron concentration to (12)x1018 cm-3 in samples allows one to reduce the anodization voltage down to 6 V and to obtain porous structures with the pore mean diameter of about 40 nm as shown in Figure 42b for a sample with electron concentration of 9x1017 cm-3 produced by annealing in a Zn+Al melt with 1 at % of Al.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Table I. Electrical parameters of n-ZnSe:Zn+X at.% Al single crystals and the mean pore diameter of porous samples produced by anodization X аt.% Al 0 0.1 0.3 0.5 1 5 10 20 40

n, cm-3 7.2·1016 9.3·1016 3.0·1017 2.1·1018 9.1·1017 5.8·1017 5.1·1017 1.7·1018 4.4·1016

σ, -1cm-1 1.49 3.15 12.64 29.30 12.77 17.22 15.03 20.10 1.07

Rσ, cm2/V·s 130 210 265 126 88 186 184 72 155

Mean pore diameter, nm 400-500 200-300 100 40 40 60 60 40 600-900

Required anodization voltage, V >30 >20 >9 >6 >6 >8 >8 >6 >50

One should mention that the variation of the electron concentration from around 1x1018 cm-3 to 2x1018 cm-3 does not change significantly the morphology of the anodized samples. Therefore, the morphology of samples produced by anodization of ZnSe crystals annealed in Zn+Al melt with 0.5, 1, and 20 at % of Al is similar to that illustrated in Figure. 42b, i. e. the morphology of the anodized samples is not determined by the concentration of Al impurity in the melt, but by the produced electron concentration in crystals. The increase of electron

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 189 concentration in crystals from 7x1016 to (1-2)x1018 cm-3 allows one to reduce the pore diameter from 400 - 500 nm to 40 nm. The width of the porous skeleton walls correlates with the diameter of pores, i.e. in all porous samples the width of the skeleton walls proves to be nearly equal to the pore diameter.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 42. SEM image taken in cross section from porous ZnSe samples produced by anodic etching of crystals annealed in a pure Zn melt (a), and annealed in a Zn+Al melt with the concentration of Al equal to 1 at % (b). Inserted is the top view.

The possibility to control the diameter of pores just by changing the applied potential during anodic etching allows one to prepare multilayer porous structures in one technological process. Figure 43a shows an example of a porous structure that consists of three layers exhibiting pores with different transverse dimensions: the average diameter is 500 nm, 200 nm and 60 nm for the upper, middle and lower layer, respectively. As evidenced by EDX analysis, the porous skeleton in the whole porous structure is characterized by stoichiometric composition inherent to the ZnSe compound. It is interesting to note that the morphology of the multilayer porous structure does not depend upon the mode of the transition from one value of the applied voltage to another one: it can be both sharp and with a temporary cease of the anodic etching. This is quite different from the case of n-InP where consecutive on–off switching of the etching process leads to the alternation of porous layers exhibiting by turns crystallographically oriented and current-line oriented pores [138]. Although the formation of uniformly distributed pores exhibiting features of short-range order is inherent to anodic etching of n-ZnSe [136], long-range order in pore distribution was not reached yet. The reason could be the absence of crystallographically oriented pores in ZnSe. As shown in previous subchapters the long-range order in pore distribution evidenced in n-InP is favored by the network of crystallographically oriented pores [47]. In case this network is well developed, there is a general tendency of current-line oriented pores to form rows oriented along direction. Noteworthy that porous ZnSe, in spite of the absence of crystallographically oriented pores, exhibits pore rows oriented along direction (see arrow in the insert of Figure 43a). This is a hint to the possibility for the occurrence of longrange order in the self-organized distribution of pores introduced in n-ZnSe substrates by anodic etching.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

190

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 43. SEM images taken from multilayer porous structures in cross-section: three layers (a) and two layers (b). The insert shows the alignment of pore rows along the crystallographic direction in ZnSe (see text for details). Reproduced with permission from [155], E. Monaico et al., Phys. Stat. Sol. RRL 3, 97 (2009). © 2009, Wiley-VCH Verlag GmbH & Co.KGaA.

Successive anodization of ZnSe substrates at varied applied voltage results in layer porosification at different length scales. The approach is demonstrated by the image presented in Figure 43b, the sample being subjected to anodic etching in two steps at applied voltages of 15 V and 8 V, respectively. The first anodization step results in the formation of a porous layer with the transverse dimensions of pores and pore walls as high as several hundreds of nanometers (see the upper porous layer in Figure 43b). The second anodization step has a double function: first, it leads to the formation of a new porous layer with pores of about 60 nm in diameter (see the lower porous layer in Figure 43b) and, second, it simultaneously generates similar pores in the thick walls of the upper porous layer. The observed successive porosification of the same layer at two different length scales opens new possibilities for the design and fabrication of device structures based on porous semiconductor compounds. Note that a similar type of successive porosification at two length scales was applied recently to macroporous Si by Lehmann [139]. Figure 44 compares the Raman spectrum measured from a (111) cleavage of initial bulk ZnSe sample annealed in a Zn+Al melt with the concentration of Al equal to 1 at % with the Raman spectrum of a porous structure produced from this material. The RS spectrum of the as-annealed sample is dominated by the TO mode at 205 cm-1. Apart from the TO mode, a complex RS band structure is observed in the region of the LO phonon. In doped polar semiconductors the plasmon from free carriers interacts with the LO phonon and gives rise to the LO-phonon-plasmon coupled (LOPC) modes in the RS spectra, usually denoted as L+ and L- [140] The L+ mode is observed in the RS spectrum of the as-annealed sample at 273 cm-1. The ―unscreened‖ LO phonon at 253 cm-1 comes from the thin surface depletion layer. The origin of the Raman signal on the high wavenumber side of the L+ mode is unclear. This signal may be due to the second order scattering contributions, some local Raman modes, or due to resonant electronic Raman scattering by bound electrons as previously observed in ZnSe crystals [141]. A resonantly enhanced 2LO phonon peak dominates the high wavenumber region of the spectrum, since the energy of the excitation laser line is close to the energy of electronic transitions in ZnSe. The intensity of second order LO-scattering is decreasing with increasing the wavelength of the used light excitation (not shown here).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 191 The electron concentration in the as-annealed sample was estimated from the position of the L+ mode in the Raman spectrum according to the relationship [141]

 2p   2L

 2LO   2L , 2 TO   2L

(17)

where p is the free electrons plasma frequency in the doped sample, TO and  LO are the TO and LO frequencies in a semi-insulating sample, respectively, and L+ is the frequency of the L+ LOPC mode. The plasma frequency is related to the electron concentration

 2p 

ne 2 ,   m

(18)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where  and m* are the high frequency dielectric constant and the conduction band effective mass, respectively. With  = 6.1 and m* = 0.16 [142] we obtain the value of n = 8.8x1017 cm3 which is close to the value of 9.1x1017 cm-3 obtained from Hall measurements

Figure 44. RS spectra of bulk and porous ZnSe samples prepared by annealing in a Zn+Al melt with the concentration of Al equal to 1 at % and anodic atching at 10 V. Inserted is the curve fit analysis of the spectrum of the porous sample.

There are several differences in the RS spectrum of the porous sample as compared to that of the bulk material. First, the intensity of the RS signal in the region of LO phonon is comparable with the intensity of the TO peak, which may be related to the porosity induced relaxation of the selection rules [18]. Second, the 2LO/1LO intensity ratio in the porous sample is much lower in comparison with that inherent to the bulk sample. This effect may be attributed to the nanostructuring induced decrease of the coupling strength of the electronphonon interaction. Within the Frank-Condon approximation [143], the electronic oscillation

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

192

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

strength distribution over the nth phonon mode is defined as I  Sne-S/n!, where S is the Huang-Rhys parameter which is a measure of coupling strength of the electron to LO phonon. The decrease of the ratio between the second- and the first-order Raman scattering cross section with the decrease of the crystallite size was previously found in ZnO [144,145], CdS [146], CdSe [147], and InP [148] nanostructures. Third, the RS spectra in the region of LO phonon mode in bulk and porous samples are rather different. The insert in Figure 44 presents the curve fit analysis of the RS signal in the porous sample. The ―unscreened‖ LO phonon intensity is higher in comparison with that of the bulk material. This can be explained by the large internal surface of the porous material, and, respectively, the increase of contributing volume of surface depletion layer to the total Raman signal in porous samples. Another observation is the disappearance of the RS signal on the high wavenumber side of the L+ mode. This may be related to the neutralization of centers responsible for this signal during the anodization of the initial bulk sample. One should note also the shift of the L+ mode position towards lower wavenumbers in comparison with the bulk material, which may indicate on the decrease of the free electron concentration in the porous sample. The free electron concentration in the porous sample estimated from the relationships (17) and (18) equals 4.6x1017 cm-3, i. e., by a factor of about 2 lower than in the bulk sample. Finally, the most important observation is the emergence of the Fröhlich type surface-related mode (marked as S mode) characteristic for nanostructured polar materials. It was observed that annealing of bulk single crystalline ZnSe wafers with the orientation (111) in oxygen ambient results in the formation of a uniform porous material with the mean size of grains between 100 and 200 nm. The possibility to control the porosity of the ZnSe template by the variation of anodization conditions allows one to prepare ZnO templates with controlled porosity by means of subsequent annealing as shown in Figure 45.

Figure 45. SEM image taken from a ZnSe template produced by electrochemical etching (a), and a porous ZnO template produced by annealing of a ZnSe template (b).

The XRD characterization demonstrates the gradual oxidation of the ZnSe wafer and the transformation of the initial zincblende ZnSe structure to the wurzite ZnO structure. Figure 46a shows the XRD patterns of the material annealed at 600 and 800 °C in comparison with the XRD spectrum of the initial ZnSe single crystal. When the ZnSe crystal is annealed at 500 °C, the ZnSe begins a transformation to ZnO, indicated by a mixed diffraction pattern of ZnS and ZnO in the XRD spectrum. (100), (002), and (101) ZnO diffraction peaks are clearly seen in the XRD spectrum for the sample annealed at 600 °C. The XRD pattern for the samples

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 193 annealed at Ta > 700 °C consists of only ZnO diffraction peaks, indicating that the ZnSe fully transforms into porous ZnO with a hexagonal wurtzite structure. This transformation is confirmed by the PL analysis.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 46. XRD (a) and cw PL (b) spectra of a ZnSe wafer (1), and of the material produced by annealing of this wafer at 600 oC (2) and 800 oC (3). The spectra (2) and (3) are shifted along the y-axis for the sake of clarity.

Figure 46b shows the low temperature cw PL spectra of the ZnSe wafer annealed at different temperatures in the range from 600 °C to 800 °C. The PL spectrum of the initial ZnSe single crystal consists of two bands at 2.79 eV and 2.73 eV associated with the recombination of excitons and donor-acceptor pairs. Annealing at 600 oC leads to the emergence of PL bands related to ZnO crystallites. Only PL bands of the ZnO component remain in the PL spectrum of samples annealed at 800 oC. These results indicate that highquality ZnO material with uniform porosity is easily obtained by using the technological approach involved.

3. METALLO-DIELECTRIC NANOCOMPOSITES 3.1 InP-Metal Composites Nowadays two types of templates are widely used for nanofabrication purposes, namely porous Al2O3 and etched ion track membranes based either on inorganic materials or on organic polymers [149–152]. Both porous Al2O3 and etched ion track membranes, however, exhibit high resistivity and therefore they often play a passive role in nanofabrication processes. In particular, templated growth of nanowires via electroplating is provided usually by the metal contact deposited on the back side of the high-resistivity membranes, while electroplating of metal nanotubes requires additional technological steps e.g. chemical modification of the inner surface of the pores prior to electrodeposition which leads to the incorporation of spurious phases in the nanotube walls [150]. In this connection an important technological task is the development of cost-effective semiconductor nanotemplates which

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

194

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

properties could be easily controlled by external illumination, applied electric fields etc. Recently, the feasibility of indium phosphide nanotemplates for electrochemical deposition of arrays of platinum nanotubes with diameters both larger and smaller than 100 nm was demonstrated [153]. Pt electroplating was performed in 30-µm thick porous InP layers exhibiting current line oriented pores with diameters of 70 nm (Figure 47) and 140 nm (Figure 48), i.e. with aspect ratio exceeding 400 and 200, respectively. The duration of the cathodic pulse was 100 and 300 µs for Pt deposition in pores with diameters of 70 and 140 nm correspondingly. Pieces of Pt nanotubes getting out from pores are clearly seen in the cross-sectional view taken from a cleaved sample with 70-nm pores, see Figure 47a. To study the distribution of nanotubes in depth of the porous layers, some samples were cleaved along planes perpendicular to pores. As shown in Figure 47b, many of Pt nanotubes get out from pores and are suspended in air. The thickness of the nanotube walls is of about 10 nm. The quality of nanotubes is indicative of high uniformity of metal deposition on inner surface of pores.

Figure 47. SEM images taken from cleaved porous n-InP samples with pore diameters of 70 nm after pulsed electrodeposition of Pt: general view of the 30-lm thick layer (a); top view after the sample was additionally cleaved along a plane perpendicular to pores (b). Reprinted with permission from [153], Ion Tiginyanu et al., Electrochemistry Communications 10 (2006). © 2006, Elsevier

Figure 48. SEM images taken from cleaved porous n-InP samples with pore diameters of 140 nm after pulsed electrodeposition of Pt: top view (a); cross-sectional view (b). The insert illustrates a sequence of Pt nanotube rows in n-InP matrix. Reprinted with permission from [153], Ion Tiginyanu et al., Electrochemistry Communications 10 (2006). © 2006, Elsevier. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 195 It is interesting to note that in SEM images the Pt nanotubes look bright in comparison with the porous n-InP skeleton walls. This is a consequence of the charging phenomenon caused by the potential barrier at the Pt-InP interface. Indeed, taking into account that pulsed electrochemical deposition of Pt on n-InP leads to the formation of Schottky barrier with the height up to 0.65 eV [154], it is obvious that the negative charge accumulates in metal nanotubes during morphology study by SEM. Pt electroplating in nanotemplates with pore diameter of 140 nm also leads to uniform metal deposition on inner surface of pores (Figure 48a and b). Practically all pores prove to contribute to the formation of quasi ordered 2D array of Pt nanotubes, the thickness of the nanotube wall being of about 30 nm. We found that packs of rows of Pt nanotubes in semiconductor envelope can be easily cleaved from the sample. A sequence of Pt nanotube rows in InP matrix is illustrated in the insert of Figure 48b. Actually the semiconductor nanotemplate with the embedded array of metal nanotubes behaves like a layered crystal, the role of individual layers being played by the rows of Pt nanotubes in n-InP envelopes. This is the result of ordered growing of current line oriented pores in the InP-template along direction as discussed in the previous subchapters. In consequence of this, the Pt nanotube rows prove to be oriented along direction, individual rows being aligned in different {011} planes which are cleavage planes in single crystalline InP. This is the reason that thin films consisting of packs of rows of Pt nanotubes in n-InP envelope can be easily cleaved from the sample. EDX analysis of chemical composition of samples after electroplating for different time periods shows that Pt deposition takes place uniformly on the inner surface of pores under pulsed voltage regimes applied. In particular, the presence of P, In and Pt is confirmed for sample shown in Figure 48a in concentrations 24.5, 25.0 and 50.5 at.%, respectively. In depth of the porous layers, the Pt deposition occurs uniformly with a deviation in concentration within the limits of ±20%. This is the result of relatively good conductivity of the porous InP skeleton which provides conditions for uniform metal deposition without preliminary technological steps like surface sensitization and activation often leading to severe contamination with impurities. Another important point is the short duration of voltage pulses and relatively long delay time applied, assuring recovery of the ion concentration in the electrolyte along the whole depth of pores.

3.2 ZnSe-Metal Composites The electronic band gap of indium phosphide is 1.3 eV, which means that the nanotemplates based on InP are opaque in the visible region of the spectrum. Among III–V and II–VI semiconductors one may consider the wide band gap compounds GaN (Eg = 3.3 eV), ZnO (3.3 eV) and ZnSe (2.7 eV) as good candidates for the fabrication of conductive nanotemplates transparent in the visible region. However, GaN crystalline substrates are not yet widely commercially available, while ZnO, according to preliminary studies, seems to be inappropriate for electrochemical pore growth. At the same time, the possibility to use a ZnSe porous matrix for the purpose of electroplating arrays of metal nanotubes was recently demonstrated [155]. Pt electroplating was carried out in 20 µm thick porous ZnSe layer with diameters of pores of about 400 nm. The electrochemical deposition of Pt was performed at 40 °C for 8 h in a common two-electrode plating cell containing 2 g/l Pt where the porous sample served as

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

.

196

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

working electrode, while a platinum wire was used as counter electrode. A pulsed voltage regime with rectangular pulses was provided by a home-made generator. During the 200 µs pulse time a cathodic potential of –40 V was applied between the two electrodes to electrochemically reduce the metal species on the inner surface of the porous matrix in contact with the electrolyte. After each pulse a delay time as long as 1 s was used at zero external voltage applied to allow ions to diffuse into pore regions depleted during the deposition pulse. Besides, magnetic stirring was applied to provide recovery of the ion concentration in the electrolyte along the whole depth of pores. As one can see from Figure 49a, electrochemical deposition of Pt resulted in the formation of metal nanotubes with the wall thickness of about 50 nm. Pieces of Pt nanotubes getting out from pores are clearly seen in the cross-sectional view taken from a cleaved sample, see insert in Figure 49a. The quality of nanotubes is indicative of good uniformity of metal deposition on the inner surface of pores. The uniformity of Pt deposition as a function of depth was proved also by EDX analysis of chemical composition. The amount of deposited Pt proved to be proportional to the deposition time. Note that in SEM images Pt nanotubes look bright in comparison with the porous n-ZnSe skeleton walls. Like in similar Pt–InP structures [153], this seems to be a consequence of the charging phenomenon caused by the potential barrier at the Pt–ZnSe interface.

Figure 49. (a) SEM image taken from cleaved porous n-ZnSe sample after pulsed electrodeposition of Pt. The insert illustrates a top view after the sample was additionally cleaved along a plane nearly perpendicular to the pores. Reproduced with permission from [155], E. Monaico et al., Phys. Stat. Sol. RRL 3, 97 (2009). © 2009, Wiley-VCH Verlag GmbH & Co.KGaA. (b) SEM image of a ZnSe templae with Pt nanodotes.

By reducing the width of the pulse to values smaller that 100 s one can deposit metal dotes instead of nanotubes into the porous ZnSe template as shown in Figure 49b. The size of the metallic dot is controlled by the deposition time. Thus, the feasibility of ZnSe-based nanotemplates for nanofabrication was demonstrated. The high conductivity of the nanotemplate skeleton provides conditions for uniform electrochemical deposition of metal species on the inner surface of pores, resulting in the formation of arrays of metal nanotubes embedded in wide-bandgap semiconductor matrix. These metallo–semiconductor structures are promising for the elaboration of photonic crystals and negative refractive index metamaterials, in particular of novel focusing elements for applications in the visible region of the spectrum [156].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 197

3.3. Negative Index Material Lenses Based on Metallo-Dielectric Nanotubes

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Negative-index materials (NIMs) are an emergent class of synthetic materials increasingly used for the development of optical elements. NIMs provide the opportunity for building a ―perfect lens‖ that can focus electromagnetic waves to a spot size much smaller than a wavelength [97,157]. NIMs have been designed on the basis of composite wire and split ring resonator structures [157,158], backward-wave transmission lines [159], and photonic-band-gap crystals [160,161]. It was recently proposed to design NIM lenses from dielectric rods with a specific dielectric constant profile approximating a fish-eye one [162,163]. This design was shown to provide improved focusing with a much thinner flat lens as compared to that assembled from homogeneous dielectric rods it demonstrating also higher tolerances to the induced disorder in the rod assembly. The approach of designing NIMs on the basis of dielectric rods with a gradient of the dielectric constant was tested experimentally at microwave frequencies [164]. Metallodielectrics are also widely used for the development of photonic crystals and optical elements [165]. Tubular dielectric structures, particularly titania nanotubes, are readily prepared by cost effective electrochemical methods as will be shown in a next subchapter. These nanotubes can be easily covered by metallic films by electrodeposition as demonstrated for InP and ZnSe nanotemplates. A highly efficient and accurate multiple-scattering approach [99] is used to calculate propagation of electromagnetic waves through the materials designed. Primarily, the light scattering properties of individual dielectric nanotubes with inner and outer surfaces covered by metallic films are analyzed. A simplified method was also demonstrated to be effective [156], which consists in the analysis of a parameter f describing the difference from the point of view of light scattering properties between the investigated rod and a rod with identical radius but consisting of material with the refractive index n = -1 f = max|Dmni - Dm|10>m>-10

Figure 50. The spectral dependence of the parameter f. Reproduced with permission from [156], V.V. Sergentu et al., Phys. Stat. Sol. RRL 2, 242 (2008). © 2008, Wiley-VCH Verlag GmbH & Co.KGaA.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

198

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where Dmni and Dm are the parameters determining the light scattering properties [99] of the cylinder made from the material with n = -1 and the investigated cylinder, respectively; m is the index of the cylindrical function [99]. The calculations have been done for titania nanotubes (n = 2.6) with the inner diameter of 80 nm and the outer diameter of 160 nm the outer and inner surfaces being covered with an Ag film with thickness of 12 nm. Nanotubes with this geometry are usually produced by electrochemical treatment of Ti foils in a mixture of HF and H3PO4 solutions in ethylene glycol. The optical constants of Ag were those from ref. [166]. The electric field vector E was considered to be parallel to the cylindrical axis. The spectral dependence of f exhibits two minima at photon energies 1.5 and 3.2 eV (Figure 50). Better focusing properties are expected for radiation with photon energy of 3.2 eV since the minimum at this photon energy is deeper. The focusing properties of lenses have been investigating by calculating the transmitted through the lens electromagnetic power T = (E/E0)2, where E is the electric field amplitude of the radiation passed through the lens, and E0 is the electric field amplitude without the lens. The flat and concave lenses assembled from rods with n = -1 and diameter of 160 nm arranged in a regular triangular lattice demonstrate focusing in a wide spectral interval. However, focusing properties are better at longer wavelength (h = 0.5 eV) as observed from the comparison of Figure 51a and 51b.

Figure 51. Electric field intensity map of a cross-sectional view of the 2D source-image system when imaging by a triangular-lattice photonic crystal lens consisting of rods with refractive index n = -1: (a), (b) flat lens with ordered arrangement of rods; (c) flat lens with disordered rods; (d) concave lens with disordered rods. Reproduced with permission from [156], V.V. Sergentu et al., Phys. Stat. Sol. RRL 2, 242 (2008). © 2008, Wiley-VCH Verlag GmbH & Co.KGaA.

There is a clear superlensing effect, S/2 600 °C [205]. The different crystalline structures of titania have different material properties including density, index of refraction, and catalytic properties. Recently, titania also gained interest as a material used in photonic band gap crystals for the visible spectrum of light due to its high index of refraction (nrutile ≈ 2.9) [206] and low absorption [207,208]. Electrochemical oxidation of Ti foils allows one to prepare a variety of

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 211

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

porous titania structures therefore enlarging the area of TiO2 applications in optoelectronic and photonic devices [156]. Taking into account the possibility of doping porous titania templates with rare earth and transition metal ions and the morphology controlled light scattering properties, one can expect that luminescent materials prepared on porous TiO2 templates are prospective for random laser applications. The morphology, structural and optical properties of porous titania layers have been investigated as a function of technological conditions of preparation and postelectrochemical-oxidation thermal treatment. For this purpose, technological conditions for the preparation of porous TiO2 layers with controlled morphology and porosity on the basis of Ti foils were developed. Technological conditions for the preparation of porous TiO2 layers on the basis of Ti foils include rinsing and sonicating in isopropyl alcohol, drying and anodizing. The samples were anodized in aqueous HF solutions with various additives. Morphologies in the form of arrays of nanotubes were produced. The optimum concentration of the electrolyte was found to be 0.5 wt %. It was found that by applying various anodizing conditions, it is possible to control the diameter and the length of nanotubes. The diameter of nanotubes increases monotonously from 30 to 100 nm with increasing the voltage from 5 to 30 V at fixed treatment duration of 30 min (see Figure 65). At the same time, the length of nanotubes increases from 70 nm to 400 nm. The geometrical parameters of the produced structures are also controlled by the duration of anodization. The increase of the anodization duration from 3 min to 30 min at constant voltage of 10 V results in the increase of the nanotube diameter from 30 to 60 nm, and the length from 70 to 200 nm.

Figure 65. TiO2 nanotubes produced by etching Ti foils in 0.5 wt % aqueous HF solutions.

A mixture of HF and H3PO4 solutions in ethylene glycol was shown to be suitable for the preparation of TiO2 nanotubes with diameters up to 250 nm and length up to 40 µm. For this, the Ti foils were treated under 120 V during 4 hours (see Figure 66).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

212

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

Figure 66. TiO2 nanotubes produced by etching Ti foils in a mixture of HF and H3PO4 in ethylene glycol.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The influence of thermal treatment upon the structural properties of TiO2 nanotubes was investigated by means of X-ray diffraction analysis and Raman scattering. The analysis of Raman spectra (Figure 67a) demonstrates that the as prepared samples are amorphous. With increasing the temperature of annealing to 300 oC an anatase structure is formed. Starting from 500 oC a rutile structure is produced which coexist with the anatase structure. A complete phase transition to the rutile structure occurs at 700 oC.

Figure 67. (a) Raman spectra of TiO2 nanotubes as-grown (1); annealed at 300 oC (2), 600 oC (3), 700 o C (4). (b) XRD pattern of TiO2 nanotubes annealed at 500 oC.

Anatase is tetragonal, with two TiO2 formula units (six atoms) per primitive cell. The space group is D4h19 (I4/amd). The 18-dimensional reducible representation generated by the atomic displacements contains the zone-center (k=0) modes: 3 acoustic modes and 15 optical modes. The irreducible representations corresponding to the 15 optical modes are 1A1g + 1A2u + 2B1g + 1B2u + 3Eg +2Eu. Three modes are infrared active, the A2u mode and the two Eu modes. The B2u mode is silent. The remaining six modes corresponding to symmetries A1g + 2B1g + 3Eg are Raman active. The Raman shift for these phonons is 514 cm-1 for the A1g mode, 399 cm-1 and 514 cm-1 for the B1g modes, and 144 cm-1, 197 cm-1 and 639 cm-1 for the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 213 Eg modes [209]. Therefore, the A1g and one of the B1g modes overlap. The two Eg modes at 144 cm-1, 197 cm-1 are outside of the range of measured Raman shifts. The rutile structure of titania belonging to the space group D4h14 with two TiO2 molecules per unit cell [210]. The cations are located at sites with D2h symmetry and the anions occupy sites with C2v symmetry. The Ti-ions are surrounded by six oxygen ions at the corners of a slightly distorted octahedron, while the three Ti-ions coordinating each oxygen ions lie in a plane at the corners of a nearly equilateral triangle. According to the factor group analysis, there are fifteen optical phonon modes with the irreducible representation as given in [211]. There are four Raman active modes with symmetries B1g, Eg, A1g, and B2g. The Raman shift for these phonons is 143 cm-1 for the B1g mode, 447 cm-1 for the Eg mode, 612 cm-1 for the A1g mode, and 826 cm-1 for the B2g [212]. The B1g mode at 143 cm-1 and the B2g mode at 826 cm-1 are outside of the range of measured Raman shifts. The XRD analysis corroborates the Raman data. The XRD pattern of TiO2 nanotubes annealed at 500 oC (Figure 67b) demonstrates the coexistence of anatase and rutile phases. Figure 68a presents the PL spectra of TiO2 nanotubes annealed at different temperatures. The temperature dependence of a sample annealed at 500 oC is shown in Figuer 68b. The luminescence from both anatase and rutile phases is observed at low temperatures (10 K). For samples annealed at temperatures up to 400 oC the luminescence measured in the spectral range from 370 to 500 nm is dominated by the near bandgap emission from the anatase phase which includes two narrow lines at 371 nm (3.34 eV) and 372 nm (3.33 eV) followed by several phonon replica with phonon energy equal to 50 meV. The luminescence of samples annealed at temperatures above 700 oC comes from the rutile phase and it consists of a near bandgap emission band at 402 nm and a wide blue band with the maximum around 423 nm al low temperatures. The near bandgap emission is quenched with increasing temperature, while the blue band is persistent up to room temperature it being red-shifted by increasing temperature. Both phases contribute to the low temperature luminescence in samples annealed in the temperature range from 400 oC to 600 oC, the room temperature luminescence being always determined by the rutile phase.

Figure 68. (a) PL of TiO2 nanotubes annealed at different temperatures. (b) PL of TiO2 nanotubes annealed at 500 oC measured at different temperatures.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

214

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

As concerns the nature of the observed PL bands, previously two sharp lines peaking at 3.31 and 3.37 eV have been observed in the near bandgap PL spectra of anatase titania [213]. These lines were interpreted as defect-trapped-exciton related although the free-exciton origin of the 3.31 eV peak was also argued. Apart from this possible nature of the PL lines at 3.34 eV and 3.33 eV observed in our samples, their relation to free-to-bound transitions cannot be excluded. A band at 402 nm and another one at 439 nm have been previously observed in the cathodoluminescence spectra of rutile phase TiO2 [214]. The low-temperature photoluminescence spectrum of rutile TiO2 was found to comprise a peak at 3.031 eV (409 nm) which was attributed to 2pxy dipole-allowed second-class excitonic transitions [215]. A band was observed at 450 in the cathodoluminescence spectra of polycrystalline rutile at room temperature [216]. Previous studies report the presence of shallow traps or deep defect levels associated with the presence of oxygen vacancies which are formed in reduced or oxidized rutile crystals and films [217,218]. The energies of the shallow traps range from 0.27 to 0.87 eV below the conduction band. Taking this into account, one can suggest that the PL band observed at 402 nm is excitonic, while the band at 423 nm in our rutile samples as well as the previously observed cathodoluminescence bands at 439 nm and 450 nm can be attributed to free to bound electronic transitions involving traps below the conduction band. Thus, electrochemical treatment allows one to fabricate porous titania templates with controlled morphology and crystallographic structure. The morphology is controlled by the technological conditions of electrochemical oxidation of Ti foils, while the crystallographic structure is controlled by the conditions of thermal treatment.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

5. CONCLUSIONS This review demonstrates that electrochemistry is an efficient and cost-effective approach for nanostructuring semiconductor surfaces and preparation of porous dielectric oxide surfaces on metallic foils. Current line oriented pores, or crystallographically oriented pores can be produced in semiconductor substrates depending on the processes which govern the dissolution of the material during electrochemical treatment. Crystallographically oriented pores are produced if direct dissolution of the material dominates the electrochemical processes. On the other hand, current liner oriented pores are formed if the dissolution via oxide formation is dominating the whole dissolution process. An important feature of the current line oriented pores is that they self-arrange locally in a hexagonal closed packed lattice. Therefore, a variety of architectures were demonstrated, including periodic spatial distribution of pores achieved by anodic etching governed by self-organization phenomena. The sharp transition from current-line oriented pores to crystallographically oriented pores can be used to induce periodicity in the third dimension, and therefore to produce a threedimensional periodic structure. Therefore, 2D- and 3D photonic crystals can be produced. In order to make electrochemistry widely accessible it should be environmentallyfriendly. Anodization in salty water proves to be a cost-effective and environmentallyfriendly tool for spatial nanostructuring of materials and nonlithographic manufacturing of semiconductor nanotemplates for nanofabrication. Preparation of nanostructured surfaces in salty water has been demonstrated with InP, InAs, GaAs, and CdSe. Self-organized quasi-

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 215 ordered two-dimensional hexagonal arrays of pores with diameters as low as 70 nm have been prepared on n-InP substrates subjected to anodic etching in aqueous solution of NaCl. Nanostructuring of surfaces induces new physical properties many of which are favourable for space applications. For instance, because of the nanoscale nature of light absorption and photocurrent generation in solar energy conversion, the advent of methods for controlling inorganic materials on the nanometer scale opens new opportunities for the development of future generations of solar cells. It was shown that the introduction of porosity in the GaAs photoelectrode leads to a considerable photosensitivity increase in the longwavelength region near the bandgap which results in an increase of the output power by a factor of four in comparison with the cell based on bulk GaAs electrode. To promote the development of photoelectrochemical cells based on porous semiconductors, technological conditions for effective surface passivation should be assured. An effective passivation of the surface was shown to occur during anodization of many semiconductors in different electrolytes, including aqueous solution of NaCl. Apart from the importance of light trapping in porous networks for photocurrent generation in solar energy conversion, the enormous light scattering in the porous medium is very important for the development of random laser media. For this purpose, methods of introducing optical gain properties to highly scattering medium should be developed. Doping of porous semiconductor GaP, GaAs and InP as well as dielectric Al2O3 and TiO2 templates with rare earth elements (Eu, Er) and transition metals (Cr, Ti) has been proposed to address this issue. Nanostructuring of semiconductor surfaces was shown to significantly enhance also the nonlinear optical properties. An efficient optical second harmonic generation was observed in porous GaP and InP membranes. Enhanced THz emission was reported in nanostructured InP layers under excitation by ultrashort laser pulses. It was suggested that the enhanced nonlinear optical response is related to strong enhancements of the local field within the porous network. Apart from that, electrochemistry proved to be a powerful tool for introducing optical anisotropy in semiconductor materials. Enhanced optical nonlinearities accompanied by artificial anisotropy make electrochemically etched III-V compounds promising for advanced nonlinear optical applications. Extremely important for space applications is the demonstration of enhanced radiation hardness induced by nanostructuring of GaN and ZnO surfaces. More than one order of magnitude enhancement of radiation hardness as compared to bulk material has been observed in nanostructured GaN and ZnO layers. This observation is especially important taking into account that ZnO and GaN are radiation hard materials even in the bulk form as compared to other semiconductors. The strong radiation hardness of ZnO and GaN coupled with excellent optical and electrical properties suggest that devices based on these materials are promising for space applications. Nanostructured semiconductor and dielectric templates represent additionally an important basis for the production of novel nanocomposites, including smart and negativerefractive index materials for optoelectronic and photonic applications. Especially important in this regard are wide-band gap compounds such as ZnSe, GaN, ZnO, and TiO2.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

216

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

REFERENCES [1] [2] [3] [4] [5] [6] [7] [8]

[9]

[10]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[11]

[12] [13]

[14] [15] [16]

[17] [18]

L.T. Canham. Silicon quantum wire array fabrication by electrochemical and chemical dissolution of wafers. Appl. Phys. Lett. 57 1046 (1990). V. Lehmann and U. Gösele. Direct observation of self-focusing near the diffraction limit in polycrystalline silicon film. Appl. Phys. Lett. 58 856 (1991). T. Yonehara, presented at Porous Semiconductors Science and Technology (PSST 2002), Tenerife, Spain 11-15 March (2002). O. Jessensky, F. Müller, and U. Gösele. Microstructure and photoluminescence. Thin Solid Films 297 224 (1997). J. S. Shor, I. Grimberg, B.-Z. Weiss, and A. D. Kurtz. Appl. Phys. Lett. 62 2836 (1993). R.B. Wehrspohn and J. Schilling. Electrochemically prepared pore arrays for photonic crystal applications. MRS Buletin 26 623 (2001). A.I. Belogorokhov, V.A. Karavanskii, A.N. Obraztsov, and V.Yu. Timoshenco. Intense photoluminescence in porous gallium phosphide. JETP Lett. 60 264 (1994). A.Anedda, A. Serpi, V.A. Karavanskii, I.M. Tiginyanu, and V.M. Ichizli. Timeresolved blue and ultraviolet photoluminescence in porous GaP. Appl. Phys. Lett. 67 3316 (1995). K. Kuriyama, K. Ushiyama, K. Ohbora, Y. Miyamoto, and S. Takeda. Characterization of porous GaP by photoacoustic spectroscopy: The relation between band gap widening and visible photoluminescence. Phys. Rev. B 58 1103 (1998). B.H. Erne, D. Vanmaekelbergh, and J.J. Kelly. Morphology and strongly enchanced photoresponce of GaP electrodes made porous by anodic etching. J. Electrochem. Soc., 143 305 (1996). F. Iranzo Marin, M.A. Hamstra, and D. Vanmaekelbergh. Morphology and Strongly Enhanced Photoresponse of GaP Electrodes Made Porous by Anodic Etching. J. Electrochem. Soc. 3 1137 (1996). F.J.P. Schuurmans, D. Vanmaekelbergh, J. van de Lagemaat, and A. Lagendijk. Strongly Photonic Macroporous Gallium Phosphide Networks. Science 284 141 (1999). F.J.P. Schuurmans, M. Magens, D. Vanmaekelbergh, and A. Lagendijk. Light Scattering near the Localization Transition in Macroporous GaP Networks. Phys. Rev. Lett. 83 2183 (1999). J. Gomez Rivas, A. Lagendijk, R. W. Tjerkstra, D. Vanmaekelbergh, and J.J. Kelly. Tunable photonic strength in porous GaP. Appl. Phys. Lett. 80 4498 (2002). I.M. Tiginyanu, G. Irmer, J. Monecke, and H.L. Hartnagel. Micro-Raman-scattering study of surface-related phonon modes in porous GaP. Phys. Rev. B 55 6739 (1997). I.M. Tiginyanu, V.V. Ursaki, Y.S. Raptis, V. Stergiou, E. Anastassakis, H.L. Hartnagel, A. Vogt, B. Prevot, and C. Schwab. Raman Modes in Porous GaP under Hydrostatic Pressure. Phys. Status Solidi B 211 281 (1999). A. Liu and C. Duan. Preparation and Raman scattering study of pore arrays on an InP (100) surface. Physica E 9 723 (2001). A. Sarua, J. Monecke, G. Irmer, I.M. Tiginyanu, G. Gärtner, and H.L. Hartnagel. Frohlich modes in porous III–V semiconductors. J. Phys.: Condens. Matter 13 6687 (2001).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 217 [19] A. Milekhin, V. Ursaki, L. Sirbu, A. Toropov, I. Tiginyanu, D. R. T. Zahn. Raman scattering by porous structures with InAs quantum dots. Phys. Status Solidi C 6 883885 (2009). [20] E. Monaico, I. M. Tiginyanu, V. V. Ursaki, A. Sarua, M. Kuball, D. D. Nedeoglo and V. P. Sirkeli. Photoluminescence and vibrational properties of nanostructured ZnSe templates. Semicond. Semicond. Sci. Technol. 22 1115–1121 (2007). [21] G. Irmer, E. Monaico, I.M. Tiginyanu, V.V. Ursaki, G.V. Kolibaba, D.D. Nedeoglo. Fröhlich vibrational modes in porous ZnSe studied by Raman scattering and Fourier transform infrared reflectance. J. Phys. D: Appl. Phys. 42 045405 (2009). [22] I.M. Tiginyanu, I. V. Kravetsky, J. Monecke, W. Cordts , G. Marowsky, and H. L. Hartnagel. Semiconductor Sieves as Nonlinear Optical Materials. Appl. Phys. Lett. 77 2415-2417 (2000). [23] I.M. Tiginyanu, I.V. Kravetsky, S. Langa, G. Marowsky, J. Monecke, H. Föll. Porous III-V compounds as nonlinear optical materials. Phys. Status Solidi A 197 549-555 (2003). [24] M. Reid, I. Cravetchi, R. Fedosejevs, I. M. Tiginyanu, L. Sirbu and Robert W. Boyd. Enhanced nonlinear optical response of InP(100) membranes, Phys. Rev. B 71 081306(R) (2005). [25] M. Reid, I. V. Cravetchi, R. Fedosejevs, I. M. Tiginyanu and L. Sirbu. Enhanced terahertz emission from porous InP (111) membranes. Appl. Phys. Lett. 86 021904 (2005). [26] E. Kukino, M. Amiotti, T. Takizawa, and S. Arai. Anisotropic refractive index of porous InP fabricated by anodization of (111) surface. Jap. J. Appl. Phys. 34 177 (1995). [27] V. V. Ursaki, N. N. Syrbu, S. Albu, V. V. Zalamai, I. M. Tiginyanu and R. W. Boyd. Artificial birefringence introduced by porosity in GaP. Semicond. Sci. Technol. 20 745–748 (2005). [28] S. Singh. Handbook of Lasers with Selected Data on Optical Technology. Ed. R.J. Pressley, Cleveland, Ohio, 1971. [29] S. Langa, S. Lölkes, J. Carstensen, M. Hermann , G. Böttger, I. M. Tiginyanu, and H. Föll. Engineering the morphology of porous InP for waveguide applications. Phys. Status Solidi C 2 3253–3257 (2005). [30] S. Langa, S. Frey, J. Carstensen, H. Foll,I. M. Tiginyanu, M. Hermann,and G. Bottgerd. Waveguide Structures Based on Porous Indium Phosphide. Electrochemical and SolidState Letters 8 C30-C32 (2005). [31] C. Youtsey, I. Adesida, and G. Bulman. Highly anisotropic photoenhanced wet etching of n-type GaN. Appl. Phys. Lett. 71 2151 (1997). [32] C. Youtsey, L. T. Romano, and I. Adesida. Gallium nitride whiskers formed by selective photoenhanced wet etching of dislocations. Appl. Phys. Lett. 73 797 (1998). [33] A. R. Stonas, P. Kozodoy, H. Marchand, P. Fini, S. P. DenBaars, U. K. Mishra, and E. L. Hu. Backside-illuminated photoelectrochemical etching for the fabrication of deeply undercut GaN structures. Appl. Phys. Lett. 77 2610 (2000). [34] A. R. Stonas, T. Margalith, S. P. DenBaars, L. A. Coldren, and E. L. Hu. Development of selective lateral photoelectrochemical etching of InGaN/GaN for lift-off applications. Appl. Phys. Lett. 78 1945 (2001).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

218

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

[35] C. Díaz-Guerra, J. Piqueras, V. Popa, A. Cojocaru, and I. M. Tiginyanu. Spatially resolved cathodoluminescence of GaN nanostructures fabricated by photoelectrochemical etching. Appl. Phys. Lett. 86 223103 (2005). [36] S. Nakamura, G. Fasol, and I. Davies. The Blue Laser Diode: GaN Based Light Emitters and Lasers. Berlin: Springer-Verlag Telos, June 1, 1997, 343p. [37] B. Gil. Group III Nitride Semiconductor Compounds: Physics and Applications. Series on Semiconductor Science and Technology, vol. 6. Oxford University Press, May 1, 1998, 470p. [38] H. Morkoc. Nitride Semiconductors and Devices. Springer Series in Materials Science, vol 32. Berlin: Springer-Verlag, November 1, 1999, 488p. [39] F. A. Harraz, K. Kamada, T. Kobayashi, T. Sakka, and Y. H. Ogata. Random Macropore Formation in p-Type Silicon in HF-Containing Organic Solutions. J. Electrochem. Soc. 152 C213 (2005). [40] K. Rajeshwar, N. R. D. Tacconi, and C. R. Chenthamarakshan. Semiconductor-Based Composite Materials: Preparation, Properties, and Performance. Chem. Mater. 13 2765 (2001). [41] R. T. Sherman, Jr. and W. T. Ford. Semiconductor Nanoparticle/Polystyrene Latex Composite Materials. Langmuir 21 5218 (2005). [42] R. W. Boyd, R. J. Gehr, G. L. Fischer, and J. E. Sipe. Nonlinear optical properties of nanocomposite materials. Pure Appl. Opt. 5 505 (1996). [43] P. V. Kamat. Photoinduced transformations in semiconductormetal nanocomposite assemblies. Pure Appl. Chem. 74 1693 (2002). [44] G. Sauer, G. Brehm, S. Schneider, K. Nielsch, R. B. Wehrspohn, J. Choi, H. Hofmeister, and U. Gosele. Highly ordered monocrystalline silver nanowire arrays. J. Appl. Phys. 91 3243 (2002). [45] H. Zeng, S. Michalski, R. D. Kirby, D. J. Sellmyer, L. Menon, and S Bandyopadhyay. Effects of surface morphology on magnetic properties of Ni nanowire arrays in selfordered porous alumina. J. Phys.: Condens. Matter. 14 715 (2002). [46] J. Oh, Y. Tak, and J. Lee. Electrodeposition of Cu2O Nanowires Using Nanoporous Alumina Template. Electrochem. Solid-State Lett. 7 C27 (2004). [47] H. Föll, S. Langa, J. Carstensen, M. Christophersen and I.M. Tiginyanu. Pores in III-V Semiconductors. Advanced Materials 15 183 (2003). [48] S. Langa, I.M. Tiginyanu, J. Carstensen, M. Christophrsen and H. Föll. Self-organized growth of single crystals of nanopores. Appl. Phys. Lett. 82 278 (2003). [49] J. Carstensen, M. Christophersen, G. Hasse, and H. Foell. Parameter dependence of pore formation in silicon within the model of local current bursts. Phys. Status Solidi A 182 63 (2000). [50] G.S. Popkirov and R.N. Schindler. A new approach to the problem of good and bad impedance data in electrochemical impedance spectroscopy. Electrochimica Acta 39 2025 (1994). [51] F. Muller, A. Birner, U. Goesele, V. Lehmann, S. Ottow, and H. Foell. Structuring of macroporous silicon for applications as photonic crystals. Journal of Porous Materials 7 201 (2000). [52] I. M. Tiginyan, E. Monaico, S. Albu, and V. V. Ursaki. Environmentally friendly approach for nonlithographic nanostructuring of materials. Phys. Status Solidi RRL 1 98–100 (2007).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 219 [53] E. Yablonovitch, T. J. Gmitter, and K. M. Leung. Photonic band structure: The facecentered-cubic case employing nonspherical atoms. Phys. Rev. Lett. 67 2295 (1991). [54] P. Schmuki, L. E. Erickson, D. J. Lockwood, J. W. Fraser, G. Champion, and H. J. Labbe. Formation of visible light emitting porous GaAs micropatterns. Appl. Phys. Lett. 72 1039 (1998). [55] I. M. Tiginyanu, C. Schwab, J.-J. Grob, B. Prevot, H. L. Hartnagel, A. Vogt, G. Irmer, and J. Monecke. Ion implantation as a tool for controlling the morphology of porous gallium phosphide. Appl. Phys. Lett. 71 3829 (1997). [56] M. N. Ruberto, X. Zhang, R. Scarmozzino, A. E. Willner, D. V. Podlesnik, and R. M. Osgood. The Laser-Controlled Micrometer-Scale Photoelectrochemical Etching of III– V Semiconductors. J. Electrochem. Soc. 138 1174 (1991). [57] H. Föll. Properties of silicon-electrolyte junctions and their application to silicon characterization. Appl. Phys. A: Mater. Sci. Proc. 53 8 (1991). [58] X. G. Zhang. Mechanism of Pore Formation on n-Type Silicon. J. Electrochem. Soc. 138 3750 (1991). [59] P. C. Searson, J. M. Macaulay, and F. M. Ross. Pore morphology and the mechanism of pore formation in n-type silicon. J. Appl. Phys. 72 253 (1992). [60] H. Gerischer, P. Allongue, and V. Costa Kieling, Ber. Bunsenges. Phys. Chem. 97 753 (1993). [61] S. Langa, J. Carstensen, M. Christophersen, H. Föll, I. M. Tiginyanu. Observation of crossing pores in anodically etched n-GaAs. Appl. Phys. Lett. 78 1074-1076 (2001). [62] J. Carstensen, M. Christophersen, and H. Föll. Pore formation mechanisms for the SiHF system. Mater. Sci. Eng. B 69–70 23 (2000). [63] I. M. Tiginyanu, V. V. Ursaki, E. Monaico, E. Foca, and H. Föll. Pore Etching in III-V and II-VI Semiconductor Compounds in Neutral Electrolyte. Electrochemical and Solid-State Letters 10 D127-D129 (2007). [64] J.M. Bonard, K.A. Dean, B.F. Coll and C. Klinke. Field Emission of Individual Carbon Nanotubes in the Scanning Electron Microscope. Phys. Rev. Lett. 89 197602 (2002). [65] Z. Xu, X.D. Bai and E.G.Wang. Field emission of individual carbon nanotube with in situ tip image and real work function. Appl. Phys. Lett. 87 163106 (2005). [66] V.G. Litovchenko and A.A.Evtukh. Effect of electron field emission enhancement in structures with quantum well. Phys. Low-Dimens. Struct. 3/4 227 (1999). [67] V. G. Litovchenko and A.A. Evtukh A A. Vacuum Nanoelectronics. Handbook of Semiconductor Nanostructures and Nanodevices vol 3 Spintronics and Nanoelectronics, Ed. A A Balandin and K L Wang (Los Angeles: American Scientific Publishers) pp 153 234 (2006). [68] V. Litovchenko, A. Evtukh, M. Semenenko, A. Grygoriev, O. Yilmazoglu, H. L. Hartnagel, L. Sirbu, I. M. Tiginyanu and V. V. Ursaki. Electron field emission from narrow band gap semiconductors (InAs). Semicond. Sci. Technol. 22 1092–1096 (2007). [69] V. Litovchenko. Analysis of the band structure of tetrahedral diamondlike crystals with valence bonds: Prediction of materials with superhigh hardness and negative electron affinity. Phys. Rev. B 65 153108 (2002). [70] V. Bugrov, M. Levinstein, S. Rumyantserv and A. Zubrilov. Advanced Semiconductor Materials (New York: Wiley) pp. 1–30 (2001).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

220

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

[71] H. Hartnagel 1997 High temperature electronics based on compound semiconductors HT Electronics (London: Chapman and Hall) p. 162 (1997). [72] M. Shur and M.A. Khan. GaN based FE transistor HT Electronics (London: Chapman and Hall) p 296 (1997). [73] C.J. Edgcombe and U. Valdre. Microscopy and computational modelling to elucidate the enhancement factor for field electron emitters. J. Microsc. 203 188 (2001). [74] V. Litovchenko, A. Evtukh, O. Yilmazoglu, K. Mutamba, H. Hartnagel and D. Pavlidis. Gunn effect in field-emission phenomena. J. Appl. Phys. 97 044911 (2005). [75] F. Genereux, S.W. Leonard, H.M. van Driel, A. Birner, and U. Gösele. Large birefringence in two-dimensional silicon photonic crystals. Phys. Rev. B 63 161101 (2001). [76] D. Kovalev, G. Polisski, J. Diener, H. Heckler, N. Künzner, V. Yu. Timoshenko, and F. Koch. Strong in-plane birefringence of spatially nanostructured silicon. Appl. Phys. Lett. 78 916 (2001). [77] N. Kuenzner, D. Kovalev, J. Diener, E. Gross, V.Yu. Timoshenko, G. Polisski, F. Koch, and M. Fujii. Giant birefringence in anisotropically nanostructured silicon. Opt. Lett. 26 1265 (2001). [78] L.A. Golovan, V. Yu. Timoshenko, A.B. Fedotov, L.P. Kuznetsova, D.A. SidorovBiryukov, P.K. Kashkarov, A.M. Zheltikov, D. Kovalev, N. Künzner, E. Gross, J. Diener, G. Polisski, and F. Koch. Appl. Phys. B 73 31 (2001). [79] P.M. Hui and D. Stroud. Theory of second harmonic generation in composites of nonlinear dielectrics. J. Appl. Phys. 82 4740 (1997). [80] D.A.G. Bruggeman. The calculation of various physical constants of heterogeneous substances. I. The dielectric constants and conductivities of mixtures composed of isotropic substances. Ann. Phys. Lpz. 24 636 (1935). [81] D. J. Bergman.The dielectric constant of a composite material - A problem in classical physics. Phys. Rep. 43 377 (1978). [82] W.N. Herman and L.M. Hayden. Maker fringes revisited: second-harmonic generation from birefringent or absorbing materials. J. Opt. Soc. Amer. B 12 416 (1995). [83] L.A. Golovan, S.O. Konorov V.A. Melnikov, A.B. Fedotov, S.A. Gavrilov, A.M. Zheltikov, P.K. Kashkarov, V. Yu. Timoshenko, G.I. Petrov, L. Li, and V.V. Yakovlev. Efficient second-harmonic generation. JETP Letters 78 229 (2003). [84] V.A. Melnikov, L.A. Golovan, S.O. Konorov, D.A. Muzychenko, A.B. Fedotov, A.M. Zheltikov, V.Yu. Timoshenko, and P.K. Kashkarov. Second-harmonic generation in strongly scattering porous gallium phosphide. Appl. Phys. B. 79 225 (2004). [85] B. I. Greene, P. Saeta, D. Dykaar, S. Schmitt-Rink, and S. ChuangFar-infrared light generation at semiconductor surfaces and its spectroscopic applications. IEEE J. Quantum Electron. 28 2302 (1992). [86] S. Chuang, S. Schmitt-Rink, B. Greene, P. Saeta, and A. Levi. Optical rectification at semiconductor surfaces. Phys. Rev. Lett. 68 102 (1992). [87] D. You, R. Jones, P. H. Bucksbaum, and D. Dykaar. Coherent generation of farinfrared radiation from InP. J. Opt. Soc. Am. B 11 486 (1994). [88] X.-C. Zhang and D. H. Auston. Optoelectronic measurement of semiconductor surfaces and interfaces with femtosecond optics. J. Appl. Phys. 71 326 (1992). [89] T. Dekorsy, H. Auer, H. J. Baker, H. G. Roskos, and H. Kurz. THz electromagnetic emission by coherent infrared-active phonons. Phys. Rev. B 53 4005 (1996).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 221 [90] P.C.M. Planken, H.-K. Nienhuys, H.J. Bakker, and T. Wenckebach. Measurement and calculation of the orientation dependence of terahertz pulse detection in ZnTe. J. Opt. Soc. Am. B 18 313 (2001). [91] M. Nakajima, M. Hangyo, M. Ohta, and H. Miyazaki. Polarity reversal of terahertz waves radiated from semi-insulating InP surfaces induced by temperature. Phys. Rev. B 67 195308 (2003). [92] M. Nakajima, Y. Oda, and T. Suomoto. Competing terahertz radiation mechanisms in semi-insulating InP at high-density excitation. Appl. Phys. Lett. 85 2694 (2004). [93] W.Lukosz. Light emission by magnetic and electric dipoles close to a plane dielectric interface. III. Radiation patterns of dipoles with arbitrary orientation J. Opt. Soc. Am. 69 1495 (1979). [94] J. Charrier, E. Le Gorju, L Haji, and M. Guendouz. Optical Waveguides Fabricated from Oxidised Porous Silicon. J. Por. Mat. 7 243 (2000). [95] S. Enoch, G. Tayeb, and D. Maystre. Numerical evidence of ultrarefractive optics in photonic crystals. Opt. Com. 161 171 (1999). [96] B. Gralak, S. Enoch, and G. Tayeb. Anomalous refractive properties of photonic crystals. J. Opt. Soc. Am. A 17 1012 (2000). [97] J. B. Pendry. Negative Refraction Makes a Perfect Lens. Phys. Rev. Lett. 85 3966 (2000). [98] V.G.Veselago. The electrodynamics of substances with simultaneously negativ values of  and µ. Sov. Phys. Usp. 10 509 (1968). [99] Lie-Ming Li and Zhao-Qing Zhang. Multiple-scattering approach to finite-sized photonic band-gap materials. Phys. Rev. B 58 9587 (1998). [100] V.V. Sergentu, E. Foca, S. Langa, J. Carstensen, H. Foell, and I.M. Tiginyanu. Focusing effect of photonic crystal concave lenses made from porous dielectrics. Phys. Status Solidi A 201 R31 (2004). [101] J.D. Joannopoulos, R.D. Meade, and J.N. Winn. Photonic Crystals: Modeling the Flow of Light. Princenton University Press, Princenton, NJ, 1995. [102] Ph. Lalanne. Effective medium theory applied to photonic crystals composed of cubic or square cylinders. Applied Optics 35 5369 (1996). [103] H. Cao, in Progress in Optics, E. Wolf, Ed., Vol. 45 (North-Holland, Amsterdam, 2003). [104] X.H. Wu, A.Y. Yamilov, H. Noh, H. Cao, E.W. Seelig and R. Chang. Random lasing in closely packed resonant scatterers. J. Opt. Soc. Am. B 21 159 (2004). [105] H. Cao. Review on latest developments in random lasers with coherent feedback. J. Phys. A: Math. Gen. 38 10497 (2005). [106] C. Yuen, S. F. Yu, E. S. P. Leong, H. Y. Yang, S. P. Lau, N. S. Chen and H. H. Hng. Low-loss and directional output ZnO thin-film ridge waveguide random lasers with MgO capped layer. Appl. Phys. Lett. 86 031112 (2005). [107] H.-C. Hsu, C.-Y. Wu and W.-F. Hsieh. Stimulated emission and lasing of randomgrowth oriented ZnO nanowires. J. Appl. Phys. 97 064315 (2005). [108] V.M. Markushev, V.F. Zolin and Ch. M. Briskina. Luminescence and stimulated emission of neodymium in sodium lanthanum molybdate powders. Sov. J. Quantum Electron. 16 281 (1986).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

222

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

[109] Ch. M. Briskina, V. M. Markushev and N. E. Ter-Gabrielyan. Use of a model of coupled microcavities in the interpretation of experiments on powder lasers. Sov. J. Quantum Electron. 26 923 (1996). [110] S. V. Frolov, Z. V. Vardeny, K. Yoshino, A. Zakhidov and R. H. Baughman. Stimulated emission in high-gain organic media. Phys. Rev. B 59 R5284 (1999). [111] R. C. Polson, J. D. Huang and Z. V. Vardeny. Random lasers in π-conjugated polymer films. Sinth. Met. 119 7 (2001). [112] S. Klein, O. Gregut, D. Gindre, A. Bboeglin and K.D. Dorkenoo. Random laser action in organic film during the photopolymerization process. Opt. Express 13 5387 (2005). [113] R. C. Polson and Z. V. Vardeny. Organic random lasers in the weak-scattering regime. Phys. Rev. B 71 045205 (2005). [114] V.S. Letokhov. Generation of light by a scattering medium with negative resonance absorption. Zh. Eksp. Teor. Fiz. 53 1442 (1967) [Sov. Phys. J. Exp. Theoret. Phys. 26 835 (1968)]. [115] L. Sirbu, V. V. Ursaki, I. M. Tiginyanu, K. Dolgaleva and R. W. Boyd. Er- and Eu doped GaP-oxide porous composites for optoelectronic applications. Phys. Status Solidi RRL 1 R13 (2007). [116] L. Sirbu, V. V. Ursaki, I. M. Tiginyanu, K. Dolgaleva and R.W. Boyd. Red and green nanocomposite phosphors prepared from porous GaAs templates. J. Opt. A: Pure Appl. Opt. 9 401 (2007). [117] G. M. Williams, P. C. Becker, N. Edelstein, L.A. Boatner and M.M. Abraham. Excitation profiles of resonance electronic Raman scattering in ErPO4 crystals. Phys. Rev. B 40 1288 (1989). [118] J. Dexpert-Ghys, R. Mauricot and M.D. Faucher. Spectroscopy of Eu3+ ions in monazite type lanthanide orthophosphates LnPO4, Ln = La or Eu. J. Luminescence 69 203 (1996). [119] F. Zenia, C. Levy-Clement, R. Triboulet, R. Könenkamp, K. Ernst, M. Saad, and M. C. Lux-Steiner. Electrochemical texturization of ZnTe surfaces. Appl. Phys. Lett. 75 531 (1999). [120] R. Garuthara, M. Tomkiewicz, and R. Tenne. Photoluminescence of CdSe: The effect of photoetching. Phys. Rev. B 31 7844 (1985). [121] I. M. Tiginyanu, E. Monaico, V. V. Ursaki, V. E. Tezlevan, R.W. Boyd. Fabrication and photoluminescence properties of porous CdSe. Appl. Phys. Lett. 86 063115 (2005). [122] E. Monaico, V. V. Ursaki, I. M. Tiginyanu, Z. Dashevsky, V. Kasiyan, R. W. Boyd. Porosity-induced blueshift of photoluminescence in CdSe. J. Appl. Phys. 100 053517 (2006). [123] R. G. Wheeler and J. O. Dimmock. Exciton Structure and Zeeman Effects in Cadmium Selenide. Phys. Rev. 125 1805 (1962). [124] E. Monaico, V. V. Ursaki, A. Urbieta, P. Fernandez, J. Piqueras, R. W. Boyd, and I. M. Tiginyanu. Porosity-induced gain of luminescence in CdSe. Semicond. Sci. Technol. 19 L121 (2004). [125] R. Jäger-Waldau et al. Thin-film CdSe: Photoluminescence and electronic measurements. J. Appl. Phys. 64 2601 (1988). [126] Y. P. Varshni. Temperature dependence of the energy gap in semiconductors. Physica (Amsterdam) 34 149 (1967).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 223 [127] M.-C. Kuo, K.-C. Chiu, T.-H. Shih, Y.-J. Lai, C.-S. Yang, W.-K. Chen, D.-S. Chuu, M.-C. Lee, W.-C. Chou, S.-Y. Jeng, Y.-T. Shih, and W.-H. Lan, Optical Properties of Zn1-xCdxSe Epilayers Grown on (100) GaAs by Molecular Beam Epitaxy. Jpn. J. Appl. Phys., Part 1 43 5145 (2004). [128] O. Maksimov, W. H. Wang, N. Samarth, M. Munoz, and M. C. Tamargo. Optical properties of Zn0.5Cd0.5Se thin films grown on InP by molecular beam epitaxy. Solid State Commun. 128 461 (2003). [129] Al. L. Efros and M. Rosen. The electronic structure of semiconductor nanocrystals. Annu. Rev. Mater. Sci. 30 475 (2000). [130] H. J. Lozykowski and V. K. Shastri. Excitonic and Raman properties of ZnSe/Zn1−xCdxSe strained-layer quantum wells. J. Appl. Phys. 69 3235 (1991). [131] T. Li, H. J. Lozykowski, and J. L. Reno. Optical properties of CdTe/Cd1-xZnxTe strained-layer single quantum wells. Phys. Rev. B 46 6961 (1992). [132] Y. Marfaing. Self-compensation in II–VI compounds. Prog. Cryst. Growth Charact. 4 317 (1981). [133] B. Reinhold and M. Wienecke. Post growth n-type doping of ZnSe-bulk single crystals. J. Crystal Growth 204 434 (1999). [134] M. Prokesch, K. Irmascher, J. Gebauer and R. Krause-Rehberg. Reversible conductivity. J. Crystal Growth. 214/215 988 (2000). [135] G. N. Ivanova, D. D. Nedeoglo, N. D. Negeoglo, V. P. Sirkeli, I. M. Tiginyanu and V. V. Ursaki. Interaction of intrinsic defects with impurities in Al doped ZnSe single crystals. J. Appl. Phys. 101 063543 (2007). [136] E. Monaico, I. M. Tiginyanu, V. V. Ursaki, A. Sarua, M. Kuball, D. D. Nedeoglo and V. P. Sirkeli. Photoluminescence and vibrational properties of nanostructured. ZnSe templates. Semicond. Sci. Technol. 22 1115–1121 (2007). [137] G. Irmer, E. Monaico, I.M. Tiginyanu, G. Gartner, V.V Ursaki,.G. V. Kolibaba and D. D. Nedeoglo. Frohlich vibrational modes in porous ZnSe studied by Raman scattering and Fourier transform infrared reflectance. J. Phys. D: Appl. Phys. 42 045405 (2009). [138] M. Christophersen, S. Langa, J. Carstensen, I. M. Tiginyanu, and H. Föll. A comparison of pores in silicon and pores in III-V compound materials. Phys. Status Solidi A 197 197-203 (2003). [139] V. Lehmann. Porous silicon matrix for chemical synthesis and chromatograpy. Phys. Status Solidi A 202 1365 (2005). [140] G. Abstreiter, M. Cardona and A. Pinczuk. Light Scattering in Solids ed M Cardona (Berlin: Springer) p 147 (1975). [141] D.J. Olego, T. Marshall, J. Gaines and K. Shahzad. Phonon coupling to excitations of free and localized electrons in n-type ZnSe. Phys. Rev. B 42 9067 (1990). [142] H. Hartmann, R. Mach and B. Selle. Current Topics in Material Science vol 9, ed E Kaldis (Amsterdam:North-Holland) (1982). [143] K. Huang and A. Rhys. Theory of Light Absorption and Non-Radiative Transitions in F-Centres. Proc. R. Soc. Lond. A 204 406 (1950). [144] H.-M. Cheng, K.-F. Lin, H.-C. Hsu and W.-F. Hsieh. Size dependence of photoluminescence and resonant Raman scattering from ZnO quantum dots. Appl. Phys. Lett. 88 261909 (2006). [145] R.P. Wang, G. Xu and P. Jin. Size dependence of electron-phonon coupling in ZnO nanowires. Phys. Rev. B 69 113303 (2004)

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

224

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

[146] D. Routkevitch, T.L. Haslett, L. Ryan, T. Bigioni, C. Douketis and M. Moskovits. Synthesis and resonance Raman spectroscopy of CdS nano-wire arrays. Chem. Phys. 210 343 (1996). [147] A.P. Alivasatos, T.D. Harris, P.J. Carroll, M.L. Steigerwald and L.E. Brus. Electron– vibration coupling in semiconductor clusters studied by resonance Raman spectroscopy. J. Chem. Phys. 90 3463 (1989). [148] M.J. Seong, O.L. Micic, A.J. Nozik and A. Mascarenhas. Size-dependent Raman study of InP quantum dots. Appl. Phys. Lett. 82 185 (2003). [149] I.U. Schuchert, M.E. Toimil Molares, D. Dobrev, J. Vetter, R. Neumann, M. Martin. Electrochemical Copper Deposition in Etched Ion Track Membranes. J. Electrochem. Soc. 150 C189 (2003). [150] W. Lee, R. Scholz, K. Nielsch, U. Gosele. A Template-Based Electrochemical Method for the Synthesis of Multisegmented Metallic Nanotubes. Angew. Chem. Int. Ed. 44 6050 (2005). [151] G. Sharma, M.V. Pishko, C.A. Grimes. Fabrication of metallic nanowire arrays by electrodeposition into nanoporous alumina membranes: effect of barrier layer. J. Mater. Sci. 42 4738 (2007). [152] M. Motoyama, Y. Fukunaka, T. Sakka, Y.H. Ogata. Initial stages of electrodeposition of metal nanowires in nanoporous templates. Electrochim. Acta 53 205 (2007). [153] Ion Tiginyanu, Elena Monaico, Eduard Monaico. Ordered arrays of metal nanotubes in semiconductor envelope. Electrochemistry Communications 10 731–734 (2008). [154] D.C. Dumca, R. Riemenschneider, J. Miao, H.L. Hartnagel, B.R. Singh. Electrochemically Fabricated High-Barrier Schottky Contacts on n-InP and Their Application for Metal-Semiconductor-Metal Photodetectors. J. Electrochem. Soc. 143 1945 (1996). [155] E. Monaico, P. Tighineanu, S. Langa, H.L. Hartnagel, and I.Tiginyanu. ZnSe-based conductive nanotemplates for nanofabrication. Phys. Status Solidi RRL 3 97– 99 (2009). [156] V. V. Sergentu, I. M. Tiginyanu, V. V. Ursaki, M. Enachi, S. P. Albu, and P. Schmuki. Negative index material lenses based on metallo-dielectric nanotubes. Phys. Status Solidi RRL 2 242 (2008). [157] D. R. Smith, W. J. Padilla, D. C. Vier, S. C. Nemat-Nasser, and S. Schultz. Composite Medium with Simultaneously Negative Permeability and Permittivity. Phys. Rev. Lett. 84 4184 (2000). [158] R. A. Shelby, D.R. Smith, and S. Schultz. Experimental Verification of a Negative Index of Refraction. Science 292, 77 (2001). [159] A.K. Iyer and G. V. Eleftheriades, IEEE MIT-S International Microwave Symposium Digest 2002, Vol. 2, p. 1067. [160] M. Notomi. Theory of light propagation in strongly modulated photonic crystals: Refractionlike behavior in the vicinity of the photonic band gap. Phys. Rev. B 62 10696 (2000). [161] C. Luo, S.G. Johnson, J.D. Joannopoulos, and J.B. Pendry. Subwavelength imaging in photonic crystals. Phys. Rev. B 68 045115 (2003). [162] V. V. Sergentu, V. V. Ursaki, I. M. Tiginyanu, F. Foca, H. Föll, and R. W. Boyd. Design of negative-refractive-index materials on the basis of rods with a gradient of the dielectric constant. Appl. Phys. Lett. 91 081103 (2007).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 225 [163] V.V. Sergentu, V.V. Ursaki, I.M. Tiginyanu, F. Foca, H. Föll, and R. W. Boyd. Focusing slabs made of negative index materials based on inhomogeneous dielectric rods. Phys. Status Solidi A 203 R48 (2006). [164] V.V. E. Foca, V. V. Sergentu, F. Daschner, I. M. Tiginynau, V. V. Ursaki, R. Knöchel, and H. Föll. Superlensing with plane plates consisting of dielectric cylinders in glass envelopes. Phys. Status Solidi A 206 140– 46 (2009). [165] I. Bulu, H, Caglayan, and E. Ozbay. Negative refraction and focusing of electromagnetic waves by metallodielectric photonic crystals. Phys. Rev. B 72 045124 (2005). [166] P. B. Johnson and R.W. Christy. Optical Constants of the Noble Metals. Phys. Rev B 6 4370 (1972). [167] A. I.Cabuz, D. Felbacq, and D. Cassagne. Spatial dispersion in negative-index composite metamaterials. Phys. Rev. A 77 013807 (2008). [168] C. Youtsey, L. T. Romano, R. J. Molnar, and I. Adesida. Rapid evaluation of dislocation densities in n-type GaN films using photoenhanced wet etching. Appl. Phys. Lett. 74 3537 (1999). [169] P. Visconti, M. A. Reshchicov, K. M. Jones, D. F. Wang, R. Cingolani, R.J. Molnar, and D. J. Smith. Highly selective photoelectrochemical etching of nitride materials for defect investigation and device fabrication. J. Vac. Sci. Technol. B 19 1328 (2001). [170] J. L. Weyher, F. T. Tichelaar, H. W. Zandbergen, L. Macht, and P. R. Hageman. Selective photoetching and transmission electron microscopy studies of defects in heteroepitaxial GaN. J. Appl. Phys. 90 6105 (2001). [171] C. Díaz-Guerra. J. Piqueras, V. Popa, A. Cojocaru, and I. M. Tiginyanu. Spatially resolved cathodoluminescence of GaN nanostructures fabricated by photoelectrochemical etching. Appl. Phys. Lett. 86 223103 (2005). [172] V. Popa, I. M. Tiginyanu, V. V. Ursaki, O. Volciuc and H. Morkoc. A GaN-based twosensor array for methane detection in an ethanol environment. Semicond. Sci. Technol. 21 1518–1521 (2006). [173] U. Ozgur, Ya. I. Alivov, C. Liu, A. Teke, M.A. Reshchikov, S. Dogan, A. Avrutin, S.-J. Cho, M. Morkoc. A comprehensive review of ZnO materials and devices. J. Appl. Phys. 98 041301 (2005). [174] C. Jagadish and S.J. Pearton S J (ed). 2006 Zinc Oxide Bulk, Thin Films and Nanostructures, Processing, Properties and Applications (Amsterdam: Elsevier) p. 589 (2006). [175] Z.L. Wang. Zinc oxide nanostructures: growth, properties and applications. J. Phys.: Condens. Matter 16 R829 (2004). [176] V.V. Ursaki, E.V. Rusu, A. Sarua, M. Kuball, G.I. Stratan, A. Burlacu and I.M. Tiginyanu. Optical characterization of hierarchical ZnO structures grown with a simplified vapour transport method. Nanotechnology 18 215705 (2007). [177] Lionel Vayssieres. Growth of Arrayed Nanorods and Nanowires of ZnO from Aqueous Solutions. Adv. Mater. 15 464 (2003). [178] A.N. Redkin, Z.I. Makovei, A.N. Gruzintsev, S.V. Dubonos and E.E. Yakimov. Vapor phase synthesis of aligned ZnO nanorod arrays from elements. Inorganic Materials 43 253 (2007).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

226

I. M. Tiginyanu, V. V. Ursaki and E.V. Rusu

[179] A. Burlacu, V.V. Ursaki, V.A. Skuratov, D. Lincot, T. Pauporte, H. Elbelghiti, E.V. Rusu and I.M. Tiginyanu. The impact of morphology upon the radiation hardness of ZnO layers. Nanotechnology 19 215714 (2008). [180] A. Burlacu, V.V. Ursaki, D. Lincot, V.A. Skuratov, T. Pauporte, E. Rusu and I.M. Tiginyanu. Enhanced radiation hardness of ZnO nanorods versus bulk layers. Phys. Status Solidi RRL 2 68 (2008). [181] B.K. Meyer B K et al. Bound exciton and donor-acceptor pair recombinations in ZnO. Phys. Status Solidi B 241 231 (2004). [182] V.V. Ursaki et al. Multiphonon resonant Raman scattering in ZnO crystals and nanostructured layers. Phys. Rev. B 70 155204 (2004). [183] A.F. Kohan, G. Ceder, D. Morgan and C.G. Van de Walle. First-principles study of native point defects in ZnO. Phys. Rev. B 61 15019 (2000). [184] A. Zubiaga, F. Tuomisto, F. Plazaola, K. Saarinen, J.A. Garcia, J.F. Rommeluere, J. Zuniga-Perez and V. Munoz-Sanjose. Zinc vacancies in the heteroepitaxy of ZnO on sapphire: Influence of the substrate orientation and layer thickness. Appl. Phys. Lett. 86 042103 (2005). [185] V. V. Ursaki, V. V. Zalamai, A. Burlacu, J. Fallert, C. Klingshirn, H. Kalt, G. A. Emelchenko, A. N. Redkin, A. N. Gruzintsev, E. V. Rusuand I M. Tiginyanu. A comparative study of guided modes and random lasing in ZnO nanorod structures. J. Phys. D: Appl. Phys. 42 095106 (2009). [186] H. Zhou, M. Wissinger, J. Fallert, R. Hauschild, F. Stelzl, C. Klingshirn and H. Kalt. Ordered, uniform-sized ZnO nanolaser arrays. Appl. Phys. Lett. 91 181112 (2007). [187] R. Hauschild and H. Kalt. Guided modes in ZnO nanorods. Appl. Phys. Lett. 89 123107 (2006). [188] C. Klingshirn, Semiconductor Optics, 3rd ed. Springer-Verlag, Berlin (2007). [189] V.M. Markushev, M.V. Ryzhkov, Ch.M. Briskina, H. Cao, L.A. Zadorozhnaya, E.I. Givargizov, H. Zhong, S.-W. Wang and W. Lu. Laser Phys. 17 1109 (2007). [190] Y. Ling, H. Cao, A.L. Burin, M.A. Ratner, X. Liu and R.P.H. Chang. Investigation of random lasers with resonant feedback. Phys. Rev. A 64 063808 (2001). [191] S. M. Khanna, J. Webb, H. Tang, A. J. Houdayer, and C. Calone. 2 MeV proton radiation damage studies of gallium nitride films through low temperature photoluminescence spectroscopy measurements. IEEE Trans. Nucl. Sci. 47 2322 (2000). [192] F. Gaudreau, C. Carlone, Alain Houdayer, and S. M. Khanna. Spectral properties of proton irradiated gallium nitride blue diodes. IEEE Trans. Nucl. Sci. 48 1778 (2001). [193] V. V. Ursaki, I. M. Tiginyanu, O. Volciuc, V. Popa, V. A. Skuratov, H. Morkoc. Nanostructuring induced enhancement of radiation hardness in GaN Epilayers. Appl. Phys. Lett. 90 161908 (2007). [194] I. M. Tiginyanu, V. V. Ursaki, V. V. Zalomai, S. Langa, S. Hubbard, D. Pavlidis, and H. Föll. Luminescence of GaN nanocolumns obtained by photon-assisted anodic etching. Appl. Phys. Lett. 83 1551 (2003). [195] V. V. Ursaki, I. M. Tiginyanu, V. V. Zalamai, S. Hubbard, and D. Pavlidis. Optical characterization of AlN/GaN heterostructures. J. Appl. Phys. 94 4813 (2003). [196] D.C. Look, D.C. Reynolds, J.W. Hemsky, R.L. Jones and J.R. Sizelove. Production and annealing of electron irradiation damage in ZnO. Appl. Phys. Lett. 75 811 (1999).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

High-Performance Nanostructured Semiconductor and Metallo-Dielectric Layers … 227 [197] T.N. Morgan. Broadening of Impurity Bands in Heavily Doped Semiconductors. Phys. Rev. 139 A343 (1965). [198] W.P. Hsu, R. Yu, and E. Matijevic. Well-defined colloidal pigments. ii: monodispersed inorganic spherical particles containing organic dyes. Dyes Pigments 19 179 (1992). [199] M. Ferroni, V. Guidi, and G. Martinelli. Characterization of a nanosized TiO2 gas sensor. NanoStruct. Mater. 7 709 (1996). [200] V. Guidi, M. C. Carotta, M. Ferroni, G. Martinelli, L. Paglialonga, E. Comini, and G. Sberveglieri. Preparation of nanosized titania thick and thin films as gas-sensors. Sens. Actuators B 57 197 (1999). [201] N. Bonini, M. C. Carotta, A. Chiorino, V. Guidi, C. Malagu, G. Martinelli, L Paglialonga, and M. Sacerdoti. Doping of a nanostructured titania thick film: structural and electrical investigations. Sens. Actuators B 68 274 (2000). [202] Z. Ma, Y.Yue, X. Deng, and Z. Gao. Nanosized anatase TiO2 as precursor for preparation of sulfated titania catalysts. J. Molecular Catal. A: Chem. 178 97 (2002). [203] U. Bach, D. Lupo, P. Comte, J.E. Moser, F. Weissoertel, J. Salbeck, H. Spreitzer, and M. Grätzel. Solid-state dye-sensitized mesoporous TiO2 solar cells with high photon-toelectron conversion efficiencies. Nature 395 583 (1998). [204] H.W. Jaffe, Crystal Chemistry and Refractivity (Dover Publications, 1996). [205] F.A. Hummel, Introduction to Phase Equilibria in Ceramic Systems (Marcel Dekker, 1984). [206] G.R. Fowles, Introduction to Modern Optics (Dover Publications, New York, 1975). [207] R. Biswas, M. M. Sigalas, G. Subramania, and K-M. Ho. Photonic band gaps in colloidal systems. Phys. Rev. B 57 3701 (1998). [208] V. N. Manoharan, A. Imhof, J.D. Thorne, and D.J. Pine. Photonic Crystals from Emulsion Templates. Adv. Mater. 13 447 (2001). [209] T. Ohsaka, F. Izumi, and Y. Fujiki. Raman spectrum of anatase, TiO2. J. Raman Spectrosc. 7 321 (1978). [210] J. H. Xu, T. Jarlborg, and A. J. Freeman. Self-consistent band structure of the rutile dioxides NbO2, RuO2, and IrO2. Phys. Rev. B 40 7939 (1989). [211] R. Loudon. The Raman effect in crystals. Adv. Phys. 13 423 (1964). [212] S. P. S. Porto, P. A. Fleury, and T. C. Damen. Raman Spectra of TiO2, MgF2, ZnF2, FeF2, and MnF2. Phys. Rev. 154 522 (1966). [213] A. Suisalu, J. Aarik, H. Mändar, and I. Sildos. Spectroscopic study of nanocrystalline TiO2 thin films grown by atomic layer deposition. Thin Solid Films 336 295 (1998). [214] J.-M. Wu, W.-T. Wu, and H. C. Shih. Characterization of Single-Crystalline TiO2 Nanowires Grown by Thermal Evaporation. J. Electrochem. Soc. 152 G613 (2005). [215] A. Amtout and R. Leonelli. Optical properties of rutile near its fundamental band gap. Phys. Rev. B 51 6842 (1995). [216] R. Plugaru, A. Cremades, and J. Piqueras, J. Phys.: Condens. Matter 16 S261 (2004). [217] A. K. Ghosh, F. G. Wakim, and R. R. Addiss Jr. Photoelectronic processes in rutile. Phys. Rev. 184 979 (1969). [218] A. Rothschild, A. Levakov, Y. Shapira, N. Ashkenasy, and Y. Komem. Surface photovoltage spectroscopy study of reduced and oxidized nanocrystalline TiO2 films. Surf. Sci. 532 456 (2003).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: High Performance Coatings for Automotive and Aerospace… ISBN: 978-1-60876-579-9 Editor: Abdel Salam Hamdy Makhlouf, pp. 229-300 ©2010 Nova Science Publishers, Inc.

Chapter 8

PLATING OF NANO-COMPOSITES- OVERVIEW AND TRENDS Subir Kumar Ghosh*1,2 and Jean-Pierre Celis**2 Materials Processing Division, Bhabha Atomic Research Centre, Trombay, Mumbai-400094, (India)1 Katholieke Universiteit Leuven, Dept. MTM, B-3001 Leuven (Belgium)2

ABSTRACT

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Composite electrolytic and electroless plating has become of a large importance since it was recognized in the 1960s as a mature technology. One of the major breakthroughs on the use of composite plating in the aerospace, automotive, and electronic industries was the development of the oleophilic, electroplated nickel containing micron-sized SiC particles. To extend the durability of engineering components operated under highly demanding mechanical and chemical conditions, the search for smart coatings characterized by high hardness, high toughness, high wear resistance, and/or low friction properties as well as a good resistance to oxidation, is still relevant in today‘s technology. The interest in the 1990s for nano-materials has generated a large research and development activity on the plating of nano-composite coatings which is achieved by the incorporation of sub-micrometer to nanometer size particles into metal matrices. Such nano-composites can be classified in the following three types: namely, (a) the incorporation of nanometer size particles in a poly-crystalline matrix to achieve an Orowan‘s strengthening, (b) the incorporation of sub-micrometer size particles in a nanocrystalline matrix, and (c) the incorporation of nanometer size particles into a nanocrystalline matrix. Depending upon the structure, surface features, and mechanical properties achieved, such coatings can be applied either as very good wear-resistant coatings, anti-friction or self-lubricating coatings, or as catalytic active surfaces for applications like inner-lining in automotive exhaust systems for minimizing environmental pollution.

*

e-mail: [email protected], [email protected] e-mail: [email protected]

**

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

230

Subir Kumar Ghosh and Jean-Pierre Celis In this chapter, the actual basic understanding on composite plating is reviewed with a focus on recently acquired insights. The various models developed over the years are highlighted. Because of the hydrophilic nature of many second phase particles, it was experienced in previous investigations that there are limitations in the incorporation from aqueous baths of such particles beyond a certain percentage. Specific issues like the electroplating of Ni-SiC, Ni-SiO2, Ni-PTFE, and Al-SiO2 nano-composites are discussed. The effect of various process parameters like weight percent of particles suspended in the electrolyte, current density, additives, and hydrodynamic conditions on the incorporation of particles is discussed. The resulting structural and functional properties like hardness, wear-resistance, and friction are illustrated. Special attention is given on the tribological and tribocorrosion properties of such nano-composite coatings. To circumvent limitations in the co-deposition of nanometer size particles, new trends in the electroplating of nano-composite coatings from non-aqueous solutions are reviewed, and their actual limitations highlighted. Keeping these difficulties and the large interest for smart nano-composite coatings in mind, the quite recent concept of composite plating achieved by the in-situ formation of nano-particles at the electrode/electrolyte interface and their subsequent co-deposition is presented.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION The plating of composite coatings refers to processes in which micron, submicron, or nanometer size particles are suspended in an electrolyte. These particles are embedded into the metal matrix during metal-ion reduction, imparting special properties to the coatings depending on the degree and type of particles incorporated. The solid particles are kept in solution either by mechanical or air agitation, or by the addition of chemicals (surfactants) to create appropriate surface properties on particles to keep them suspended in the electrolyte. The nature of particles can be metallic, ceramic, organic or inorganic compounds, and minerals, and these particles should be insoluble in nature in the electrolyte. Even though the origins of composite electrodeposition lie in the early 1920s, most of the applications and modeling developments have been achieved over the last 40 years. The electrodeposition of metallic coatings containing inert particles can be traced back to studies on Cu-graphite coatings for self-lubricating surfaces in car-engines [1]. A significant development of electrodeposition of composites was noticed in the early 1950s and late 1960s [2]. The major breakthrough of composite plating took place in the automobile industry with the use of electrolytic Ni-SiC coatings in a generation of motor engines made of weightsaving aluminum alloys [3]. Some of these Ni-SiC coatings have been introduced under trade marks as e.g. NIKASIL, ELNISIL, etc. Many Europeans, Americans, and Japanese automobile companies have introduced this technology in their high-class motor-car models, and even in high speed and high performance engines. This break-through of Ni-SiC composite coatings is probably the best guarantee for a further increasing introduction by platers and designers of other composite coatings with unique physical, chemical, and/or mechanical properties. Besides the successful electrolytic co-deposition one should also note the development of electroless composite plating for the production of coatings like Ni-SiC, Ni-PTFE having outstanding wear resistance [3-4]. A list of industrial and technological applications of electrolytic and electroless composite coatings is provided in Chart-1.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

231

Chart-1. Various applications of composite coatings Materials

Co-CrC

Ni-P-diamond or SiC or WC or B4C

Ni-Al2O3

Ni-SiC

Ni-PTFE/Ni-PPTFE

Ni-TiO2 Ni-diamond

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Au-Co-Al2O3

Ni-Cr2O3-BN Ni-P-ZrO2 Co-P-IF-MoS2 Ni-P-SiO2-Rucomplex

Ni-P-containing light emitting particles

Co-IF WS2

Applications Bores in aircraft components: Bores coated to size with Co-CrC to reduce wear and fretting at high temperatures and to help eliminate costly welding and grinding operations. Piston rings: Cast iron piston rings coated with Co-CrC on external diameter for improved wear resistance/anti-scuffing. Aircraft nose wheel steering body: Coated with Co-CrC in the bore to provide wear resistance in sliding contact with Al alloy pistons. Aerospace air conditioning system pump components: Co-CrC to resist dry rubbing wear and fretting at 500 0C Aircraft components: Bearing journals, Piston heads, Hydraulic actuators splines, Struts, Engine mounts, Engine shafts, Landing gear components, hot zone hardware Plot tubes, Gyro components etc. coated with wear resistant coatings. Automobile shackle pin: EN58 hardened pins coated with Ni-Al203 on outside to provide resistance to wear, corrosion and abrasion. Glass moulds: Ni-Al2O3 deposited onto the face of cast iron moulds to provide oxidation and wear resistance. Automotive engine cylinder bores: Al-Cylinder bores coated to size with Ni-SiC to reduce wear at elevated temperatures and weight of the engine. Rotary engines: Ni-SiC particles used on rotary engines to reduce wear on rotating apex seal of Wankel engines Heat exchangers: Ni-PTFE to promote dropwise condensation and, hence, more efficient heat transfer than a continuous film Automotive clutch components: Ni-PTFE to control friction, corrosion and wear resistance on valves, carburetor components, pump rotors, aluminum air cylinders, mould cores, in certain precision applications, and in standard nuts and bolts). Aerospace industry: Servo valves, Gryo components coated with lubricating coatings. Automotive exhaust: Acts as catalyst and reduces the pollution level Electrothermal actuators for MEMS technology Electronic Contact: Au-Co-Al2O3 and Au-Co containing Cynanide complex for wear resistant electrical contact Helical Gear System in Military Aircraft: Electroplated Ni incorporating Cr2O3 hard nano-particles and BN soft lubricant performs as self-lubricated wear resistant coating (GE, Pratt Whitney). It also sowed validation in Automotive (Fernando Racing, Timkin) machinery and or aerospace (Eaton, Vetco), Medical (products) (Baxter). Catalytic electrode for hydrogen evolution reaction Self lubricating coating for dental applications Luminescent based Oxygen sensor Aerospace industry It offers an indicator layer, warning when the coating has worn off and replacement or recoating is necessary. The presence of colored emission from the coating can be visible in authenticating parts from a distinct source. A technician can verify the authenticity of a replacement part before installation on an aircraft. Thereafter, inspection by airline personnel or regulatory agencies is made easier. Any mandatory periodic replacement of aircraft parts is facilitated by this clear method of indication. Self-lubricating coating on orthodontic wire in medical applications

Apart from those effective or potential industrial applications, preliminary investigations of a number of other composite coatings show potential applicability in the industry as well as in high-end technology, e.g.: 1) Ni-Co-P-SiC: Electromagnetic compatibility purposes especially at GHz [6] 2) Ni-P-PTFE: Anti-fouling coatings in heat-exchangers and other areas [7-8]

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

232

Subir Kumar Ghosh and Jean-Pierre Celis 3) Ni-Carbon Nanotubes: Gas sensors [9] 4) Ni-nanodiamond: Microactuators [10].

With the progress in material science in all respects and the upsurge interest towards developing materials from micrometer to nanometer scale since the early 1990s, the plating of composites has evolved towards nano-composite plating [11-12]. The research on nanocomposite plating is today fueled by the increasing demand for very high wear-resistant, high temperature compatibility, high oxidation-resistant, and sometimes for very good, autocatalytic and self-lubricating properties for technological applications. Notwithstanding that, the research on nano-composite plating existed before the start of the ‗nano-era‘ in material‘s sciences [13-20]. For example, the electrodeposition of  -alumina particles with an average grain size of 50 nm into gold and Co-hardened Au matrix [13-18] was reported in the 1980s as wear-resistant, hardened composite coatings for electrical contact devices. Of course, with the availability and the development of efficient techniques for making nanoparticles of different sizes and morphologies over the last one and half decade, a lot of research was done aiming to the synthesis of composite coatings with superior properties. Plating processes that produce nano-composites can be classified as follows: a) the incorporation of nanometer size second phase particles in a polycrystalline matrix, b) the incorporation of submicron size particles into a nano-crystalline matrix, c) the incorporation of nm size particles into a nano-crystalline matrix, and d) the incorporation of nm size particles into an amorphous matrix.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Today, several techniques are available for depositing nano-composite coatings on substrates, namely: (i) Vapour Phase Deposition (e.g. Sputtering, Chemical vapour deposition, Ion-beam assisted deposition, Ion implantation,) (ii) Wet phase Deposition (e.g. Electrolytic deposition, Electroless deposition, Chemical bath deposition or sedimentation, Sol-gel method) (iii) Solid phase deposition (e.g. Air Plasma spray, Vacuum plasma spray, High-velocityoxy-fuel spray). The electrolytic and electroless deposition offer advantages over competing technologies such as physical and chemical vapor deposition since they require simpler instrumentation and operating conditions, and can be used to deposit coatings onto irregularly shaped substrates and in deep recesses [21-25]. Most vapor phase techniques are line-of-site processes, need high vacuum, are expensive, and require specific handling to coat engineering components with complicated shapes. Moreover, recent uses of such nano-composites in MEMS technology with structures of high aspect ratio demand advanced plating technologies, i.e. electrolytic and electroless deposition. However, a systematic development of nano-composite coatings for engineering applications needs an inter-correlation between synthesis, structural properties of coatings, functionality, and finally industrial validation (Fig. 1).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

233

Applications

Structural Aspects

Synthesis

Functionality

Industrial Validation

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 1. Schematic representation of development steps in nano-composite coatings for industrial purposes.

Previously, Celis et al. [26-30] have published several review papers on the electrochemical composite plating. In these papers, the use and the mechanism of electrolytic composite plating are discussed. Hovestad and Janssen have also published review articles regarding experimental facts on electrolytic composite coatings containing mainly micron size particles [31-32]. Recently, Low et al. [33] reviewed nano-composite plating where nano-sized particles are suspended in the electrolyte, and co-deposited with a metal. They discussed issues related to the effect of different experimental parameters on the incorporation of particles during electrolytic co-deposition. This chapter focuses on the effect of different deposition parameters and techniques (both electrolytic and electroless processes) on particle incorporation, i.e. to increase the particle content, to avoid agglomeration, their limitations, and future directions. A special highlight is given to different uses of composite/nano-composite coatings in the automotive and aerospace industry. Starting from basic concepts on strengthening mechanism and its dependence on particle number density, especially true for ‗nano-size‘ particles, a correlation between literature on micro-composites and recent developments on nano-composites, is made. Besides electrodeposition, the fundamental concepts of particle stability to retard particle agglomeration at smaller length scales is also provided so that electroplaters in laboratory/industry can get handful basics on nano-composite deposition. Different techniques like pulse current, reverse pulse electrolysis, and two step deposition are discussed in the case of composite plating. Amiability of various kinds of hydrodynamics to date to boost particle transport mechanism and hence the particle content on the coating are also brought out. Limitations of aqueous and non-aqueous deposition techniques and ways to overcome these difficulties are provided. In addition, the introduction of a new concept, namely the in-situ formation and the subsequent co-deposition of particles during electrolysis, is also presented.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

234

Subir Kumar Ghosh and Jean-Pierre Celis

2. STRENGTHENING MECHANISM IN NANO-COMPOSITES The incorporation of second phase particles in a metal matrix allows improving the intrinsic properties of materials. It increases hardness, tensile strength, fatigue strength, etc. [34-35]. These properties can even be further enhanced by a nano-structuring of materials and coatings. It is now established that nano-composite coatings exhibit a hardness significantly exceeding the one given by the rule of mixture: H(AaBb)=[aH(A)+bH(B)]/(a+b)

(1)

where H(A) is the hardness of phase A, H(B) is the hardness of phase B, and a the composition ratio of A and b the composition ratio of B in the mixture. H(AaBb) is the hardness of the mixture. It is, therefore, necessary to understand the basic mechanisms behind the enhancement of various intrinsic properties of nano-composites. These basic mechanisms are: (a) dispersion strengthening or Orowan strengthening, (b) Hall-Petch strengthening, and (c) multiphase-multi-component strengthening.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.1 Dispersion Strengthening Mechanism It is well known that hardness is closely linked to plastic deformation of a material under an applied load. Plastic deformation of a material is a measure of paths available for dislocation, a kind of line defect in the regular arrangement atoms inside the crystal, to move in a material. So, the softer a material is, the more is the plastic deformation, and the opposite is true for hard material under an identical load. Therefore, in order to enhance hardness, the basic idea is to arrest or block the paths for dislocation to move in crystals. Among various possibilities, it can be done by the incorporation of second phase particles in metal matrices. There is a general agreement among investigators that inert dispersoids block, or at least delay, the movement of dislocations in a metal matrix, and thus inhibit plastic deformation [Fig.2]. This mechanism, called Orowan strengthening, relates the plastic shear strain rate,  , to the moment of dislocation as:

  bv where

(2)

 is the geometric orientation factor, b the Burgers vector,  the density of mobile

dislocations, and v the average dislocation velocity. So the plastic shear strain rate is determined by the production of mobile dislocations, and the average dislocation velocity at the applied shear stress, plastic strain, plastic strain rate, and temperature. It allows a description of the macroscopic plastic flow either qualitatively or quantitatively. By incorporating particles into a monolithic matrix, the shear strength of the matrix,  m , becomes the shear strength of a composite,  Com . The difference,  Orowan , represents the extent of strengthening:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

 Orowan   Com   m

235 (3)

 Orowan is described by the Orowan-Ashby equation [2]:  

0.13Gmb  r  ln    b

where r is the particle radius,

(4)



the inter-particle spacing, b the Burgers vector of the

matrix, and Gm the shear modulus of the matrix. The inter-particle spacing,  , can be expressed [36-38] in terms of particle radius namely, r  d p / 2 with d p the average diameter of the particles: 1   3  1   d p  2V   1 p  

(5)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where V p is the volume fraction of reinforcing particles in a matrix. Therefore, it is clear that the strength of composites will increase by reducing the inter-particle spacing i.e. by increasing the volume fraction of particles or by incorporating smaller particles. That can be achieved by the inclusion of nanometer size particles into a metal matrix. In a similar theoretical calculation, Malone [39] showed that in order to effectively constrain all dislocations, the inter-particle spacing of dispersoids (inert particles) should be small, preferably less than one micron. He showed that if one co-deposits 1 µm particles, the desired volume percent of the dispersed oxide particles should be nearly 40 vol% in order to obtain an inter-particle spacing able to cage dislocations.

Figure 2. Mechanism of dispersion strengthening. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

236

Subir Kumar Ghosh and Jean-Pierre Celis

This would result in the case of electrodeposition or electroless plating in an extremely high brittleness. By using particles with a diameter less than 0.03 µm, the desired particle spacing can be achieved with approximately 2 vol% dispersiod oxide particles. In a similar calculation, Steinhouser et al. [40] showed that the optimum dispersion hardening effect in composite materials appears when a ductile matrix contains 10 nm particles spaced 100 nm apart. In that case, only 0.86 vol% particles in the composite material are required which is a really ―low‖ quantity of incorporated particles. Similarly, Greco [41] showed that for micron and submicron size particles, dispersion strengthening is achieved by a dispersion of fine particles with a diameter ranging from 0.01 to 1 m, and vol% ranging from 1 to 15. The optimum performance was found at an inter-particle spacing,  , between 0.5 and 5 m. Literature indeed predicts this low quantity, attributing it to physical and chemical effects between the particle and growing surface [42]. However, it has to be noted that an indefinite refinement of the second phase particles would not lead to an unlimited strengthening. Many investigators have shown that the Orowan particle strengthening mechanism is valid down to a grain size of the matrix of ~70 nm which contains second phase particles down to ~7 nm. A further refinement in the grain size of matrix and second phase particles would not lead to the expected strengthening [43-45] due to a particle agglomeration or a reverse Hall-Petch behavior [46].

2.2 Hall-Petch Strengthening Classically, one would expect an increase in yield stress,  y , with a decrease in grain

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

size, d , according to Hall-Petch equation [47-48]:

 y   0  kdn

6)

 o is the yield stress of the material with infinite grain size, i.e. in absence of any grain boundary, sometimes called lattice friction stress to move individual dislocations, k a constant, d the mean grain size, and n the mean grain size exponent (generally -1/2). where

Hardness, H , a measure of resistance to plastic deformation, can be correlated with yield stress as:

H  H 0  kdn

(7)

where H 0 is the hardness of matrix with infinite grain size, k and n are constants. In other words, the yield stress and, hence, the hardness increase as grain size decreases because pile-ups in fine-grained materials contain fewer dislocations, the stress at the tip of the pile-up decreases, and, thus, a larger applied stress is required to generate dislocations in adjacent grains. This particular phenomenon can operate in materials with nanometer size grains as well [34, 49], and indeed the Hall-Petch relationship is valid where the stress required for dislocations to propagate from one grain to another increases as grain size

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

237

reduces. In very small grains, this mechanism will break down because grains are unable to support dislocation pile-ups since Frank-Read dislocation loop source and dislocation multiplications can not operate. Then instead of increase, hardness was found to decrease with decreasing grain size. The slope k has then a negative value, and this has been termed as the inverse Hall–Petch relationship. Apart from the sign of k , there are also reports showing that the grain size exponent, n , can have very different values, e.g., -1, -1/2, -1/3 and -1/4. Some investigators have modified the normal Hall–Petch equation by modifying either the grain size exponent or the sign and magnitude of the slope, k . Typically, this is expected to occur for grain sizes below 10 nm for most metals [38, 39, 43-48]. As a consequence, a threshold value is expected at which a maximum yield stress can be achieved. However, experimentally, several systems such as Cu and Ni exhibit a reduced Hall–Petch slope ( k value in Eq. (7)) or even a negative one below a certain grain size [49-55]. Furthermore, the transition in the Hall–Petch slope normally occurs at grain sizes above 10 nm, where dislocation pile-ups are still possible. It was also noticed that the deformation mechanism evolves from a ‗dislocation controlled slip‘ to a ‗grain boundary sliding‘, thereby increasing plasticity [34]. Besides dislocation based models [56-58], there are diffusion based models [59-60], grain-boundary-shearing models [61-64], and two phase based models [65-67] describing the inverse Hall-Petch relation.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.3 Multiphase Multi-component Strengthening Recently in designing the technologically-superior nano-composite coatings for engineering applications, ultra-high hardness and high toughness are equally considered [6869]. To achieve that, a nano-composite coating should comprise at least two phases, namely a nano-crystalline phase and an amorphous phase, or two different nano-crystalline phases. The mechanical and tribological properties of nano-composite coatings depend on grain boundary effects and on synergistic effects of the composite constituents owing to the size effect [70]. The hardening or strengthening mechanisms are again related to the blocking of dislocation movements inside the coatings. Hindering dislocation movements in these coatings can be achieved either by grain boundary hardening, solid solution hardening, age hardening, or compressive stress hardening.

2.3.1 Grain Boundary Hardening In a recent review [71], Veprek analyzed the design criteria to produce super-hard or even ultra-hard nano-composite coatings. The combination of nano-crystalline transitionmetal nitrides and carbides with very thin amorphous Si3N4 or BN as grain boundary phase, seemed to end in ultra high hardness. Here basically, transition metal nitrides and carbides are chosen because these compounds are very low-compressibility solids that have very a high bulk modulus, E , or shear modulus, G . The very thin amorphous phase at grain boundary acts as a glue to hold the nano-crystals and to achieve ultra hardness, thus, avoiding the inverse Hall-Petch effect even for very small grain size crystals. However, these kinds of coatings are generally processed via plasma assisted deposition techniques. Details can be found in earlier reports [72-75].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

238

Subir Kumar Ghosh and Jean-Pierre Celis

2.3.2 Solid Solution Hardening Solid solution hardening results from a lattice distortion as a result of the insertion of alloying elements at interstitial locations or substitutional ones for some host atoms. It is, perhaps, the oldest hardening method used in bulk solids. In nano-composite thin films or coatings, the same principle works, too. This is basically achieved by the insertion of a C atom in an AB matrix to from a non-equilibrium, supersaturated solid solution, say ACB. Material ACB becomes harder compared to AB, since the gliding of dislocations eventually formed inside the crystallites, is hindered by the strain exerted by the insertion of C. Such coatings like TiCN [76], TiAl-N [77], and CrZrN [78] are very popular examples.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2.3.3 Age Hardening Age hardening in nano-composite coatings is achieved by annealing non-equilibrium supersaturated solid solutions. This process leads to the formation of fine grain crystals resulting in a hardness increase due to spinodal decomposition. This is very similar to bulk material ―age‖ hardening. Examples are (Ti,Al)N [79] and Ti(B,N) [80] obtained by nonequilibrium deposition techniques. 2.3.4 Compressive Stress Hardening During the plasma processing of nano-composite coatings, ion bombardment at low temperature was used to increase density, and to modify the morphology of films [81]. This process densifies materials by removing the loose atoms from the surface because of a momentum transfer, and sometimes leads to the rearrangement of atoms within the crystal generating compressive stresses in the coatings, so the coatings harden. This is called compressive stress hardening. It can be seen in coatings like (Ti,Al,V)N and TiN [82-83]. Similarly, the toughness of nano-composites can also be increased by different mechanisms as putting various ductile phases, phase transformation, putting high compressive stress, and forming graded inter-layers in the coatings. Details can be found in a review article by Zhang et al. [84].

2.4 Conclusion The aforementioned strengthening mechanism in nano-composite coatings highlights a few factors regarding the incorporation of nano-size particles: a) A very low volume fraction of nano-size particles is sufficient to achieve similar or even superior mechanical properties in comparison to micron size particles. This is possible thanks to a reduced inter-particle distance and hence a higher ‗number density‘ and better entrapment of particles in crystals or intra-crystalline regions unlike micron size particles generally trapped at the inter-crystalline region. Indeed, dislocations are caged in a small region surrounded by fine second phase particles and get much less room for movement under an applied load, thereby strengthening the whole matrix. However, the entrapment of micron size particles may lead to a ‗dispersion strengthening’ or ‗particle strengthening’ depending upon their distribution in a matrix. A homogeneous distribution across the matrix is here a more

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

239

crucial factor in comparison to nano-size particles. Whereas, the ‗number density‘ of the nanosize particles within the matrix is the more important factor in determining intrinsic (strength, hardness, toughness, etc.), and extrinsic properties (wear, coefficient of friction, corrosion, oxidation resistance, etc.) of the composite rather than classical ‗volume fraction‘ or ‗weight fraction‘ of particles. b) Because of their very high surface area, nano-size particles have a very high tendency to agglomerate in electrolytes. The smaller the particle size is, the larger the affinity to form an agglomerate. Besides this, for a given electrolyte, the concentration of ions, the dielectric constant of the solution, and the particles zetapotential dictate particle agglomeration. So, in order to deposit nano-size particles as single entities at electrode surfaces, efforts have to be directed to avoid assembling due to Brownian motion or due to forced mechanical movement, by selecting a suitable bath chemistry and localized stirring as we will next section. For industrial applications, one has to keep in mind that a stable electrolyte is required for continuous operation.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

3. PROCESS CONTROL Previous discussion on strengthening mechanisms in nano-composite coatings emphasizes that the particle number density, i.e. the number of incorporated nano-size particles per unit volume without or with a minimum agglomeration, determines the composite characteristics. Therefore, to optimize a particular end-use extrinsic property or functional property, one has to achieve a controlled particle incorporation rate maintaining homogeneity and number density. Before going into a detailed discussion on the various process parameters that affect the particle deposition rate, let us visualize the basic processes involved when a particle is put into an electrolyte till its incorporation into a metal matrix. The whole process can be divided into three main steps, namely:   

Step 1: Particle stability in the electrolyte, Step 2: Particle transportation to the electrode surface, and Step 3: Particle entanglement into metal matrix.

Fig.3 represents schematically these three steps as a parallel process to metal ion reduction and eventually hydrogen evolution, etc. (not shown here) at the electrode surface. A detailed microscopic picture of particle stability in an electrolyte, transportation to electrode surface, and its subsequent incorporation into a metal matrix, is illustrated in Fig. 4 considering different length scales for particle size and other associated phenomena. This schematic visualization will help one to understand and to get control over various factors responsible for particle incorporation as will be discussed further in this section.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

240

Subir Kumar Ghosh and Jean-Pierre Celis

Adsorption + reduction

Substrate

M+Particle

Particle

Step 3

Electrolyte

+

Mn+ Mn+ M n Mn + H2O H2 O M+n Mn+ H2 O Mn M n+ Mn+ M n+ M n M n+ H2O + n+ H2O M H

Diffusion +

Mn+ Mn+ M n Mn + H2O H2 O M+n Mn+ H2 O Mn M n+ Mn+ M n+ M n M n+ H2O + n+ H2O M H

Step 2

Convection

+

Mn+ Mn+ M n Mn + H2O H2 O M+n Mn+ H2 O Mn M n+ Mn+ M n+ M n M n+ H2O H2O + Mn+ H

Ionic cloud formation

Step 1

Figure 3. Schematic representation of particle co-deposition mechanism [20] consisting of three major steps: (1) Step 1 corresponds to particle stability in the electrolyte, (2) Step 2, particle transport from bulk of the electrolyte to electrode surface (convection and diffusion), and (3) Step 3, to particle adsorption and entrapment into metal matrix.

In fact, a lot of scientific challenges are involved in Step 1 and Step 3, whereas Step 2 relies on engineering efforts especially particle flow and hydrodynamics based on electrode geometry. However, it is pertinent to say that the final step, viz. particle incorporation in a metal matrix, strongly depends on previous two steps and a necessary attention to these steps is also required in order to obtain sound metal-particle composites with homogeneous particle distribution.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

3.1 Fundamentals of Particle-Ion Interactions For depositing composite coatings either by electrolytic or electroless process, the first and foremost criteria is that the particles should be kept suspended in the electrolyte. Upon putting into the electrolyte, depending upon the particle type, surface characteristics (amphoteric or negative or positive charge, hydrophobic, or hydrophilic), particle size, and shape, it undergoes physicochemical processes with ions and molecules present in the solvent in order to minimize its energy as seen in Fig.4. This physicochemical process involves (i) adsorption of ions and molecules on the particles, and (ii) agglomeration of particles or to remain suspended in the solution. These two later processes highly depended on particle ‗zeta-potential‘ (-potential). Again, the -potential of a particle in a given solution is also a function of particle characteristics. Naturally, particle deposition rate and particle entrapment into a metal matrix during electrodeposition are highly influenced by bath composition and plating conditions. A brief review of fundamental issues is given hereafter.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Scenario I when

< 1m

Agglomerates surrounded by adsorbed ions

Without surfactant

Reduction of metal ions Mn+

e-

Agglomerate deposition

H+

Embedment of particles into growing metal matrix

With surfactant

Dispersed particle surrounded by surfactant ions

H+ Mn+

Desorption of surfactant ions

Particle deposition

Metal matrix Substrate

Scenario II when

> 1m Reduction of metal ions

e-

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Mn+ Embedment of particles into growing metal matrix Particle with adsorbed ion cloud

H+

Particle deposition

Metal matrix

Substrate

Figure. 4 Schematic representation of particle stability, particle transport and embedment into metal matrix. Note the differences in particle agglomeration behaviour depending upon their size and in presence of surfactant in an electrolyte and their co-deposition behaviour. Smaller particles form aggregate quickly without surfactant, however, bigger particle takes some time. While embedding, note the differences between smaller and bigger particles: very thin metal layer deposition is sufficient to engulf the smaller particles but larger particles need thick enough metallic layers to cover it sufficiently.

242

Subir Kumar Ghosh and Jean-Pierre Celis

3.1.1 Electrolyte Modification Due to Ion Adsorption on Particle Most second phase particles are electrically charged. Therefore, when a solid particle is immersed in an electrolyte, there is a tendency for ions of a given sign to be preferentially adsorbed onto the solid and for the oppositely charged ions to remain in the neighboring electrolyte. This ‗first ion layer adjacent to particle‘ is called double layer or ‗Stern layer‘. Depending upon the ionic concentration and ionic strength of the electrolyte, the ionic layer can be compact or diffuse. The distance over which this occurs depends on the electrolyte concentration e.g. ~ 1 nm at concentrations of ~1 M and ~100‘s nm at concentrations of ~10-5 M. This later charge arrangement is called the diffuse electrical double layer around the particle. Adding salt to a colloid suspension causes the double layer to shrink around the particles; this is known as double-layer compression. It has a profound effect on electrolyte pH and on its -potential. The net charge, and hence the electrostatic potential on the particle surface, relative to the surrounding electrolyte, is strongly dependent on the balance between positive and negative ions – the potential-determining ions – in the solution. For example, most metal oxides have a surface layer of the metal hydroxide that is amphoteric and can become either positively or negatively charged, by taking up a proton or by proton abstraction, depending on the pH as given below:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

M-OH + H+  MOH2+ M-OH + OH-  M-O- + H2O

(8) (9)

Besides H+ and OH- in solution, metal ions are also present and take part in these adsorption processes. The particular pH at which the positive and negative charges are balanced, so that there is no net charge on the particle surface, is called the point of zero charge (pzc) or iso-electric point. In fact, it has been seen [85] that there is a competition between Ni+2 ion and H+ ion adsorption on -Al2O3 particles depending upon the pH of the electrolyte. In that study, Bhagwat, Celis, and Roos showed that the pH changes of the electrolyte on addition of Al2O3 particles substantially depend upon the initial pH of the electrolyte. The low pH of sulphamate electrolyte increases up to a high value and then stabilizes, whereas in high pH electrolytes, the pH decreases to a lower value before stabilizing at a critical value near pH 4 as shown in Fig.5. Apart from this, it highlights that the electrolyte pH change depends on aging time, Ni+2 ion concentration, alumina content, and temperature. It shows that a higher particle concentration and a longer aging period lead to a more extended change in pH. However, higher nickel ion concentration leads to a higher adsorption of H+ ions on a particle surface because of the higher ionic strength. Interesting observations were derived from plots of initial pH vs. final pH as shown in Fig. 6. The extent of pH changes, pH, allows us to divide that plot into three zones as follows:   

Zone 1 0< pH < 0.10 Zone 2 Sub-zone (i) 0.10 < pH < pH Sub-zone (ii) pH > pH > 0 Zone 3 pH < 0.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

243

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The quantity of 0.10 units was chosen arbitrarily since the lower point of inflection for these curves lies around this value. At these points where the curves cut through the diagonal of the figure (drawn at 450), pH is zero and that pH is called the critical pH (pHl) because above this point pH is positive. Similarly at the upper side, a critical point exists designated as pHu in Fig.6. In the second zone, the point at which pH is maximum, is noted as the middle critical point (pHm), at which tendencies of pH change reverse in the opposite direction. The different curve represents the variation of pH with different alumina content in the electrolyte. A close look onto these curves gives a qualitative explanation of the basic processes involved. In this sulphamate electrolyte, Ni+2 and H+ are the main cations and OHand sulphamate ions are the principle anions. Instantaneous changes in pH caused by the addition of alumina are due to either the removal of active protons and/or because the protons are made more active by the removal of the nickel ions. At low initial pH, the increase in pH upon alumina addition is caused by the adsorption of protons by alumina because of their high abundance. At higher initial pH, alumina adsorbs nickel ions, thus preventing them from participating in the equilibrium of the bulk electrolyte. This raises the activity coefficient of protons that is observed as a drop in pH. In the intermediate pH range both the adsorption processes balance each other, resulting in the pH change as appearing in Fig.6. This abundance of Ni+2 or H+ ions adsorbed on the particle surface dictates the adsorption of particles onto the electrode surface while the entrapment into the metal matrix will be further discussed in §4.2.2.

Figure 5. pH vs. time diagram for 600 g/l nickel sulphamate solution during mixing. [85]. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

244

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 6. Initial pH vs. final pH diagram for 600 g/l nickel sulphamate solution. [85].

3.1.2 Zeta-Potential (-Potential) In the previous section, it was shown that the adsorption of ions on second phase particles is a first process taking place in the electrolyte. As a consequence, a fundamental parameter called ‗zeta-potential‘ has to be taken into account. The -potential is known as the electrical potential that exists at the interface between a solid surface and its liquid medium, and is a function of the surface charge of a particle, any adsorbed layer at the interface, and the nature and composition of the surrounding medium in which the particle is suspended. The magnitude of the -potential of a particle is a measure of the particle interaction, which means that the -potential can be used to predict the long-term stability of a suspension. If the particles in a suspension have a large negative or positive -potential, then they will tend to repel each other, and resist forming aggregates. However, if the particles have a low potential value, i.e. close to zero, then there will be nothing to prevent the particles to approach each other and to form aggregates. The dividing line between stable and unstable suspensions is generally taken at either +30 mV or −30 mV [86]. Particles with -potential more positive than +30 mV or more negative than −30 mV are normally considered as stable ones. Most materials when immersed in water exhibit a -potential, and that -potential is affected by pH, concentration of an additive, or the ionic strength of the medium. For example in reactions described above (1) and (2), when the pH increases on addition of alkali, the surface will become more negative, or at least less positive as per equation (2), and the potential will track this. If acid is added, ionization will cause the loss of hydroxyl ions what will make the surface more positive. By this simple manner, therefore, the materials are more positive at a low pH, and more negative at a high pH. In describing the stability of colloidal particles in solution (Fig.7), the curve passes through zero called the iso-electric point (-potential = 0). This is very important for practical purposes since it is normally the pH where the colloidal system is least stable [87]. As a result, the solution pH has a marked influence on the zeta potential (), and hence on the incorporation rate as it is related to electrophoretic mobility (e) as

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

  eI /  r 0 where

245 (9)

 I is the liquid viscosity, and  r and  0 are the dielectric constant of the medium and

ionic strength

free space respectively.

unstable stable

stable

pH

Brownian agglomeration

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 7. Particle stability map with pH and ionic strength of the medium.

Critical size

Particle size Figure 8. Variation of Brownian agglomeration with particle size.

Similarly, the surface charge on particles when it is very low (three orders of magnitude less in comparison to Ni+2 ions) [88], is also a governing factor in deciding the incorporation rate of particles in line with Stokes electrophoresis model [89]. Assuming that the shape of the colloidal particles is spherical, the electrophoretic velocity, ve , in an electric field, E , can be expressed as:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

246

Subir Kumar Ghosh and Jean-Pierre Celis

ve  e E  where

q 6 r

E

(10)

 e is the electrophoretic mobility, q is the charge of the particle, r is the radius of

the particle, and  is the viscosity of the suspension. The Stokes model indicates that under identical electrophoretic field conditions, particles with high q move faster than particles with low q .

3.1.3 Particle Agglomeration and Stability Agglomeration is a mass-conserving, but number-reducing process that shifts the particle distribution towards larger sizes. This can have important consequences for the transport of particles (e.g. aerosol or colloid) since larger particles tend to settle more rapidly under gravity, and diffuse more slowly in the electrolyte. Agglomeration also reduces the particle surface area available for condensation and/or chemical reaction. It has a deleterious effect on the properties of the coatings. Generally, particle agglomeration in solution occurs along four different mechanisms, namely:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

(i) Brownian motion, (ii) Gravitational motion, (iii) Turbulent motion during hydrodynamic condition, and (iv) Electrostatic attraction. Brownian agglomeration is probably the best understood agglomeration mechanism as it occurs via particle Brownian motion. It is highly dependent on particle size and increases with decreasing particle size because of an enhanced translational motion compared to larger counterparts. The typical Brownian agglomeration probability variation with particle size is shown in Fig. 8. Note that below a critical size, the agglomeration probability increases exponentially and requires an imminent particle surface modification, especially for smaller particles to impede such processes. For example, Xiang et al. [90] measured the average particle size of nano-diamond (ND) of 8.4 nm before putting them into an electroless Ni-bath. After ultrasonic vibration in distilled water, the mean particle size of ND aggregates becomes 0.47 m. In a recent report, Chen et al. [91] reported that nano-size; inert SiC particles significantly agglomerate in a plating bath even at a low concentration of 0.002 vol%. The details of other agglomeration mechanisms can be found elsewhere [92]. Generally, the particle stability is described by DLVO theory [93-94] considering electrostatic attraction and Van-der Waals attractions forces. From various investigations, it is now well-known that with the decrease in particle size from micron to nanometer size, the tendency to agglomeration and sedimentation increases, and that the co-deposition efficiency decreases. It is therefore necessary to overcome these two barriers so that the desired property of composite coatings can be attained without disturbing much the metal ion deposition rate. In literature, the most adopted techniques are the modification of particles surfaces to increase repulsive forces between each other, either in-situ in the electrolyte or by ex-situ method. Among these, the most popular ones are:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

247

a) Addition of small quantity of supporting ions, b) Addition of surface active agents or surfactants, and c) Ex-situ surface modification. The details of these effects on particle content of composite are discussed in § 4.2.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

3.2 Mechanisms of Composite Deposition Mechanisms of co-deposition of second phase particles along with metal ions have been proposed based on various investigations done on micron size particles since early 1960s. These theories basically put forward the following possible mechanisms in regard to particle transport towards electrode followed by their inclusion into a metal matrix: (i) electrophoresis; (ii) mechanical entrapment; (iii) adsorption of particles into cathode; (iv) convection-diffusion, etc. A summary of these models developed over the years by various authors is presented in Chart-2 below providing information on basic assumptions, limitations, and the range of applicability. Besides these, recently, Stappers et al. [104] tried to model particle co-deposition under turbulent conditions. It is evident from this chart that most of the models developed so far are applicable to micron size particles except a few whose applicability range is sub-micron/ nanometer size particles viz. Beulens et al. (50 nm Al2O3) [20], Fransaer et al. (10 nm SiC) [99], Vereecken et al. (300 nm Al2O3) [102]. It has also to be noted that the relationships between the quantity of particles in the metal deposit and current density or hydrodynamics has been the main focus in these models. Even though these theoretical models are quite successful in describing one or two specific metal-particle systems within a limited set of operating parameters at a laboratory scale, the flexibility and reliability of each to describe the behavior of a wide range of metal-particle systems over a vast operating range, still requires validation. Quite often, the existing models involve complicated mathematical relations that are interrelated to an extensive set of operating parameters. It is, however, quite difficult to realize these parameters in practice and, in all the models available so far, the effect of electro-crystallization has not been considered. This aspect is important since the inclusion of particles into a metal deposit may cause (1) reduction of cathode surface area if the particles are non-conductive, (2) enlargement of the cathode area if the particles are conductive, or (3) modification of the deposit morphology, and (4) even affect the particle adsorption phenomenon in the case of partially-engulfed particles. Besides electro-crystallization, the future models should take into account other interactive variables such as particle characteristics (type, dimensions, and concentrations), the operating parameters (temperature, current density, pH, and hydrodynamics) and the electrolyte compositions (concentrations, presence of surfactants, and additives). The latter issue becomes even more important while considering co-deposition systems involving nanometer size particles. Reasons are as will be seen later in this chapter, that the electrolyte pH and surfactants are crucial factors in achieving or not the electrolytic co-deposition of nanometer size particles into a metal matrix.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

248

Subir Kumar Ghosh and Jean-Pierre Celis

Chart-2. Features of chronological developed electro-co-deposition model Model

Basic features of mechanism or assumptions

Systems, particle size range

Guglielmi, 1972 [95]

Considers two step deposition of particles: (a) Loose adsorption; (b) strong adsorption at the electrode surface. Particle characteristics and electrolyte conditions are accounted semiempirically. The effect of flow and the effect of particle characteristics were not considered.

Ni-TiO2, Ni-SiC 1-2 m

Celis et al. (MTM, K U Leuven) 1977 [96]

Buelens et al. (MTM, K U Leuven) 1987 [20]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Valdes et al. 1988 [97]

Fransaer et al. (MTM, K U Leuven) 1992 [98-99]

Hwang and Hwang 1993 [100]

Wang et al. 1999 [101]

Vereecken et al. 2000 [102] Barcot et al. 2002 [103]

Considers two step deposition of particles but different from Guglielmi as (a) adsorption of depositable ions on particles surface is assumed first; (b) Upon transport of the same particle-ion assembly to the electrode surface, a strong adsorption takes place, and once ions are reduced, a permanent contact between particles and cathode is realized. This model considers a five-step process as described in Fig. 3. Uses statistical approach to describe the particle incorporation at a given current density. (a) Important idea of this model is that the particles will be embedded only if a certain fraction of adsorbed ions on the particle‘s surface is reduced. (b) Mass transport of particles is proportional with the mass transport of ions towards the working electrode. (c)Wt% of particle in deposit is shown to increase under charge-transfer control, and to decrease under mass-transfer control regime. Hydrodynamics is not considered and certain parameters and coefficients are very difficult to be quantified. In this model, first time both mass-transfer and interfacial kinetic processes are clubbed together to describe the whole electrochemical system. Certain experimental results like a peak in the particle content vs. current density curve are predicted but at limiting current density rather than at low current density. Considers particle trajectory to describe mass-transport from bulk of the electrolyte to the electrode surface for composite deposition at a RDE. (a) Particle flux to the electrode surface was calculated using trajectory expression considering several forces acting on particle; (b) at electrode surface several other forces like adhesion, shear force, etc. were assumed for particle capture. However, extensive mathematical calculations and the complexity of an adaption to other geometries than RDE have prevented further quantitative investigations. This is an improved version of Guglielmi and Celis et al. models. It considers three steps (i) forced convection of particles to the cathode surface; (ii) loose adsorption of particle on the surface; (iii) irreversible incorporation of the particles by reduction of fraction of adsorbed ions on the particle. In order to explain the current density dependence, three regimes of current density; low, intermediate and high are used. There is a lack of clarity in describing the reduction of H+ ion and bond formation with metal matrix. This model is a modified version of Fransaer et al. model. The average adsorption strength of particles on the electrode is employed to describe the intensity of the interaction between particles and electrode instead of the force balance on a particle at the electrode with a distribution of adhesion forces in the model of Fransaer. The model does not fully describe the current density dependence of particle content of composite. The particle transport to the electrode surface is considered via convectivediffusion process and treats particle incorporation in terms of residence time at the surface. It is valid for particles with dimensions less than the diffusion layer thickness. An improvement of Guglielmi model, which introduces a 3rd order correction factor to account for the effects of adsorption and hydrodynamics.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Cu-- and Al2O3, -Al2O3, 50 nm -Al2O3 0.3 m

Cu--Al2O3, Au-Al2O3 50 nm

No exptl. validation is available Cu-PS (11m) Ni-SiC (0.01-10 m), Z-PS (0.8 m)

Co-SiC 3 m

Fe-P-Al2O3 7 m

Ni-Al2O3 0.3 m Ni-PTFE 0.5 m

Plating of Nano-composites- Overview and Trends

249

4. EFFECT OF VARIOUS PARAMETERS ON PARTICLE INCORPORATION Over the years numerous investigations on particle incorporation into metal matrix revealed that several factors are affecting the particle incorporation rate. Chart-3 presents several parameters that affect directly or indirectly the above-mentioned three steps controlling the incorporation rate and thus also the properties of electrolytic composite coatings.

Chart 3. Plating process parameters affecting particle incorporation Step-1

Step-2

Particle Stability in the Electrolyte

Particle Transport to the Electrode

Particle Characteristics Particle composition, Particle size, Particle shape, Particle bath concentration

Bath Composition & Conditions Bath composition, pH, Supporting ions, Surfactants, Ex-situ surface modification, Temperature

Step-3 Particle entrapment in metal matrix

Hydrodynamic Conditions

Electrode Geometry

Process Type

Laminar, Mixed or Turbulent regime

Rotating disc electrode (RDE), Rotating cylinder electrode (RCE), Plate-in-tanks, Parallel plate electrodes and many variations.

Electrolytic: Direct current, Pulse current, Reverse Pulse Current, Potentiostatic, Two-step deposition Electroless

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.1 Particle Characteristics Particle characteristics as mentioned in Chart-3 are very important not only because they affect the particle stability in the electrolyte but also because they affect the incorporation rate and thus the end use significantly. The end use of composite coatings basically dictates the selection of particles to be incorporated. Apart from this, the compatibility between particles and the matrix is a factor influencing the internal stress level and the adhesion on the substrate. Therefore, depending upon the mechanical, chemical, catalytic, and self-lubricating properties of the particles, composite coatings can exhibit a dispersion strengthening, a corrosion resistance, a wear resistance, a catalytic surface, or a self-lubricating property. Of course, particle-matrix combinations and microstructures dictate these characteristics. Once particle selection is done, other factors viz. particle size, shape, and surface conditions as well as the volume percent of particles in the matrix come into the picture, and need to be optimized. These later parameters, particle size, concentration in electrolytes strongly affect particle stability and their incorporation rate. It is worthwhile to mention that the availability of a database on particle characteristics, compatibility with a given matrix, and co-deposition levels, would enable experimentalists [35, 38, 105] to do the appropriate selection of process parameters in a predictive way. That would enhance definitely the development and industrial transfer of composite plating.

4.1.1 Compositions of Particle Dispersion strengthened coatings can be achieved by the incorporation of insoluble particles of nanometer size e.g. Al2O3, and TiO2 in metallic layers. Such composite layers generally exhibit a large increase in hardness, tensile strength, and electrical resistivity. They

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

250

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

need often to be subjected to a thermo-mechanical treatment in order to achieve the desired ductility or maximum strength. Some examples in this field are Ni-Al2O3 [106-107], and NiTiO2 [39]. In order to produce wear resistant coatings, SiC, diamond, WC, or other hard ceramic particles can be chosen as second phase material. In these cases, the electrodeposited metal acts as a ductile matrix to which hard particles provide the abrasion resistance. Typical examples are NiCo-SiC [108-109] and Co-Cr2O3 [110-111]. Composite Co-Cr2O3 containing 30 vol.% Cr2O3 embedded shows an excellent wear resistance at high temperature [112]. To obtain self-lubricating layers, inert particles with excellent anti-friction properties like e.g. MoS2, graphite, PTFE, polymers and also liquid containing polymer capsules can be incorporated in a metallic layer. Examples are electrolytic and electroless Ni-PTFE [113115], Ni-Co-MoS2 [116] Ni-P-MoS2 [117], and Ni with embedded polymeric microcapsules [118, 119]. In the field of corrosion-resistant coatings, Co-Cr3C2 [120], Ni-P-SiC [121], and Ni-PZrO2 [122] can be cited. Another well-known application is the production of micro-porous chromium layers deposited on top of a nickel deposit containing electrically non-conducting particles. Composite Zn-Al2O3, Zn-SiO2, and Zn-TiO2 [123-124] coatings also show improved corrosion resistance in comparison to pure Zn coatings. In the case of catalytic composite coatings, Ni-TiO2 [125-126] nano-composites, and NiCeO2 [127-128] are recently developed coatings, and find applications in car exhaust systems to reduce the emission to the environment, and as oxygen sensors for various applications respectively. TiO2 has a very good catalytic surface and is available as nano-size particles with a very high surface area. Similarly, cerium has an oxidation-reduction couple, i.e. Ce+4/Ce+3, enabling it to be used as oxygen sensors. These examples clearly show that the choice of particle composition is dictated by the desired properties. It is thus very important to keep in mind the effects of particle properties on the incorporation rate during composite plating.

4.1.2 Particle Size from m Down to nm Size Particle size is an important parameter that controls physicochemical properties like surface charge, zeta potential, tendency to agglomerate in a given solution, specific adsorption of ions, particle mobility, and interaction with any applied electrical field. Over the years, world-wide research on electrolytic composite plating has shown that a decrease in particle size reduces the particle content in composite coatings deposited under the same set of deposition conditions. Particle content in a metal matrix composite is currently expressed either as volume%, weight%, or atom%. Therefore, one has to be careful when comparing literature data in data banks. As mentioned in Section 2, the ‗number density‘ would be a more appropriate parameter to correlate structural and property data. Sometimes, however, it is difficult to determine experimentally the particle ‗number density‘ in a matrix. Of course, for a given particle size, under the assumption that a homogeneous distribution is achieved, the volume or weight percentage of particles is proportional to the ‗number density‘. However, if agglomeration takes place in the plating solution, the effective ‗number density‘ is not directly correlated to the volume or weight percentage. Fig. 9 shows the variation of particle content of Ni-SiC coatings with particle size at a given particle bath concentration of 20 g/l SiC. Note that for small particle sizes, the particle

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

251

content in the coatings increases till 2.5 m and then decreases at higher particle size. A similar observation was done during the co-deposition of micron size MoSi2 particles in a Ni matrix by electroless plating [129].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 9. Particle content vs. particle size at a given particle concentration of 20 g/l of SiC in electrolyte.

Figure 10. Variation of particle number density with particle size at different particle content in composites.

This particular phenomenon is confirmed by previous reports on a negligible [130] or even opposite [131-132] correlation between particle size and particle content. For very large High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

252

Subir Kumar Ghosh and Jean-Pierre Celis

particles, the particle content decreases with increasing concentration of particles in the electrolyte. This could be due to the unavailability of a sufficient surface fraction for incoming particle flux to be attached at the electrode due to the higher volume of particles and the time-lag to cover up large particles by reduced metal ions, sometimes called ‗parking problem‘ [133-135]. Therefore, even though the flow of second phase particles is continuous during electro-deposition, however, the particle-particle collisions and the parking problem or preferential co-deposition, hinder particles to find a favorable site to cling to the cathode surface. Fig. 10 represents the variation of the number density of particle in Ni-SiC coatings with particle size for a given volume fraction of particle content as presented in Fig.9. It is calculated by assuming that SiC particles are spherical in shape and homogeneously distributed in the Ni-matrix. At small particle sizes, it is evident that the number density is several orders of magnitude higher than its bigger counterparts for a given particle volume content. And it decreases exponentially as the particle size increases from sub-micron to micron size range. Beyond 1m, the number density decreases in a monotonous way with increasing particle size. And for a given particle size, the number density increase is linear with the volume content of particles in the Ni-matrix. For example, Garcia et al. [136] demonstrated that the co-deposition of SiC particles of different sizes (0.3, 0.7, 5 m) increases with particle size and concentration of particles in the bath, however, the calculated ‗number density‘ deceases with increasing particle size. Berkh et al. [137] did a similar observation for micron size SiC particles deposited with Ni. Recently, Pavlatou et al. [138] made a similar observation while electrodepositing Ni-SiC composite coatings containing SiC particles of 1 m and 20 nm, by pulse and DC plating. For sub-micron and micron sized SiC, Maurin and Lavanant [139] demonstrated that the amount of co-deposited SiC increases when particle size increases. They noticed also that the codeposition of SiC particles smaller than 100 nm was more difficult than that of 0.8 m sized ones. Kim and Yoo [140] reported that a maximum co-deposition in a sulphamate nickel bath is achieved for SiC particles of about 10 m. An increase of the SiC particle size up to 20 m resulted in a decrease of the amount of co-deposited SiC particles. The size of particles in the nanometer range also affects the rate of incorporation of particles into the metal matrix. Findings have suggested that as the size of nano-particles becomes smaller, a larger number can be incorporated into a metal deposit per unit volume similar to what was found for micron size particles. For example, the rate of incorporation of two different sizes of Al2O3 nano-particles (50 nm and 300 nm) in a nickel coating, has been studied [141-142]. Using similar electrodeposition parameters (1000 rpm, 20 mA cm−2,and 73 C cm−2), it was found that the % volume fraction of 300 nm Al2O3 in nickel was much higher compared to the one achieved with 50 nm Al2O3 particles. However, the quantity of 50 nm Al2O3 particles in the nickel was much larger than the quantity of 300 nm Al2O3 particles agreeing the number density change with particle size. Some exceptions are also reported in literature, e.g. in the case of the electrodeposition of Ni-TiO2 coatings where the weight percentage of 12 nm anatase in the nickel deposits was higher than the one achieved with 1 μm rutile particles [143].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

253

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.1.3 Particle Shape Even though not much literature is available related to the effect of particle shape on particle co-deposition, the basic idea is that the surface area/charge is basically responsible for the colloidal behavior and hence the co-deposition process because the surface charge of a particle is related to the surface area and, hence, particle shape. For example, the surface area of a rod-shaped particle is larger compared to the one of a spherical particle with an identical diameter. Thus the general perception is that particles with large specific surface area can attract a higher number of adsorbed ions, and that thereby the ion-particle assembly can feel much more the electrical force attracting them towards the cathode. Apart from this, adsorbed ions help in stabilizing the particle suspension in the electrolyte. Moreover, on adsorption of rod-shaped particles on an electrode surface the interaction between them is much higher than in the case of sphere-shaped particles, leading to a higher entrapment into the metal matrix as shown in Fig.11. However, at decreasing particle dimension, the tendency to agglomeration increases due to higher thermal motion [144]. Previously, Hovestad and Jansen [31] cited an example related to particle shape effect. The amount of co-deposition of Al2O3 particles with electroless Ni increases in the order fiber, irregular to spherical-shaped particles. This behavior can be explained by considering the specific surface area of such particles. Related to the electrolytic co-deposition of nano-rods (diameter 20-30 nm, and length 2-3 nm) and nano-spherical shape SiC particles (5 nm) in a Ni matrix, Abdel Aal et al. [145] showed in 2008 that the particle content for nano-rods is higher in comparison to the one of nano-spherical shape SiC particles under identical bath loading conditions (Fig.12). Lower incorporation of spherical particles in composite plating can be ascribed to the smaller size of the particles, which are swept away from the electrode surface compare to the rods with larger size particles [146]. Other details behind this observation are discussed in the next section.

Rod shape particle

Spherical shape particle

Electrodes Figure 11. A schematic representation showing changes in surface contact between rod-shaped or spherical-shaped particles and electrodes.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

254

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 12. The influence of SiC bath content on SiC content in Ni–W–P–SiC coatings deposited at 55 0 C, pH 6.5, and a current density of 4 A.dm-2. [145].

4.1.4 Concentration of Particles in Bath For micron size particles, generally the volume fraction of particles increases with the increase in particle concentration in the plating bath, and ultimately saturates at a certain level. This was confirmed for a wide variety of metal particle systems [20, 31, 41, 98, 99, 133, 134, 137, 147-148]. This behavior is very similar to particle adsorption on an electrode surface following the Langmuir adsorption isotherm. Fig. 13 shows the relationship found during graphite incorporation in Cu-Sn coatings at different rotation speeds [149]. From this figure, two general observations can be made: (1) the volume fraction of particles in the metal deposit increases substantially when the particles concentration in the plating solution increases, and (2) a saturation in co-deposition is reached at high particle concentration in the plating solution.

Figure 13. Langmuir-like isotherm showing the relationship between the vol.% of particles in the metal deposit and particle concentration in the electrolyte for Cu–Sn alloys containing 10 μm graphite, and the effect of electrode rotation rate on it. Speed of magnetic stirrer: ▼ 50 rpm, ○ 100 rpm and ● 200 rpm. [149].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

255

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 14. Number density of particles in Ni–SiC coatings containing particles of three different sizes plotted vs. the number density of particles suspended in the plating solutions. [136].

In the case of the electrolytic co-deposition of SiC particles with different sizes (0.3, 0.7, and 5 m), Ni, Garcia et al. [136] have shown that the volume percent of co-deposited SiC in the coatings increases with increasing concentration of particles in the plating solution what is in agreement with the observations made by several researchers [98, 145]. The highest volume percent of co-deposited SiC was obtained with 5 m particles. The re-plotting of the co-deposition data as a number density of particles in the plating solution, ns, and in the coating, nc, results into Fig. 14. The number density of co-deposited SiC particles is thus 1 to 100-fold larger than the number density of particles in the plating solution. The co-deposition efficiency, defined as the ratio between the number density of particles in the coating and in the plating solution, increases for each SiC particle size with decreasing number density of particles in the plating solution. More important is the fact that the co-deposition efficiency increases substantially with decreasing particle size. This appears in Fig.14 as a jump in the number density of co-deposited particles at a given number density of particles in the plating solution (see e.g. at 5×1014 and 5×1017/m3), but of different particle size. Therefore, in contrast to the apparent outcome of Fig. 13, the co-deposition efficiency does not decrease with decreasing particle size. On the contrary, at a given number density of particles in the plating solution, the co-deposition efficiency increases with decreasing particle size. Recent investigations on the effect of the concentration of nano-particles in the electrolyte on their content of the composite coatings have also been predicted to follow a Langmuir isotherm adsorption phenomenon [121, 142, 150-155]. The amount of nanoparticles in the solution has a significant effect on the volume percentage of the nano-particles incorporated in electrolytic deposits.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

256

Subir Kumar Ghosh and Jean-Pierre Celis

Recently Lee et al. [155] worked out a comparison between the effect of particle concentration in the electrolyte on the particle content of the composite, and the particle codeposition rate by considering two different SiC particle sizes, namely ~2.5 m and ~50 nm. For nanometer size particles, the saturation of the particle content in deposits appears much earlier than in the case of micron size particles, in line with observation done for their codeposition of Al2O3 into a Ni matrix [142, 144].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.2 Effect of Electrolyte Compositions and Electrolysis Parameters 4.2.1 Electrolyte Composition for m and nm Size Particles Various investigations have clearly demonstrated that particle incorporation depends on bath constituents and particle concentration in the plating solution. Particle surface charge and the associated adsorption of ionic species on particles influence the diffusion and the incorporation of particles into an electrodeposited metal matrix. Essentially, ‗particle potential‘ and the interaction with other bath constituents and electrode, govern to a large extent the deposition rate of particles. A higher metal ion concentration increases the particle -potential, and consequently enhances the particle-electrode attraction and the incorporation of particles [156]. Besides metal ions, several other constituents like surfactants, brighteners, or wetting agents are commonly added to the plating solutions in order to improve the structural and functional properties of electrodeposits. Surfactants have a strong influence on the incorporation of particles as will be seen in § 4.2.4. Some reports demonstrate a positive influence of brighteners on the incorporation of particles [157]. Other reports claim a positive [158] or negative [159-160] influence of wetting agents on the incorporation of particle. For same kinds of particles, the electro-deposition from different electrolytes ends up in different particle contents. For example, Al2O3 can co-deposit from Ni and Cu-baths but not from Cr-baths [160]. Cu-Al2O3 can be electrodeposited from a sulphate electrolyte of low pH, but not from a cyanide bath [160]. The addition of iron group metal ions seems to enhance the incorporation of particles like SiO2 into a Zn [161] matrix. Kariapper and Foster [162] noticed that an increased metal ion concentration leads to an increased metal ion adsorption on particles, and eventually increases the incorporation of particles. Recently, the dispersion of Al2O3 (~25 nm  and  phase) and SiO2 (7 nm) nano-particles was monitored in a nickel sulphamate electrolyte via -potential measurements and photon correlation spectroscopy (PCS) [163]. It was found that the agglomeration tendency of Al2O3 particles increases with increasing ion concentration, but SiO2 particles behave differently behavior and show only a limited agglomeration. In addition, an adsorption of Ni+2 ions on SiO2 particles was noticed. So, the incorporation rate and the particle content in composite electro-coatings differ from type to type of particles. During the electrolytic co-deposition of nano-size Al2O3 (80 nm) from a sulphamate bath by using direct current in combination with ultrasound agitation, a strong agglomeration was noticed. An agglomerate average size of 280 nm was noticed at 1.27 M Ni+2 ion [164]. That size reduced to 178 nm on dispersion in a solution of 0.2 M Ni+2 ions. At the same time, the volume content of alumina particles in these composite coatings increased from 8.37 vol.% up to a maximum of 26.78 vol.%.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

257

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.2.2 Electrolyte pH in the Case of m and nm Size Particles The -potential of particles was already shown to be highly dependent on pH. That potential indirectly dictates particle stability and surface charge, and thus by consequence also particle content in electrodeposited composite coatings. Several papers [2, 31, 85, 130, 133, 137, 148, 158, 165-173] report on the effect of pH on the co-deposition of particles into different metal matrices. Depending upon particle type, composition of the electrolyte, and deposition conditions, experimental results are not comparable. On the contrary, systematic results correlating pH, zeta-potential, and particle content, can be found. Previously, Aslanidis et al. [171], while depositing Zn-SiO2 (0.5 m) composite coatings found out that the pH of the electrolyte as well as the pH change at the cathode modify the codeposition efficiency drastically. They co-deposited either pure silica or titania modified SiO2 particles. The presence of titania on SiO2 particles enhanced the dielectric constant of the particles, and hence increased the adsorption of Zn+2 ions compared to pure SiO2 particles. At increasing pH the Zn+2 ion adsorption increases on both types of particles. However, the extent of that increase in Zn+2 ion adsorption was much higher in the case of titania modified SiO2 particles. The specific adsorption of ions that can be reduced at the applied electrode potentials also depends in the case of nano-size particles on the relative concentration of H+ ions as observed by Bhagwat et al. [85]. The effect of pH on the co-deposition of nano-size Al2O3 in the NiAl2O3 system as reported by previous researchers, is summarized in Fig.15. This simple observation was explained nicely by Bhagwat et al. [85] by considering ion adsorption on Al2O3 particles (average size 50 nm). However, at low pH, the adsorption of particles onto the cathode is hindered by the large coverage by H+ ions, as happens in ‗Zone-I‘ of Fig.6. That process is at the basis of the ‗riding problem‘ first noticed by White and Foster [174].

Figure 15. Effect of pH on co-deposition in the Ni-Al2O3 system. [85]. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

258

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 16. Variation of the zeta-potential of micro- and nano-sized SiC particles with pH. Type-A are -SiC with an average size of 2-3 m. Type-B are -SiC with an average size of 45-55 nm. [155].

At an intermediate pH (see Zone 2 in Fig.6), the simultaneous adsorption of Ni+2 ions along with H+ ions on the particles, allows a particle adsorption at the cathode, and finally an entrapment of particles into the metallic layer during electrolysis. The co-deposition reaches a maximum at ~pH =4. Above that pH, the particle content is found to decrease what could be due to the higher amount of particles at the electrode surface that results in the so-called ‗parking problem‘ (see Zone 3 in Fig.6). Very recently, Lee et al. [155] investigated the effect of pH on the zeta potential of particles and on the particle content in electrolytic composite coatings. They compared results obtained for nanometer (45-55 nm) and micron size (2-3 m) SiC particles. The iso-electric point for micron size -SiC particles was found to be at pH 3.0 while for nanometer size SiC particles it was pH 7.0. The typical zeta-potential vs. pH curves (Fig.16) show that the zeta potential for micron size particles is more negative than the one for nanometer size particles. That reveals a higher adsorption of positive ions on the particles, and therefore one can expect that in the former case, a higher deposition rate and a higher particle content in the matrix will be achieved. That was experimentally confirmed (see Fig.17). In fact, the volume percent of co-deposited micron-size SiC particles is found to increase with increasing pH but to saturate after a certain pH-value. The co-deposition rate increases largely with increasing pH in acidic solutions. Once it reaches a maximum, co-deposition decreases at further increasing pH probably due to a particle-particle steric hindrance and a parking problem at the cathode surface. That effect was found to be quite similar in the case of nanometer size SiC particles but at a reduced particle content and reduced deposition rate as expected.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

259

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 17. Deposition rate and co-deposition of SiC with pH of the plating bath; (a) for micron size SiC for type-A and (b) is for type-B [155]. Type-A are -SiC with an average size of 2-3 m. Type-B are SiC with an average size of 45-55 nm. [155].

4.2.3 Addition of Small Quantity of Supporting Ions It is known that surface charge on second phase particles might be sometimes quite small. In consequence, the particles either form agglomerate or feel a very low electrophoretic force moving them from the bulk of the electrolyte towards the electrode surface. One simple way to overcome this problem is by increasing the adsorption of oppositely charged ions onto particles. Various investigators [159, 162, 175-177,] found that the presence of suitable supporting ions, either cations or anions) in the electrolytes increases the thickness of the double layer surrounding particles. For example, small amounts of cations, like Tl+, Ce+, Rb+, and NH4+,or amines like EDTA (ethylene diamine tetra acetic acid), TEPA (tetra ethylene pentamine), alanine, etc., [32] promote co-deposition. Not only the addition of mono-valent ions, but also bi-valent ions like Mg+2 [178], Co+2 ions [179-181] were found to enhance codeposition. Recently, it was noticed that the addition of Co+2 ions enhances the co-deposition of SiC and Al2O3 particles in electrolytic nickel [179-180]. Bivalent Co+2 ions were also found to enhance the co-deposition efficiency in the case of nano-diamond particles but also the dispersion of the particles in the composite nickel coatings [181]. Apart from cationic additives, anions play also a role. For example, the co-deposition of BaSO4 in copper increases from 0.5 to 4.5 wt% in addition of 25 g.l-1 of EDTA [175]. That co-deposition is even further enhanced by the addition of ammonia salts: NH4Cl enhances the co-deposition of BaSO4 in copper matrix to a larger extent than NH4F. Similarly, in a recent study, Bund et al. [170] investigated the effect of different anions on the co-deposition of Al2O3 particles (15 nm) in electrolytic copper. They found that negatively charged particles (due to adsorbed ions) do co-deposit better than positively charged particles from acidic copper sulphate baths in which particles are positively charged, from neutral pyrophosphate baths in which particles are negatively charged, and from alkaline sorbitol based baths in which particles are less negatively charged. The highest amount of co-deposition, namely ~11 wt% Al2O3, was obtained in pyrophosphate baths. Podlaha et al. [182] also indicated that the citrate ions have a better influence than chloride on the co-deposition of Al2O3 particles (32 nm) with electrolytic nickel.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

260

Subir Kumar Ghosh and Jean-Pierre Celis

4.2.4 Addition of Surface Active Agents or Surfactants In the case of micron size particles, even though agglomeration is not that severe, surfactants were found to promote particle co-deposition up to a certain extent. For certain hydrophobic particles like PTFE [183-185], graphite [1], and B4C [186], surfactants were used to avoid agglomeration during composite plating. In the case of the co-deposition of CuAl2O3 from an acid copper sulphate plating bath, Terzieva et al. [187] verified that hydrophobic silica co-deposited in presence of cetyl trimethyl ammonium hydrogen sulphate (CTHAS) but could not be co-deposited in presence of SDS. They suggested that surfactants also modify the electrode surface, and so influence particle incorporation. Recently, Shrestha et al. [109, 186, 188-189] showed that the addition of a cationic surfactant containing azobenzene (AZTAB) resulted in a co-deposition of 71.5 vol% SiC particles (1 m) in nickel without agglomeration. AZTAB was found to act as an efficient dispersing agent in the case of Ni-B4C [186] and Ni-diamond [189] electrolytic deposition. PTFE, being hydrophobic in nature, needs to be hydrophilized in order to disperse them in electrolytic or electroless nickel plating solutions [190-192]. Generally, a surfactant containing fluorine (fluorochemical surfactant) is used to render PTFE hydrophilic because the perfluoro group in it exhibits a high adsorbability on PTFE. Previous reports showed that the addition of cationic surfactants is more useful since composite coatings containing up to 70 vol% PTFE particles [184, 193-194] were obtained. The beneficial effect of a cationic surfactant is believed to result from the achievement of a positive charge on particles as was recently confirmed by Hu et al. [184]. Kinoshita et al. [185] used a non-ionic hydrocarbon surfactant for the synthesis of electroless Ni-P-PTFE composite coatings. Besides these, polyoxyethylene alkyl ethers (non-ionic hydrocarbon) [195], cationic (tetrabutyl-ammonium chloride (TBAC), C4H9)4NCl), anionic (perfluroalkyl sulphonic acid or formulation Fn-RSO2OH), and non-ionic (perfluroalkyl betaine or formulation Fn-R-COOCH2N(CH3)3) [196] were also used as suitable surfactants for depositing PTFE with electroless Ni. An anionic surfactant like sodium dodecyl sulphate (SDS) also enhances the electrolytic co-deposition of polymeric microcapsules in nickel [197], as well as polystyrene particles in zinc [198]. Cetyl trimethyl ammonium bromide (CTAB) as cationic surfactant enhances the codeposition of nanometer size particles like SiC (250 nm) in electrolytic nickel [199], MoS2 and WS2 nano-tubes in electrolytic nickel [200-202], and Ni-CNT (carbon nano-tubes) [203] from nickel Watt‘s bath. Surfactants are also required in the case of electrolytic Ni-nanodiamond (particle size 4-8 nm) deposition [204]. Zou et al. [205] used polyethylene glycol (PEG) as dispersant agent for depositing MoS2 nano-tubes in electroless nickel. Wang [206] investigated the effect of various anionic, cationic, non-ionic, and amphoteric surfactants in Ni-MoS2 composite plating. They found that the cationic surfactant (benzyl ammonium salts; BAS) provides a maximum adhesion and particle deposition efficiency. Recently, Hazan et al. [207] showed that comb-polyelectrolyte surfactants are effective in dispersing nano-Al2O3 (14 nm  &) particles. Highly homogeneous Ni-P/Al2O3 nano-composite coatings plated from stabilized baths contained up to 50% co-deposited particles. 4.2.5 Ex-Situ Surface Modification As seen before, the addition of surfactants or the presence of supporting ions assists in particle incorporation by an in-situ changing of the particle surface chemistry. Several investigators also attempted to alter the surface of particles via an ex-situ surface

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

261

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

modification. Celis et al. [187] used hydrophobic and hydrophilic SiO2 particles to be codeposited during copper electrodeposition. The hydrophilic SiO2 was modified into a hydrophobic one by treatment with a functional short-chain siloxane (Oligodimethyl siloxane,-diol, OHDMS). The effect of two different surfactants viz. CTAHS and SDS on the codeposition behavior, was also investigated. They found that hydrophilic silica particles did not codeposit neither from surfactant-free, nor from surfactant-containing acid copper sulphate solutions, but up to 14 vol% hydrophobic silica co-deposited from solutions containing 15 g/l SiO2 particles and 10-4 M CTAHS. Previously, the same group investigated the ex-situ surface modification via applying a TiO2-layer on SiO2 particles during composite plating with Zn [171]. A better inclusion efficiency and entrapment within the metal matrix were noticed. Prior to composite plating of nickel from a sulphamate bath, Henuset and Menini [196] did an acid treatment on SiC and SiO2 particles in order to remove the possible oxide films or impurities. They also noticed drastic influences on the kind of pre-treatment like HF or HNO3 and subsequent use of surfactants on the particle incorporation during nickel electrodeposition.

4.2.6 Effect of Temperature Very similar as in metal electrodeposition, the temperature of the electrolyte sometimes profoundly affects the particle incorporation in composite plating. However, the influence is particle-electrolyte system specific and differs over a wide scale range. There is a two-fold effect of temperature on the electrolytic deposition of inert particles. At low temperature, the kinetic activity of particles increases with increasing plating bath temperature what is beneficial for the co-deposition of second phase particles [208]. According to Langmuir adsorption theory, an increase in temperature leads to a decrease in the adsorbability of particles onto electrode surfaces, a decrease of the cathodic over-potential, that makes it harder for the particles to be embedded in the coating. Besides these, temperature influences indirectly particle incorporation by affecting the surface chemistry of particles, i.e. adsorption/desorption of ions on their surface, by changing the pH of the electrolyte, the viscosity and density of the electrolyte, and hence the particle mass-transfer [99], and of course by changing the kinetics of the metal ion reduction process. Therefore, one can expect a marked influence of temperature on particle incorporation from electrolytes containing surfactants. For example, temperature was found to increase continuously the incorporation of TiC particles into nickel [209]. A maximum incorporation at 50 0C was observed for Ni-PTFE [148], Ni-V2O3 [131], and Ni-BN [210] co-deposition. In the case of Ni-Al2O3 composite plating [2,107] no effect of temperature on the percentage of embedded particles was found due to the fact that this co-deposition took place from a surfactant-free electrolyte. Tian et al. [211] reported that temperature affected the co-deposition of Al2O3 particles in a Ni-Co matrix. They found that on increasing temperature from 25 0C to 35 0C, the particle content increases at first and then decreases on further increasing temperature. Very recently, Wang [206] showed that a maximum deposition of MoS2 in nickel is achieved at 45 0C and that the amount of co-deposition decreases on a further increase in temperature possibly because the amount of absorbed surfactant, namely benzyl ammonium salts (BAS), decreases.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

262

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.3 Hydrodynamic Conditions Besides the use of surface active agents (chemical dispersion), physical dispersion, mainly bath agitation, is another way of dispersing particles, keeping them suspended in the plating solution, and preventing them from either floating or settling down the container. Common agitation techniques are: magnetic stirring, mechanical blade stirring, re-circulation of the electrolyte, air bubbling, ultrasonic agitation, and electrode mechanical rotation. Besides these techniques other recent agitation techniques viz. mechanical chemistry method (HEMC) [212-214], and impinging jet electrodes (IJE) [215-216] are also an effective means of dispersing and incorporating particles into metal matrix during electro-deposition. The effect of agitation on particle incorporation is related to the transport of particles from the bulk of the solution towards the electrode. At increasing agitation, the particle content in the coatings increases, reaches to a maximum, and then decreases on a further increase in agitation. Under vigorous stirring, ‗nascent adsorbed particles‘ are removed from the cathode surface by hydrodynamic forces before entrapment of particles into the metal matrix became effective. Details can be found in related review articles [32, 33]. Recently electrodeposition with an un-submerged IJE resulted in an incorporation of up to 30 vol% SiC particles [217] and a maximum 12 vol% of 50 nm -Al2O3 particles in a nickel matrix [218]. Ultrasonic agitation is also found to be one of the most effective ways to disperse nano-particles in electrolytes [164, 219-221]. The effect of stirring on the co-deposition of particles for micron and nano-size SiC particles into nickel is shown in Fig. 18 [155]. The content of m size particles decreases with increasing stirring rate, while in the case of nano-sized SiC particles, the co-deposition of SiC showed a maximum at a stirring rate of 100 rpm. This is due to an insufficient time for m size particles to get fixed at the surface due to the higher impinging rate of electrolyte. On the other hand, for nm size particles, at very low stirring speed, the agglomeration tendency of particles is high enough and particles settle down to the bottom of the plating tank. As the stirring rate increases, the co-deposition of SiC particles increases up to a maximum. On further increase in agitation rate, the high impinging rate does not allow particles to get incorporated in the metal matrix as was the case for micron size particles.

Figure 18. Deposition rate and co-deposition of SiC particles with Ni vs. stirring rate; (a) type-A micron size SiC particles (2-3 m), and (b) type-B nanometer size SiC particles (45-55 nm). [155].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

263

In some investigations, it was reported that the electrode rotation rate and the nanoparticles concentration in the electrolyte are interrelated, and, at a certain combination, a maximum deposition is possible [103, 143].

4.4 Electrode Geometry

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Concerning the fluid flow conditions, a plate-in-tank geometry provides a much undefined hydrodynamic. The quantitative and controllable description of the fluid flow conditions is on the contrary achieved with rotating disc electrode, rotating cylinder electrode, and parallel plate channels. Three different flow regimes can be distinguished, namely laminar, mixed (turbulent and laminar), and turbulent ones. These flow conditions are of interest to get insight into the theoretical understanding of particles trajectory in plating cells. For micron-sized particles, there is generally no significant influence of electrolyte flow on the particle content under laminar flow conditions. In the transition regime, the particle content increases at higher flow rates. In the turbulent regime, the particle content tends to decrease at higher flow rates [15] due to a continuous impinging thrust by other particles causing a poor adherence of particles onto the electrode surface and their desorption eventually, as mentioned earlier.

Figure 19. Wt.% Al2O3 embedded in copper vs. rotation speed of Riddiford type electrodes. [15].

Figure 20. Wt.% Al2O3 embedded in gold vs. rotation speed of cylindrical electrode. [15].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

264

Subir Kumar Ghosh and Jean-Pierre Celis

Buelens et al. [15] clearly showed the dependence of particle incorporation on the stirring rate during the electrolytic co-deposition of -Al2O3 from an acidified copper sulphate and additive free gold cyanide bath (Figures 19 and 20). For the co-deposition of Cu-Al2O3, they used a flat-shaped disc electrode, whereas a cylindrical disc electrode was used for the codeposition of Au-Al2O3. A comparison between Figures 19 and 20 reveals a shift in the rotation rate at which laminar, transition, and turbulent flow patterns are found. This is due to the use of disc electrodes of different shape and size. It can be seen from Fig.19 that under laminar flow conditions a constant amount of Al2O3 particles is embedded. At the start of the transition zone a marked decrease in embedded alumina is noticed. At increasing rotation rate, an important increase of embedded alumina is noticed until the turbulent zone is reached where an abrupt decrease of the amount of embedded alumina particles occurs. The increased co-deposition of alumina in the transition zone was shown to be due to the formation of Al2O3-agglomerates in the plating solution, and the subsequent co-deposition of agglomerates [222]. For Au-Al2O3 system, a similar observation was done. That study also pointed out that at very high rotation rate the particle content in the coatings decreases as shown in Fig. 20. Other details about the effect of the geometry of the electrode on the co-deposition of particles can be found in literature [31, 33, 182, 223-224].

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.5 Process Type The available literature on electrolytic and electroless composite plating over the last five decades shows that composite plating under DC is the most adopted technique followed by pulse electrolysis, even though this later became available since about 30 years. In DC plating of composites, the applied current or potential are the only parameters (external perturbation) for a given electrolyte, that affect the particle incorporation rate, and hence its contents. However, in pulse plating, parameters like pulse frequency, pulse current density and duty cycle, pulse on-time and off-time, control metal reduction at the cathode as well as particle adsorption or desorption, entrapment into the metal matrix, and its final incorporation. In addition, pulse plating permits higher current densities than the limiting direct current density to be attained, and thus nano-crystalline metals and alloys can be obtained [225-237]. Besides ordinary pulse plating, pulse reverse plating has been found to be very effective in particle incorporation. Another recent technique, namely the two step deposition which is a combination of two different DC cycles for depositing metals and particles separately, is also most promising for depositing composite coatings with high particle content. In this section, the effect of various DC, pulse, and two step deposition parameters on particle incorporation is discussed.

4.5.1 DC Plating Under DC conditions, for ionic constituents, the applied current density (mostly a used practice for electroplaters) has a direct relation as per the polarization behavior at a given condition. It can be either under charge-transfer conditions or diffusion-controlled or limiting current density conditions, and it dictates the structure and properties of the electrodeposits. However, second phase particles being inert, there is no direct correlation between particle deposition rate and applied current density over the whole range. And therefore, the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

265

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

dependence of particle incorporation rate on current density is not unique. It decreases or increases [2, 130, 158, 228] with the appearance of a few maxima in the current density vs. particle content curve. Details are discussed in a review article by Hovestad and Janssen [32]. Sometimes it gives one or two peaks over the whole current density range [15, 96, 100, 188189]. Of course, the relationship between particle content and current density depends on other factors like agitation [139, 229], particle bath concentration [148, 229-231], and particle surface conditions, etc. For example, in the electrodeposition of Ni-SiC composites [139], a continuous decrease of the particle content with current density was reported. The trend was found to increase with the appearance of maxima when the electrode rotation speed and the particle size decrease. In the case of the electrolytic co-deposition of nano-size TiO2 with Ni matrix [232], a non-exclusive dependence between particle content and current density was reported, namely the particle content decreases monotonously at increasing current density. On the other hand for Ni-CNTs composite coatings [233], the carbon content increases with increasing current density up to a maximum of 8 wt.% carbon, and then it decreases. Similarly, Chou et al. [234] found a continuous increase in particle content with current density during the DC deposition of Ni-P-SiC composites. Using a sulfamate bath containing 600 to 800 nm α-Al2O3 nano-particles, an increase in current density resulted in a rough surface microstructure and led to a lower incorporation of nano-particles in the metal deposit [235].

4.5.2 Pulse and Reverse Pulse Plating Both pulse current (PC) and pulse reverse current (PRC) enable a higher particle content as well as the synthesis of coatings with a wide range of compositions and properties. Various authors have attempted to deposit composites via pulse and DC current electrolysis, and they compared the particle content and properties. Promising improvements in terms of the amount and the distribution of incorporated alumina particles have been reported as a result of pulse as well as reverse pulse plating conditions [223, 236-239]. According to Low et al. [34], applying PRC technique for deposition of composites has the following advantages: the amount of nano-particles can be enhanced, a lower quantity of nano-particles in the electrolyte is required to obtain similar properties, and a selective entrapment of nanoparticles with similar sizes can be achieved. All these advantages are associated with the partial dissolution of the metal deposit during the anodic period causing larger sizes of nanoparticles to be lost from the deposit while smaller nano-particles are further entrapped. Of course, by varying the pulse length and duty cycle, one can also extend the entrapment capability towards micron size particles. In DC plating, nano-particles sometimes block the metal deposition, and a parking problem arises. However, the applied off-period and pulse length are limiting factor in deciding the grain size and the particle adsorption-desorption phenomenon. In fact, during DC co-deposition of nano-Al2O3 in Ni matrix, the nano-particles (25 nm Al2O3) dictate the growth of nickel grains (~ 30 nm), whereas under PC conditions, the pulse length is the limiting factor for grain size (~20 nm) [34]. Here, the selective entrapment of nano-particles occurs, but the larger particles are removed preferentially during the reverse pulse period, producing a finer deposit with a less agglomeration of nano-particles. Pulse frequency plays also a role on the incorporation of particles into a metal matrix. At increasing pulse frequency the volume fraction of sub-micron size Al2O3 in nickel increases at a constant duty cycle [240], which is in agreement with previous observation [241]. There

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

266

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

is a drop in particle incorporation till 100 Hz which is in agreement with another study by Bahrololoom and Sani [239]. The longer off-time at lower pulse frequency is responsible for a particle desorption. It is worthy to note that at increasing pulse frequency the texture of coatings evolves from (111) to a random orientation, and that the hardness of the coatings decreases slightly due to micro-structural changes even though volumetric content of particles increases. Thiemig et al. [170, 242] also obtained higher amounts of alumina particle incorporation at lower average current density. The amount of co-deposited alumina ranged from 2 to 11 vol% depending upon the current modulation and the plating parameters. Celis et al. [13, 15, 17, 18, 241] had incorporated -Al2O3 of 50 nm size in Au and Au-Co matrix to obtain hardened and wear resistant gold deposits for various electronic contact devices. In that study, a maximum 0.03 wt% of Al2O3 nano-particle could be deposited by DC plating, but it decreased under pulse plating conditions. A similar observation was done by Wang et al. [243] while incorporating ZrO2 (10-30 nm) in nickel: the particle content was higher under DC plating than under PC and PRC composite plating under the same electrodeposition conditions. Sometimes uniform and well-dispersed nano-particles in composites are obtained by a combination of ultrasonic agitation and pulse plating [244].

4.5.3 Two Step Deposition Besides pulse and DC plating, a two step deposition consisting of an electrophoretic deposition of a thin layer of particles followed by in the second step electro- or electroless deposition of a metal layer onto the cathode surface, is also an effective way to obtain composite coatings with a very high volume percent of second phase particles. Using this two step technique, Shrestha et al. [245] incorporated 63 and 67 vol% Al2O3 and BN particles respectively in nickel from electrolytes containing 5 g/l particles. From a bath containing both 5 g/l BN and 2.5 g/l Al2O3 particles in the electrophoretic bath, 55.5 and 6.2 vol% BN and Al2O3 respectively in nickel were obtained. Under the same electrolytic condition, the electrodeposition of Ni-Al2O3 and Ni-BN composites from a suspension in a single step resulted in a particle content of 18 and 8.4 vol% respectively. The wear performance of NiBN composite coatings was always better than the one of Ni-Al2O3 coatings regardless the deposition technique used. However, all three composite coatings deposited by the two step method showed a substantially better wear resistance than those prepared by the one step method. The two step deposition was also applied for depositing Ni-ZrO2 [246]. A good bonding between coating and substrate was achieved. In another attempt [178], the electrophoretic deposition of Al2O3 followed by the electroless deposition of Ni resulted in an incorporation of 23.4 vol% ceramic particles along with phosphorous in a nickel matrix. 4.5.4 Electroless Plating The electroless plating of composites is also very popular because of its few size and shape limitations, and good throwing power. However, electroless solutions are by their nature operated on a slender knife edge of stability; it is important to maintain the solution free of suspended, fine particles. However in composite plating that is what one has to do. If electroless deposition starts on the particles, then the electrolyte very rapidly degrades. When particles are partly engulfed in the metal matrix, then it is important that they are covered with an adherent metallic layer. In order to overcome these difficulties, one has to make sure

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

267

that the electroless bath is stable even when second phase particles are suspended in it. The stability of the suspension again depends on the particles surface charge and their zeta potential. The most effective way of dispersing the particles into the electroless plating solution and to avoid agglomeration, is by getting the right particle surface chemistry i.e. by using surfactants as mentioned earlier. The most common method is to use combo-surfactants what is a mixture of ionic and non-ionic surfactants in suitable proportions [183-184, 193]. The successful incorporation of micron size PTFE particles in a Ni-P matrix was reported [193, 194, 247-248] and resulted in a low coefficient of friction. A maximum incorporation of 25-30 vol% PTFE in Ni-P matrix was achieved with particles of 8-9 micrometers [193]. At decreasing particle size, the particle content in the composite was found to decrease [185]. Other studies show that from an unstirred electrolyte, the particle content in composite coatings increases with increasing bath loading and saturates at a certain level [114-115]. Coatings obtained from stirred electrolytes show a maximum content of particles at increasing bath loading [247]. The combo-polyelectrolytes were recently found to be effective in dispersing nano-size  or -Al2O3 particles in the electrolyte [249-250] and to co-deposit up to 50 vol% particles. Besides polyelectrolytes, Na3Co(NO2)6 complex was also found to promote the incorporation of SiC particles in Ni-P [150] Electroless plating of Ni-P-SiC [169, 172, 251-252], Ni-P-Si3N4 [253], Ni-P-B4C [254255], Ni-P-diamond [256], Ni-P-SiO2 [257], Ni-P-TiO2 [258], and Ni-P-graphite [259] are also popular thanks to their very good corrosion and abrasive wear resistance.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

4.6 Conclusion The effect of various process parameters on the particle content in composite coatings was reviewed. It points out that the selection of a suitable bath chemistry allows to avoid agglomeration, while particle concentration, applied current form (either DC or pulse), hydrodynamics and electrode geometry determine the level of co-deposition that will be achieved. For nano-composite coatings, bath chemistry has to be optimized and pulse plating practice seems to be most promising to achieve desired coating properties.

5. FUNCTIONAL PROPERTIES Industrial applications of composite coatings factually lie on enhanced extrinsic properties like hardness, wear, lubrication, corrosion resistance, catalytic, and various other functional properties. The functionality of such coatings depends on parameters like second phase particle size, grain size of the metallic matrix, microstructure, internal stress, and adhesion to the substrate. Therefore, in order to optimize properties, the correlation between process parameter, microstructure, and functionality, should be understood. Even though the reinforcement of metal matrices by the incorporation of particles has been known for many decades, the ever-increasing demand for better coatings to extend the lifetime of various industrial components boots the search for better performing coatings. The increasing availability of various nano-size ceramic powders and their subsequent incorporation in a

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

268

Subir Kumar Ghosh and Jean-Pierre Celis

metal matrix, further push the enhancement of properties towards new limits. Considering the incorporation of micron down to nano-sized particles, its effect on extrinsic properties like hardness, wear and lubrication, corrosion, and tribocorrosion, is reviewed hereafter.

5.1 Mechanical and Tribological Properties

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The discussion in Section 2 on the strengthening mechanisms achieved by the incorporation of second phase particles into a metal matrix, allowed to identify the factors responsible for hardening of composites, namely; (a) particle size, (b) particle number density, (c) inter-particle distance, (d) particle agglomeration, (e) grain size of the matrix metal, and (f) microstructure and texture of the metal matrix. Electrolytic and electroless composite coatings that contain hard ceramic particles like oxides [260-262], carbides [136, 149, 254], nitrides [209, 253], borides [263], silicides [264], and many others particles [265-266] leads to hardening of the metallic counterpart. Hardness, tensile strength, yield strength, and ultimate strength of such composite coatings increase, but their ductility decreases with a few exceptions in nano-composites [266]. Many reports [260, 267] indicate that with increasing particle content, the hardness increases up to a certain level, and then decreases due to an increasing brittleness of the metal matrix. Sometimes, hardness reaches more than twice the value of the pure metallic counterpart [268]. On the other hand, the incorporation of soft second phase particles, like MoS2, graphite [259], PTFE [248, 255, 271], or polymeric microcapsules, leads to a lowering of hardness and strength. Wear is a very complicated material degradation process that depends on physical, mechanical, and chemical properties of two or three contacting bodies. The following structural factors dictate to a large extent the tribological behavior (friction, wear, and lubrication) of electrolytic composite coatings, namely: a) b) c) d) e) f) g)

hardness, internal stress, second phase particle size, particle-matrix interface, particle agglomeration, self-lubrication properties, and tribocorrosion behavior of coatings.

Other synergistic effects like wear particle size, work-hardening, debris entrapment into wear tracks, phase changes, or surface film formation in sliding wear tracks affect the dominant wear process. Composite coatings containing hard particle like SiC [53, 109, 136, 267], BN [210, 245], B4C [186, 254, 255], WC [271], Al2O3 [13, 17, 260, 245], TiC [147], Si3N4 [272], diamond [273-274], CNTs [265], and ZrO2 [243] were shown to exhibit a much better abrasive wear resistance compared to pure metal matrix materials. Sometimes the improvement in wear resistance is many times higher than the one of pure metallic counterparts [147]. The other important engineering aspect of tribology is to achieve a low coefficient of friction and to lower the effective mass loss. The incorporation of soft particles like graphite

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

269

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[259], PTFE [115, 193, 269], polymer microcapsules containing lubricating oils [275], CNTs [276], and MoS2 [116], allows to reduce the coefficient of friction in sliding systems. Therefore, depending upon the applications, either hard ceramic particles or soft layered molecular particles can be embedded in a metal matrix to obtain the required functional tribological properties. Fields of applications as engine parts, saw blades, gages, and drills, require hard wear-resistant coatings. On the other hand, self-lubricating components, heat exchanger pipes, threads and water repellent component require the incorporation of soft particles. Among the various wear modes, composite coatings containing hard particles may be beneficial to prevent abrasive wear to occur, and so to protect the base materials. It seems from literature that material loss due to abrasive wear generally follows the hardness profile of composites except at high volume fraction where brittleness enhances the material loss. However, higher hardness does not necessarily lead to a lower wear loss. That is true especially for composite coatings containing submicron or nano-size particles that contain a low volume fraction of particles and thus have a still low hardness. They may show a better wear performance compared to pure metallic materials and composite coatings containing micron size particles [136]. The achievement of a uniform distribution and an optimum number density of particles in the metal matrix, are a prerequisite to achieve the best wear performance. In fact, the later is achieved when the second phase particle size is in the nanometer range. Indeed, smaller particles not only impart dispersion hardening to a matrix but lower also the degradation by abrasion. As a consequence, nano-composite coatings can show a better wear resistance than micro-composite coatings provided that an extensive particle agglomeration is avoided during electrodeposition. The functionality of a number of composite coatings of interest in transportation is discussed hereafter, namely: (a) cobalt-hardened Au, (b) Ni-SiC, (c) Ni-P-SiC, and (d) composite coatings containing nano-particles emphasizing both micro- and nano-composite aspects.

5.1.1 Tribological Properties of Cobalt-Hardened Gold Composite Coatings for Electrical Connectors Research by Celis et al. since 1980‘s [13, 16-18], clearly showed that different microstructures of electrolytic composite coatings coincide with a change in crystallographic texture induced by the incorporation of particles that plays a crucial role on the resulting micro-hardness. Recently, several investigators [138, 277-279] supported those findings. The alteration of the crystallographic texture of a metal matrix modifies the atomic density and the lattice strain [280], thereby affecting the hardness. That effect is significant when the size of the reinforcing particles is in the nanometer range. During electrolytic deposition of cobalt-hardened gold coatings, changes in the structural and textural characteristics were observed at increasing current density as shown in Figures 21 (a) and (b). It is interesting to notice that bright deposits obtained from an additive-free gold plating bath, exhibit a preferential (100) crystallographic texture, while in the case of cobalt-hardened gold, a (111) texture dominates in such bright deposits. The corresponding surface morphology also changes systematically. In cobalt-hardened gold coatings, the carbon content increases strongly at increasing current densities reaching a plateau value of 0.3 wt%, while at the same time the hardness drops from 160 HV down to 120 HV. The source of carbon is basically due to the precipitation of potassium cobalt complex species

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

270

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

KCo[Au(CN)2]2 [281] in the gold matrix. These complex species act as second phase particles resulting in a dispersion hardening of the metallic Au-Co matrix. Besides this dispersion hardening due to the formation at the cathode of Co-complex species and their subsequent co-deposition in the metal matrix, the modification of the crystallographic texture induced by this co-deposition also increases the strength to the matrix. In monolithic crystals, the atomic density varies with the crystallographic orientation. Therefore, the higher the atomic packing fraction is, the lesser dislocations can move in the crystal under applied load. Crystals with a (111) orientation have a higher atomic density and a lower movement of dislocations as compared to crystal with a (100) orientation under loading. In fact, literature on microstructure clearly demonstrates that a microstructure with a preferential (100) texture, provides deposits possessing a maximum ductility [282], a minimum hardness, and a minimum internal stress [283-284].

(a)

(b) Figure 21. (a) Overview of the structural characteristics of additive-free hard gold electrocatings, (b)Overview of the structural characteristics of cobalt-hardened gold electrocatings. [16].

Celis et al. [13, 16] also found that the incorporation of Co-complex precipitates governs the wear rate that is related to grain size, microstructure, hardness, and internal stress of electrolytic Co-hardened gold coatings. Besides a hardening effect that appeared to be of minor importance, the presence of complex cobalt species at the coating surface is essential

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

271

for achieving wear resistant coatings. This was pointed out by Solomon and Antler [285]. It was also noticed that high internal stress levels (>100 N/mm2) are required in additive-free cobalt-free gold coatings to avoid adhesive wear. According to literature [286], high internal stresses promote abrasive wear. Since the internal stress is closely related to the amount of incorporated metallic cobalt (Fig.22), a high metallic cobalt content (Com > 0.4 wt%) causes abrasive wear. The most remarkable and most important observation by Celis et al. [13, 16] was that the ratio of Cok/Com (k stands for complex and m stands for metallic) dictates the wear behaviour of cobalt-hardened gold coatings. That Cok/Com ratio is a better indication of the wear behaviour than Withlaw‘s criterion [287] stating that a ratio K/Co > 1 can be associated to wear resistant coatings, while a ration K/Co < 1 leads to galling wear. In fact, the ratio of Cok/Com was found to govern the grain size, the internal stress, as well as the wear resistance of cobalt-hardened gold coatings.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 22. Macro-residual stress v. metallic cobalt content for jet-plated cobalt-hardened gold coatings (bold symbols refer to wear-resistant coatings. [13].

Figure 23. Cok/Com vs grain size for jet-plated cobalt-hardened gold coatings (bold symbols refer to wear-resistant coatings. [13].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

272

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 24. Wear criterion for jet-plated cobalt-hardened gold coatings (bold symbols refer to wearresistant coatings as proposed by MTM, K U Leuven. [13].

Figure 25. Wear behavior vs. structural properties for jet-plated cobalt-hardened and cobalt-free gold coatings. [13].

The ratio of complex to metallic cobalt (Cok/Com) is plotted against grain size in Fig.23. It is evident that when Cok/Com > 1, wear-resistant, cobalt-hardened gold coatings are obtained. On the contrary at Cok/Com < 1, the coatings suffer from either galling or abrasive wear. Fig. 24 shows another representation of this criterion applied for the same set of samples. It shows that cobalt-hardened gold coatings that suffer from galling, contain low amounts of incorporated complex cobalt, and have a high ductility. The co-deposited complex cobalt species act either as a lubricant or as a ductility reducer. Combining all factors correlating cobalt concentration to grain size, internal stress, and wear resistance, Fig. 25 becomes a most useful graph to identify the optimized structural parameters (and thus indirectly the optimized plating parameters !) for obtaining wear-resistant, cobalt-hardened gold composite coatings.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

273

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

5.1.2 Tribological Properties of Ni-Sic Composite Coatings as Wear-Resistant Layer on Engine Cylinders As far as the effect of particle size and number density on hardness is concerned (see Section 1), it can be stated that for a given volume fraction, the lower the particle size is, the higher is the ‗number density‘, and the lower is the inter-particle spacing. The number density increases with increasing volume fraction for any kind of particle size, as far as agglomeration is not occurring. For electroplated Ni-SiC composites, Garcia et al. [136] showed that the hardness of composite Ni-SiC coatings increases with the volume fraction of particles incorporated in the Ni-matrix, and that hardness depends on the particle number density. Using three types of SiC particles of different nominal sizes namely 0.3, 0.7, and 5 m, they showed that the strengthening mechanism evolves with particle size due to an alteration of the inter-particle spacing as shown in Figures 26a and b). For particle sizes below 1 m, a dispersion strengthening mechanism dominates while above that size a particle-reinforced strengthening is significant because the inter-particle spacing is above 5 m [41]. Under such conditions, an applied load is carried both by particles and matrix unlike the former case where the applied load is carried only by the metal matrix.

Figure 26. (a) Vickers hardness of pure nickel and composite Ni–SiC coatingscontaining SiC particles of three different sizes plotted vs. the vol.%of co-deposited SiC-particles. [136].

Figure 26. (b) Vickers hardness of composite Ni–SiC coatings containing SiC particles of three different sizes plotted vs. the SiC inter-particle distance [136] in the coatings [4]. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

274

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Similarly, the authors also showed that the wear factor decreases with increasing number density both in uni- and bi-directional sliding experiments following the hardness profile as shown in Figures 27 (a) and (b). The coating containing SiC particles of 0.3 m size shows the best wear resistance as compared to nickel composite coatings containing 5 m SiCparticles. Shrestha et al. [109] showed that in presence of AZTAB a maximum 71.5 vol% of SiC particles can be incorporated in electrolytic nickel, and that the wear resistance of such composite coatings depends on the volume fraction of particles in the coating. The wear resistance increased with increasing content of SiC particles in the composites up to about 52 vol.%. The coating containing 71.5 vol.% of SiC particles exhibited a lower wear resistance than Ni-coatings free of SiC particles. When the volumetric content of particles is increased above that amount, the deposited nickel probably can not bind all particles tightly within the matrix, and the brittleness of the deposits of such coatings increases.

Figure 27.(a) Volumetric wear factor under uni-directional sliding on composite Ni–SiC coatings containing SiC particles of three different sizes, plotted vs. the number density of co-deposited SiC particles. [137].

Figure 27.(b) Volumetric wear factor under bi-directional sliding on composite Ni–SiC coatings containing SiC particles of three different sizes, plotted vs. the number density of co-deposited SiC particles. [137]. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

275

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 28. Weight loses of Ni–SiC co-deposition film after 90 min wear testing and at 89.2N loading (C1-Ni-5.2vol% SiC, C2-Ni-6.9 vol% SiC, C3-Ni-9.2 vol% SiC, D1-Ni-7.9 SiC, D2-Ni-9.8 SiC, and D3-Ni-11.5 SiC) [288].

Figure 29. Summary of mechanical properties for Ni-SiC composites for conventional and nanocrystalline Ni matrices produced by electrodeposition. [225].

The incorporation of nano-size SiC particles (293 nm) in electrolytic nickel enhances hardness and wear resistance at increasing particle content in the Ni-matrix (Fig. 28) [288]. However, the required amount of particles is much lower than in composite coatings containing micron and sub-micron size particles. Zimmermann et al. [266] showed that Ni-SiC composite coatings consisting of a nanocrystalline Ni-matrix and submicron size SiC particles at less than 2 vol% [225, 266], exhibit a significant increase in hardness, yield strength, ultimate tensile strength, along with an improved ductility compared to composites consisting of a conventional nickel matrix (Fig. 29). Pulse plating did lower the Ni grain size down to the nanometer range (10-15 nm). At a SiC content above 2 vol%, the strength and ductility decrease. That study also showed that particle agglomeration plays a significant role on the tensile properties of such composites. Such agglomerates produce locally ductile or ‗soft‘ regions in a hard, relatively brittle nanocrystalline Ni matrix, due to an enhanced void formation between neighboring particles [289290]. Such voids effectively increase the porosity of an otherwise fairly dense material, and eventually compromise mechanical properties of nano-composites.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

276

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Particle agglomeration inside conventional Ni-SiC and nano-crystalline Ni-SiC matrix explains that detrimental effect on tensile properties. Fig. 30a depicts mobile dislocations pinned by SiC particles, and SiC-agglomerates in a conventional polycrystalline Ni–SiC matrix leading to dispersion strengthening. Dislocation pile-up at grain boundaries can also occur. This leads to a large increase in strength compared to conventional polycrystalline nickel. In Fig. 30 (b) depicts a microstructure consisting of particles and agglomerates of SiC in a matrix of nano-crystalline nickel. The individual sub-micron size particles are much larger than the nano-sized nickel grains. The presence of micro-pores between adhering particles in agglomerates is also depicted. As the concentration of SiC particles in a nickel matrix made up of nano-sized grains increase, damage accumulates and porosity increases weakening the nano-composite. Therefore, small volume fractions of SiC are more effective to increase the strength of nano-composite than larger ones that are actually detrimental to strength.

Figure 30. (a) A schematic of the microstructure of a conventional Ni–SiC MMC showing SiC particles pinning dislocations. (b) A schematic of the microstructure of Nano-Ni–SiC showing the presence of micro-voids in SiC agglomerates [266].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

277

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 31. Plot of wear scar depth vs. vol.% of incorporated SiC in the Ni/SiC composite coatings [109].

Figure 32. Volumetric wear factor values of composite Ni/SiC deposits prepared under direct and pulse plating conditions, at a frequency of v = 0.1 Hz and various duty cycles [278].

In some studies [109, 271], it was pointed out that at increasing particle concentration in the electrolyte, the particle content in the coating increases. However, a very high concentration leads to agglomeration of particles in the deposit what lowers the expected extrinsic properties. For example, Shrestha et al. [109] showed that at high SiC content, the electrolytic Ni-SiC composite suffers a high wear due to the brittleness of the matrix because of agglomerated SiC particles in the deposit (Fig. 31). Recent investigations [138] on texture of Ni-SiC composites containing micron or nanosize SiC particles reveals that the crystallographic texture evolves from a (100) texture for pure nickel to a (111) texture for composite coatings [138]. Here the embedding of SiCparticles in nickel matrix causes the soft-mode (100) texture to evolve into a mixed preferred orientation of nickel (100) and (211) crystallites, and effectively increases hardness in a very similar way to the cobalt-hardened Au composites as reported previously by Celis et al. [16]. Of course, this higher hardness has also to be partly attributed to other facts like a reduction of the nickel grain size in presence of SiC-particles, and the presence of second phase particle in the nickel matrix. Such a synergistic effect on hardness arising due to changes in

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

278

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

microstructure and grain size was also noticed in electrodeposited Ni/SiC compositionallygraded, multilayered coatings [291]. Crystallographic texture and grain size modification play an important role in minimizing the wear loss on Ni-SiC coatings [278]. The SiC particle content in pulse plated Ni-SiC coatings containing either nano or micron size SiC particles, is the lowest one for nano-SiC composites electrodeposited at 50% duty cycle and f = 0.1 Hz. Surprisingly, this coating has a better wear resistance than the other ones (Fig. 32). The authors explain this observation in a similar way as hardness enhancement, namely a change in preferred crystalline orientation from (100) to (111) that causes a more compact lattice structure, a refinement in grain size, and, hence, better wear resistance.

5.1.3 Tribological Properties of Ni-P-Sic Composites as Wear-Resistant Layer on Complex-Shaped Engineering Parts The incorporation of phosphorous into electrolytic or electroless nickel increases the hardness as compared to pure electrodeposited Ni coatings. Upon annealing, the hardness increases up to a plateau value, and then decreases at a higher annealing temperature. The inflection point depends on the phosphorous content, the annealing temperature, and its duration. This increase of the hardness is attributed to the formation on annealing of a Ni3P inter-metallic phase acting as second phase particle like in composites. The addition of SiC into a Ni-P matrix further enhances the coating hardness. A heat-treatment of such a composite ends up in a composite structure containing two different second phase particles viz. NiP and SiC causing the hardness to rise further compared to as-plated Ni-P-SiC composites [292-293]. Not only hardness but also the wear modes of as-plated and heattreated Ni-SiC composites are drastically different. Research by Celis et al. [292-293] pointed out that it is not the abrasive wear alone that determines the total mass loss, but the contribution of oxidational wear is also important. This oxidation results from a burning of the top layer due to the flash heating occurring in dry sliding contacts. In the case of as-plated NiP and Ni-P-SiC composite coatings, a heat treatment at 4200C for 1 h reduces the wear loss on both coatings due to an increase of the hardness linked to the formation of a hard Ni3P phase. However, the wear loss on Ni-SiC composite coatings was larger than on NiP coatings due to an induced sensitivity for crack formation around NiP/SiC interfaces in both uni- and bi-directional sliding wear tests as can be seen from Fig. 33. The competition between oxidational and abrasive wear was noticed on both as-plated and heat treated Ni-P-SiC coatings, however, the oxidation film formed in the wear track during uni-directional sliding adheres strongly to the coatings protecting them from a high wear rate. The wear rate in bi-directional sliding tests was found to be smaller than the one in uni-directional sliding tests. This is due to the entrapment of wear debris in the sliding contact area that modifies the interaction between two rubbing bodies leading to the subsequent formation of a compact oxidation layer (Fig.34) in and around the wear track leading to a lower mass loss. That tribocorrosion determines the overall wear mass loss. A similar observation was made by Xinmin et al. [267] on electroless deposited Ni-P-SiC composite coatings.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

279

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 33. Details of crack formation in the oxide layer (black area of wear scar) showing particle removal and oxidative wear during uni-directional sliding [292].

Figure 34. Microstructure of the wear tracks on heat-treated composite NiP–SiC coatings (600 nm 80 g/l). Bi-directional wear test parameters: corundum ball as counter body, 10 N; 2 Hz, 20000 fretting contacts, ambient air 50% RH and 23 °C [293].

The crystallographic texture evolves also by the embedment of SiC particles in NiP coatings [234], similar to Ni-SiC composite coatings, and that embedment affects greatly the phosphorous content of the composites. That induces a structural transformation from (111)Ni at high phosphorous content to a more random crystallographic texture that consists of (111)Ni, (200)Ni and (220)Ni planes at lower phosphorous content. A maximum hardness of 770 HV is reached at about 3.7 wt% P. At high SiC content, a significant agglomeration of sub-micron size SiC (0.3 m) particles takes place and the hardness decreases.

5.1.4 Functionalities of Composite Coatings Containing Other Nano-particles Unlike micron size particles, a quite low particle loading is effective in the case of nanosize particles to obtain good properties. Of course, in order to disperse nano-particles in the plating solution, one or more surfactants are required in order to avoid a deleterious

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

280

Subir Kumar Ghosh and Jean-Pierre Celis

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

agglomeration either in the plating solution or on co-deposition in the electrolytic or electroless coatings. However, surfactants sometimes alter the microstructure of the electrolytic composite coatings and their functional properties. For example, during the electrodeposition of carbon nano-tubes with nickel (Ni-CNTs), it was seen that surfactants affect hardness strongly [233]. Both sodium dodecyl sulphate (SDS) and cetyl tri-methyl ammonium butyl (CTAB) reduce the grain size. The preferred (220) planes obtained in coatings in absence of surfactants becomes (200) and (111) planes in presence of SDS and CTAB, respectively. The corresponding hardness for Ni-CNT coatings obtained in presence of SDS is much higher than the one measured on coatings produced with CTAB and without surfactant as shown in Fig.35.

Figure 35. Effect of surfactants on hardness of nickel-CNTs composite coatings formed in solutions with different CNTs concentrations [233].

Figure 36. Effect of CNTs concentration on deposited coating hardness before and after heat treatment [294]. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

281

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

In case of Ni-Cu-P-CNT composites, the hardness increases with an increasing amount of CNTs in the electrolyte at first, but then decreases because of upcoming brittleness [294]. Note the very low particle concentration in the plating solution. The hardness further increases by performing a heat-treatment that leads to the formation of a dispersed hard Ni3P phase. The effect of grain refinement and second phase particle agglomeration on hardness was reported [295] for electrodeposited Ni-ZrO2 nano-composite coatings. The hardness of such Ni–ZrO2 coatings containing either mono-dispersed or agglomerated ZrO2 nano-particles, was 529 and 393 HV, respectively. The strengthening was attributed to a combination of particle reinforcement and nickel grain size refinement. Inorganic fullerenes (IF) like WS2 and MoS2 particles show an ultralow coefficient of friction [296-297], and their incorporation in a suitable matrix can result in self-lubricated coatings. Very recent studies [200-201] on self-lubricated composite coatings containing fullerene like WS2 nano-particles revealed a promising tribological properties. These nanoparticles provide lubricity even in a humid environment, and are of interest in orthodontic applications [200, 202]. Fig. 37 shows the reduction of coefficient of friction recorded on electrodeposited Co-IF-WS2 and electroless Co-IF WS2 sliding against a steel-bearing ball. A comparison is given with pure cobalt and stainless steel.

Figure 37. Time dependence of the coefficient of friction recorded on stainless steel, cobalt and two composite cobalt coatings containing inorganic fullerenes: (A) electroless Co + IF and Co coatings; and (B) electrolytic Co + IF and Co coatings under dry sliding condition. Counter body was steel bearing ball [200].

5.2 Corrosion Behavior Present-day coatings need to possess multiple functions. Not only an improvement in mechanical properties is asked for but it should be combined with a better corrosion resistance, a better electrical conductivity for electrical appliances or micro/nanotechnologies, a brightness for decorative purposes, or a better wear resistance in engineering parts, depending upon their field of applications. The high tech use of various electrolytic or electroless composite coatings in automotive, aerospace industries, heat-exchanger etc. requires an operation in direct or indirect contact with an aggressive chemical environment

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

282

Subir Kumar Ghosh and Jean-Pierre Celis

either. Therefore, the corrosion resistance of these composite coatings on various appliances is a critical issue and an improved corrosion resistance is desirable. Corrosion is an electrochemical phenomenon associated with the exchange of electrons between two parts of a metallic material. One part acts as anode (M→ Mn+ + ne) and hence dissolves material while the other part acts as cathode. The former process slows down

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1) 2) 3) 4)

by making area of anode very large compared to cathodic part, by reducing the potential gradient between two parts, by forming passive film with low conductivity, and /or by mechanically covering the exposed anodic surface.

Basically, particle incorporation in composite coatings reduces the available surface for oxido-reduction reactions [298] and the porosity [88]. However, this cover-up process depends on particle size, metal/particle interface integrity, and the creation of excess interfaces beyond metal-metal grain boundaries. In fact, the grain boundary area is in general most susceptible for a chemical attack and varies with grain size drastically on going from m down to nm. That depends on the characteristics of the second phase particles as discussed in previous sections. Sometimes, it increases the corrosion resistance thanks to a stronger passivation of the surface and a better localized corrosion resistance of a nano-crystalline or amorphous matrix [299-300]. Besides these, electrodeposition parameters and post-deposition treatment also affects much of the microstructure and the corrosion resistance of electrolytic and electroless composite coatings. Because of this complexity, the corrosion resistance of composites varies widely. So far, no unique trend has been found for a given metal-particle system. For some composites, the corrosion rate decreases [88, 301-302] whereas for others the opposite [88, 210, 302] is observed. For example, nickel composite coatings containing PTFE [303], SiC [121, 145, 210, 298], TiO2 [88, 194, 302], CNTs [233, 294], Al2O3 [88, 302], SiO2 [302], or Si3N4 [304], particles show a decreased corrosion rate under certain conditions compared to pure nickel coatings. A further enhancement of the corrosion resistance was noticed in Ni-P coatings containing PTFE, or Cu and PTFE and W and SiC nano-particles. In a recent study, Abdel Aal [145] showed that the shape of SiC nano-particles affects the corrosion resistance of NiW-P-SiC nano-composite coatings in 3.5 wt% NaCl. Such nano-composite coatings containing SiC nano-rods exhibits a higher corrosion resistance compared to deposits containing spherical SiC nano-particles and Ni-W-P alloy coatings. This higher corrosion resistance is attributed to an accelerate passivation of Ni-W-P matrix [116] by the embedment of nano-particles into crevices, micron holes and very small pores [305]. Garcia et al. [298] indicated that the micro-structural modification induced by the codeposition of SiC particles in nickel also affects the corrosion behavior. They found that the corrosion resistance of Ni-SiC composites is independent of particle size and is much higher compared to pure Ni coatings. But Ni-SiC composites containing submicron size particles are more resistant to localized corrosion compared to their micron-size counter part. This enhanced corrosion resistance of Ni-SiC containing sub-micrometric SiC-particles is linked to a change in grain morphology from elongated columnar grains to smaller equiaxed grains, in the texture of the coatings, and their internal stress. However, for Ni-P-nano SiC composite coatings, Malfatti et al. [121] showed that the presence of SiC particles increases the corrosion properties i.e. the corrosion potential of the composites compared to Ni-P alloys.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

283

With the increase in particle content, the corrosion current density increases for composite coatings containing smaller particle sizes, and after heat treatment.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 38. HRTEM examinations of the interface between nickel grains and incorporated nanoparticles (electrodeposition done at 2 A.dm2, and 50 g/l particles in the electrolyte) [88].

The nature of the particle-matrix interface also dictates the corrosion performance of composite coatings. For example, composites like Ni-nanoTiO2 and Ni-nano-Al2O3 [88] show a different corrosion resistance due to a difference in the structure of the Ni-particle interface. Ni-TiO2 coatings show a better corrosion resistance than Ni-Al2O3 composites. High resolution transmission electron microscopy (HR-TEM) revealed that the Ni/TiO2 interface is compact and pore-free. However, the Ni/Al2O3 interface is incoherent, and is found to be the possible site for crevice corrosion as shown in Fig.38. The oxidation resistance of composite coatings at high temperature is another interesting property for different applications. Several composite coatings like Ni-WC [306]], Ni-P-SiC [307]], Ni-P-Cr2O3-TiB2 [307], Ni-P-Cr2O3-TiB2-ZrO2 [307], Ni-Cr2O3 [301], Ni-Si3N4 [304], and Co-Cr3C2 [301] show an enhanced high temperature oxidation resistance. At high temperature, metal oxide scales are formed at the metal-air interface. In composites, the interdiffusion between particle and matrix at elevated temperature changes the chemistry of the interface and thus its oxidation behavior. For example, in Co-Cr3C2 [301] composite coatings the carbide particles supply chromium to a mixed cobalt/chromium oxide layer that improves the corrosion protection. The oxidation resistance further improves if the heat-treatment is done in an oxygen-free atmosphere that enables the formation of a cobalt-chromium alloy showing a better corrosion resistance than conventional cobalt-chromium alloys [301]. In some studies [307], it was shown that a mixture of second phase particle shows a better oxidation resistance than a single one. For example, a Ni-P alloy containing a mixture of Cr2O3 and TiB2 particles oxidizes at a slower rate in comparison to composites containing only one of these particles. The addition of ZrO2 particles into Ni-P-Cr2O3-TiB2 composites was found to improve further the oxidation resistance due to a suppression of the negative effect of TiB2. Besides this, Ni-P-Cr2O3-ZrO2 composite coatings also show a very good

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

284

Subir Kumar Ghosh and Jean-Pierre Celis

oxidation resistance. However, in some cases due to a breakdown of the metal-particle interface at elevated temperature metal oxidation takes place faster and the coatings show a poor high temperature corrosion resistance. In the same way, on Ni-SiC [304] and Ni-TiC [147] coatings, the formation of nickel oxide is favored and these composite coatings have a poor corrosion resistance at high temperatures.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

5.3 Tribocorrosion Behavior Tribocorrosion or corrosion-wear describes the material behavior in systems where interactions between tribological and corrosion processes take place simultaneously. In such tribocorrosion systems, materials are subjected to mechanical, chemical, and electrochemical loadings. Tribocorrosion is defined as the chemical–electrochemical–mechanical process leading to a degradation of materials in sliding, rolling, or erosion contacts immersed in a corrosive environment. That degradation results from the combined action of corrosion and wear. The mechanism of tribocorrosion is not yet fully understood due to the complexity of the chemical, electrochemical, physical, and mechanical processes involved [308-310]. Examples of the occurrence of tribocorrosion in service are the accelerated corrosion of steel conveyors exposed to ambient air of high relative humidity, the fall out of electrical connectors in the automobiles, the degradation of hip prosthesis and dental fillers, the erosion wear of turbine blades, and in many industrial components, e.g. bearings, pumps, and rolling mill bearings that operate in water that is directly or indirectly introduced as a coolant or present as a working fluid. Passive metals are particularly sensitive to tribocorrosion because rubbing can destroy their passive surface film so that both corrosion and wear rate are large before the surface re-passivates. It was observed that the material removal in a tribocorrosion system usually exceeds the sum of mechanical and corrosion contributions measured separately [311312]. Over the past few years, the mechanisms of wear and corrosion-wear of stainless steels were studied by different techniques [313-318]. Concerning the tribocorrosion behavior of composite coatings, Benea et al [319] recently showed a few results on the tribocorrosion behavior of Ni-SiC composite coatings containing micron and nano-size particles. For Ni-SiC composite coatings containing micron size particles (30 m), the study showed that the incorporation of non-uniform SiC agglomerates causes a most active surface and a high material loss due to the exposure of maximum-bared Ni-surface during friction. The coatings containing a uniform distribution of SiC particles show a very good wear-corrosion resistance (Fig. 39). Of course, in both cases the appearance of passive-active-repassive potentials during unloaded-loaded-unloaded cycles indicates the sensitivity of the top layer for degradation. A similar but clearer pattern was shown by considering Ni-nano-SiC composites [320] as shown in Fig.40. A passivation-depassivation taking place during each cycle is made evident by the introduction of a latency time at the end of each sliding cycle allowing the passivation kinetics to form a thin layer before its removal by mechanical scratching. This study also highlights that at increasing applied load, the wear volume increases as evidenced by potentiodynamic polarization plots and wear-corrosion rates (Figures 40 and 41). No evidence of localized corrosion was observed. Very recently, Benea [321] showed a similar accelerated material loss on Ni-ZrO2 composites during tribocorrosion tests in 0.5 M K2SO4 in a pin-on-disc tribometer.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

285

Figure 39. Variation of the open-circuit potential of Ni–SiC nano-structured coatings immersed in 0.5M K2SO4. (a) Variation of free potential after immersion in the solution without friction; (b) variation of the free potential during intermittent friction test with a latency time of 20 s; and (c) variation of the free potential before loading (area A), during continuous friction (area B), and after stopping the friction (area C) [319]. Without friction Continuous friction 10N

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Continuous friction 15N

Figure 40. Potentiodynamic polarization curves of Ni–SiC nano-structured composite coatings immersed in 0.5M K2SO4 recorded by direct potential scan at 0.1V per minute: Curve (1), no sliding applied; Curve (2), continuous sliding at 10 N; 120 rpm; Curve (3), continuous sliding at 15 N, 120 rpm [320].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

286

Subir Kumar Ghosh and Jean-Pierre Celis

Figure 41. Wear-corrosion rate of nano-structured Ni–SiC (20 nm) composite coatings in 0.5M K2SO4 tested at different applied sliding loads for continuous and intermittent sliding (latency time of 20 s) tests [320].

6. MAIN LIMITATIONS OF ACTUAL COMPOSITE PLATING SYSTEMS

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

6.1 Limitations of Electrolytic Co-deposition Process from Aqueous Electrolytes The detailed overview of the co-deposition mechanisms and the effect of several process parameters on the electrolytic co-deposition of second phase particles into a metal matrix given in the previous sections, allows to identify that the particle surface nature (both physical and chemical), particle size and interactions with not only ions, but also water molecules and electrode, as well as the reduction reaction of metal ions are the main governing factors acting on co-deposition steps like the adsorption of particles onto the growing metal surface, the entrapment and embedment of these particles into a metal matrix. With respect to the interaction with water molecules, the hydrophobic or hydrophilic nature of particles determines their incorporation rate, and thus their content in composite coatings. Hydrophilic particles do not incorporate due to the hydration layer that acts as an ionic screen on their surface [171, 187, 322-324]. The hydration layer prevents hydrophilic particles from making a ―real‖ contact with the electrode, and hence particles and electrodes remain separated by a small gap. Recently, it was shown that particles start to ―ride‖ surrounding the growth front formed by metal deposition [322]. Metal ions diffuse into the small gap existing between particles and electrode, and are reduced underneath the particles. This pushes particles up by the moving metal/electrolyte interface, instead of being incorporated in the growing metal deposit. From detailed studies done at MTM-KULeuven (Belgium) on hydrophobic/hydrophilic particle co-deposition aspects [171, 187, 322, 324], it is now quite clear that the electrolytic co-deposition of non-Brownian particles from aqueous solutions is governed to a large extent by the hydration force [98, 325]. By measuring the adhesion force between particle and electrode, it was found that the electrolytic co-deposition is governed by the DLVO interactions plus an additional short-range repulsion. This force was tentatively identified as the hydration force [98]. This would explain why highly hydrophilic materials such as oxides,

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

287

have a small tendency to co-deposit, while hydrophobic materials such as plastics, graphite, etc. do co-deposit readily. In order to avoid particle hydrophilicity, several investigators have tried to co-deposit them either from non-aqueous electrolytes [152, 171, 271] or by modifying the particle surface nature from hydrophilic to hydrophobic. For instance, as discussed in § 3, Terzieva et al. [171] showed that hydrophilic SiO2 particles do not co-deposit from a surfactant-free or a surfactant containing acid copper sulphate solutions. However, they were able to co-deposit SiO2 by decreasing the hydrophilicity of these particles by using cetyl trimethyl ammonium hydrogen sulphate (CTAHS) from aqueous electrolytes.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 42. Fractured cross section of a composite Al–SiO2 coating obtained from a AlCl3–DMSO2 (write in full text!) electrolyte operated at 110 oC, 8 A.dm-2 and containing 53 gl-1 200 nm SiO2 particles [152.].

Figure 43. Concentration (in volume percent) of 200 nm silica particles in Al–SiO2 coatings obtained from a 2:10 AlCl3–DMSO2 (dimethyl sulphoxide) electrolyte operated at 110 oC, at 11 A.dm-2 as a function of the particle concentration in the electrolyte in gl-1. Key: (d) experimental plot; (- - - -) Langmuir isotherm [152].

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

288

Subir Kumar Ghosh and Jean-Pierre Celis

Besides this and considering particle size, it may be stated that a risk of particle agglomeration especially for nano-size particles, dictates the particle distribution, homogeneity, final agglomerate size, and content into composite coatings. In aqueous electrolytes of high ionic strength, particles have a tendency to agglomerate easily due to a compression of the diffuse double layer surrounding particles. This effect is more pronounced for particles of sub-micrometer size (i.e. < 10-6 m) as the shearing forces on the agglomerates created by the agitation of the plating bath, decreases with particle size. As a consequence, the co-deposition of agglomerated particles takes place and the expected mechanical, chemical and/or physical properties are not reached. Secondly, the rate of co-deposition of particles decreases with size. e.g., the volume percent of co-deposited particles in aqueous electrolytes drops from 5 to 15 vol% for micrometer particles down to 0.1 vol% or less for submicrometer particles [152, 322]. An alternative way is to perform co-deposition from non-aqueous electrolytes. That process appeared to be effective for the co-deposition of hydrophilic particles without any agglomeration. Researchers at MTM-KULeuven [152] were able to deposit significant amounts of hydrophobic SiC (mean size 700 nm), TiB2 (mean size 1.9 m), h-BN (mean size 4 m) and hydrophilic SiO2 (mean size 200 nm) and Al2O3 (mean size 50 nm) particles into an aluminum matrix from non-aqueous molten salt electrolytes. Important is to stress that the nano-size particles were deposited as non-agglomerated particles as can be seen from Fig.42. The relationship between particle content and current density revealed a Langmuir adsorption-like behavior as shown in Fig.43. Another approach of composite plating would be to co-deposit from organic baths. Particles like Al2O3 (mean size 1 m), SiC (mean size 1 m), B4C (mean size 1.08 m), diamond (mean size 2-4 m), SiO2 (mean size 0.28 m), WC (mean size 1 m), and TiC (mean size 1 m) can be co-deposited into a nickel matrix [10] using ethanol as solvent. It has to be noted that most hydrophilic particles like SiO2 do easily co-deposit from organic baths. Sometimes a grain size refinement of a nickel matrix electroplated from a non-aqueous electrolyte was reported [326-328].

6.2 Limitations of Electrolytic Co-deposition from Non-Aqueous Electrolytes Even though a few successes in depositing metal-ceramic particle composite coatings from non-aqueous electrolytes, not all combinations of second phase particles and metal matrix can be obtained in that way. The disadvantages of organic non-aqueous or molten salt electrolytic techniques are: a) b) c) d)

The hygroscopic nature of the electrolytes, Molten salt electrolysis requires a high temperature operation, The complicated experimental set-up, The non-availability of suitable molten salts or organic baths for any given metalceramic particle pair, and e) Expensive in comparison to aqueous processes.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

289

It is thus indicated to search for a novel approach of composite plating which e.g. could be the ‗in-situ‘ formation of particles in the diffuse double layer on cathodes and the subsequent co-deposition of such particles that might even be nano-sized ones.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

7. FUTURE TRENDS The co-deposition of particles along with metal ions during electrolytic processes is a complicated matter, given the fact that several parameters like particle characteristics, process parameters, bath constituents and their concentrations, and the hydrodynamic conditions have a significant influence on the amount of particles in the deposits. Recent studies have highlighted that it is the number density of the second phase particle incorporated without agglomeration, is a pre-requisite to enhance mechanical/chemical properties in affecting the coating microstructure, and especially in caging the ―dislocations‖. It also enhances the surface area of coatings covered by particles thereby increasing the catalytic activity of the composite coatings. Therefore, future thrust on composite plating should be focused on achieving the co-deposition of numerous as small as possible (especially nano-size) second phase particles into a metal matrix to impart substantial improvements in functionality. That implies that particle agglomeration and suspension in plating solutions of high ionic strength are the matter to be tackled as a first priority. Despite the many limitations of aqueous electro-co-deposition like hydrophilicity, particle agglomeration tendency, and difficulties in getting suspended particles duly engulfed in the metallic matrix, it is much more easy and versatile than non-aqueous co-deposition. Both problems of hydrophilicity and agglomeration tendency become even more serious in the case of nano-particles. Such commonly experienced problems at lab scale and in industrial plants in aqueous processes, can be solved, as anticipated by Celis et al. a decade ago [42], by adopting a smart electrochemical process in which particles are generated ‗in-situ’ at the electrode surface as a process taking place simultaneously with the reduction of metal ions. In the electrical double layer in front of the electrode (cathode for electrolytic process), the particle chemical constituents come close to each other and with the help of a reducing environment, they can start to react and to form nascent molecules or particles. Before being incorporated into the growing metal matrix, these single molecules or particles may form small agglomerates or may remain as single entities, and get finally embedded into the growing metal matrix. This process would not only avert the problem of significant agglomeration but also would allow to control and to enlarge the number density of small sized, even sub-micron-sized particles, inside the metal matrix. Recently, a joint research project funded by the CEC (Contract Marie Curie IIF: 220002) between MTM-KULeuven (Belgium) and Bhabha Atomic Research Centre (India) has been launched to incorporate in that way MoS2 nano-particles into a metal matrix to obtain selflubricating nano-composite coatings. The research is focused on the growth of composite coatings from a single electrolyte containing metal ions and basic constituents for the formation of MoS2. In order to meet future industrial needs on novel and performant coatings, there is a strong impetus to develop such ‗in-situ’ co-deposition processes resulting in a nanostructuring of the metal matrix and the incorporation of nano-sized particles.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

290

Subir Kumar Ghosh and Jean-Pierre Celis

8. SUMMARY

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

In summary, the following points are highlighted from an in-depth literature survey on nano-composite plating: a) Uniform distribution of particles without or with a minimum agglomeration in a metal matrix is a prerequisite to obtain desired properties, b) Henceforth, it will be much more appropriate to consider the ‗number density‘ rather than classical vol% or wt% of particles in composite coatings in particular in the case of nano-composites. Superior functional properties can be achieved even with a low vol% co-deposited particles, c) The use of reliable surfactants that don‘t show any short-term degradation is a must to minimize agglomeration of nano-particles since agglomerated particles in composite coatings badly affect the functional properties, d) Pulse and reverse pulse electrolysis (PRC) are a good option to incorporate particles of smaller size at a higher efficiency, e) The lack of knowledge on the particle entrapment/engulfment into an ever-fresh metal layer and also on particle-electrode interactions indirectly imposes a large process control. Future theoretical models should include bath chemistry, and electro-crystallization phenomena in the description of the whole electrodeposition and electroless process, f) Related to properties of electroplated composite coatings, most effort was paid to hardness, tribology, and corrosion, while up to now little attention was given to ‗tribo-corrosion‘ even though composite coatings are being used under aggressive mechanical and chemical environments. So, intense investigations of the tribocorrosion properties of composite coatings are required, g) Because of the limitations of composite plating from aqueous solutions due to agglomeration, and the limited availability of non-aqueous solvents, research should be directed in the future on the ‗in-situ’ formation of particles at an electrode surface and the subsequent incorporation of such particles into a metal matrix.

ACKNOWLEDGMENTS S. K. Ghosh thanks the European Commission for a Marie Curie International Incoming Fellowship (Contract PJ_REF:220002) funded under FP7 Framework Program. Part of this work was done within the Scientific Research Community on Surface Modification of Materials funded by the Flemish Science Foundation (contract WOG -FWO-Vl), as well as COST 533 on Biotribology and the Network of Excellence on Complex Metallic Alloys (NoE-CMA Contract No. NMP3-GT-2005-500140), both funded by the European Commission under FP6 Framework Program.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

291

REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]

Fink, C. G.; Prince, J. D. Trans. Am. Electrochem. Soc., 1928, 54, 315. Sautter, F. K. J. Electrochem. Soc., 1963, 110 (6), 557-560. Busson, M.; Perinaud, M. Proc. Colloque Innovationa Technologiques dans les Traitements de Surfaces, Paris, Dec. 9-10, 1981. Feldstein, N.; Lancsek, T.; Lindsay, D.; Salerno, L. Proc. 1st AES Electroless Plat. Symp., St. Luis, Mo, March 23-24, 1982. Helle, K.; Opschoor, A.; Proc. Interfinish ’80, Kyoto, 1980, 234-236. Li, Y.; Wang, R.; Qi, F.; Wang, C. Appl. Surf. Sci., 254, 2008, 4708-4715. Zhao, Q.; Liu, Y.; Muller-Steinhagen, H.; Liu, G. Surf. Coat. Technol., 155, 2002, 279284. Zhao, Q.; Liu, Y.; Wang, C.; Wang, S.; Muller-Steinhagen, H. Chem. Engg. Sci., 60, 2005, 4858-4865. Sean Brahim, S.; Colbern, S.; Robert, G.; Moser, A.; Grigorian, L. Nanotechnol. 20, 2009, 235502-8. Tsai, Li-N.; Shen, G-R.; Cheng, Y-T.; Hsu, W. J. Microelectromecha. Sys., 15, 2006, 149-158. Erb, U.; El-Sherik, A. M.; Palumbo, G.; Aust., K. T. Nanostruct. Mater., 2, 1993, 383390. Webb, P.R.; Robertson, N. L.; J. Electrochem. Soc., 141(3), 1994, 669-673. Celis, J. P.; Roos, J.R.; Van Vooren, W.; Vanhumbeeck, J. Trans. Inst. Met. Finish, 67, 1989, 70-72. Buelens, C.; Celis, J.P.; Roos, J. R. Trans. Inst. Met. Finish., 63(1), 1985, 6-10. Buelens, C.; Celis, J.P.; Roos, J. R. J. Appl. Electrochem., 13, 1983, 541-548. De Bonte, M.; Celis, J. P.; Roos, J. R.; Vanhumbeeck, J.; Precious Metals-Modern Technologies and Aplications, November 28-29, 1991, 81. Roos, J.R.; Celis, J.P.; Van Vooren, W.; Buelens, C. Proc. Of the Eleventh Intrnal. Precious Metals Institute Conf., Brussels, Belgium, 1987, 93-107. R. De Donker, G. Ide, J. Vanhumbeeck, J. P. Celis and J. R. Roos, Symposiym on ―Advances in Surface Treatment Technology‖, April, 8-10, 1986. Celis, J.P.; Roos, J.R.; Van Vooren, W. AESF 3rd Intl.PulsePlat. Symp., October 28-29, 1986. Celis, J.P.; Roos, J.R.; Buelens, C. J. Electrochem. Soc., 134(6), 1987, 1402-1408. Lozano-Morales, A.; Fitzgerald, J.; V. Singh, V.; Xie, X.; Podlaha, E.J. J. Electrochem. Soc., 153 (8), 2006, C567-C574. Romankiw, L.T. Electrochimica Acta, 42, 1997, 2985-3005. Datta, M.; Landolt, D. Electrochemica Acta, 45, 2535 (2000, 2535-2558. Romankiw, L.T. Plat. Surf. Finish., 84, 1997, 10-15. Ali, H.O.; Christie, I. R. A., Gold Bull., 17(4) 1984, 18-127. Celis, J.P.; Roos, J. R.; Buelens, C.; Fransaer, J. Trans. Inst. Met. Finish., 69, 1991, 133-139. Fransaer, J.; Celis, J.P.; Roos, J. R.; Met. Finish., 91, 1993, 97-100. Celis, J.P.; De Bonte, M.; Roos, J. R. Trans. Inst. Met. Finish. 72, 1994, 89-93. Celis, J.P.;Fransaer, J. Trans. Inst. Met. Finish. 75, 1997, 118-121.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

292

Subir Kumar Ghosh and Jean-Pierre Celis

[30] Buelens, C.; Fransaer, J.; Celis, J.P.; Roos, J. R. Bull. Electrochem., 8, 1992, 371-375. [31] Hovestad, A.; Janssen, L. J. J. J. Appl. Electrochem., 25, 1995, 519-527. [32] Hovestad, A.; Janssen, L. J. J in Modern Aspects of Electrochemistry, edited by B.E. Conway et al, Kluwer Academic/Plenum Publishers, NY, 2005, 38, 475-532. [33] Low, C.T.J.; Wills, R.G.A.; Walsh, F.C. Surf. Coat. Technol., 201, 2006, 371-383. [34] Moris, D. G. Materials Science Foundation (monograph Series), Trans Tech publ., ISBN 087849, Vol 2, 1998. [35] Dieter, G.E. Mechanical metallurgy (3rd ed.), McGraw-Hill, NY, 1986, 212–20. [36] Zhang, Q.; Chen, D.L. Scr. Mater., 51, 2004, 863–867 [37] Meyers, M.A.; Chawla, K.K. Mechanical behaviour of materials, Prentice Hall, Saddle River (NJ), 1999, 492–4. [38] Zhang, Q.; Chen, D.L. Scr. Mater., 54, 2006, 1321-26. [39] Malone, G. A. Plat. Surf. Finish., 8, 1991, 58-62. [40] Steinhauser, S.; Wielage, B.; Zschunke, A.; Jakob, C.; Erler, F.; Nutsch, R. Galvanotechnik, 92, 2001, 78-84. [41] Greco, V. P. Plat. Surf. Finish., 7, 1989, 62-67. [42] Fransaer, J.; Celis, J. P. DECHEMA/DGO-Symposium, Franfurt/Main, Germany, 11/1999, 49. [43] Moris, D. G.; Morris, M.A. Acta Metall. Mater., 39, 1991, 1763-1772. [44] Lu, K. Nanostructured Mater., 2, 1993, 643-51. [45] Hall, E.O. Proc. Phys. Soc., B64, 1951, 747-58. [46] Petch, N. J. J. Iron Steel Inst., 174, 1953, 25-34. [47] Volp, T.; Gring, E.; Kuschke, W.M.; Artz, E. Nanostructured Mater., 8, 1997, 855-65. [48] Nieh, T.G.; Wadsworth, J.; Scr. Metall. Mater., 25, 1991, 955-58. [49] Choksi, A.H.; Rosen, A.; Karch, J.; Gleiter, H. Scr. Metall., 23, 1989, 1679-83. [50] Nieman, G. W.; Weertman, J. R.; Siegel, R. W. J. Mater. Res., 6, 1991, 1012-1027. [51] Fougere, G. E.; Weertman, J. R.; Siegel, R. W.; Kim, S. Scr. Metall. Mater., 26, 1992, 1879-1884. [52] El-Sherik, A. M.; Erb, U.; Palumbo G.; Aust, K. T. Scr. Mater., 27, 1992, 1185-1188. [53] Gertsman, V. Y.; Hoffmann, M.; Gleiter, H; Dirringer, R. Acta. Mater., 42, 1994, 35393544. [54] Sanders, P. G.; Eastman, J A., Weertman, J. R. Acta Mater., 10, 1997, 4019-4025. [55] Schuh, C. A.; Nieh, T. G., Yamasaki, T. Scr. Mater., 46, 2002, 735-740. [56] Scattergood, R.O.; Koch, C.C. Scr. Metall. Mater. 27, 1992, 1195-1200. [57] Yamakov, V.; Wolf, D.; Phillpot, S.R.; Mukherjee, A.K. Nat. Mater., 1, 2002, 1-4. [58] Schiotz, J.; Jacobsen, K.; Science, 301, 2003, 1357-1359. [59] Masamura, R.A.; Hazzledine, P.M.; Liaw, P.K.; Lavernia, E.J. Acta Metall., 13, 1998, 4527-4534. [60] Conrad, H.; Narayan, J. Scr. Mater., 42, 2000, 1025-1030. [61] Van Swygenhoven, H.; Spaczer, M.; Caro, A. Acta Mater., 47,1999, 3117-3126. [62] Takeuchi, S. Scr. Mater., 44, 2001, 1483-1487. [63] Yamakov, V.; Wolf, D.; Salazar, M.; Phillpot, S.R.; Gleiter, H. Philos. Mag. Lett. 83, 2003, 385-393. [64] Yamakov, V.; Wolf, D.; Phillpot, S.R.; Mukherjee, A.K.; Gleiter, H. Nat. Mater., 3, 2004, 43-47. [65] Lu, K.; Sui, M.L.; Scr. Mater., 28, 1993, 1465-1470.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98]

293

Wang, N.; Wang, Z.; Aust, K. T.; Erb, U. Acta Metall. Mater., 43, 1995, 519–523. Konstantinidis, D.A.; Aifantis, E.C. Nanostruct Mater., 10, 1998, 1111-1118. Zhang, S.; Sun, D.; Fu, Y.; Du, H. Surf. Coat. Technol., 198, 2005, 2-8. Musil, J.; Baroch, P.; Vlcek, J.; Nam, K.H.; Han, J.G.; Thin Solid Films, 475, 2005, 208-218. Gusev, A. I. Usp. Fiz. Nauk, 168, 1998, 55-83. Veprek, S.; Veprek-Heijman, M. G. J.; Karvankova, P.; Prochazka, J. Thin Solid Films, 476, 2005, 1-29. Veprek, S. J. Vac. Sci. Technol. A, 17, 1999, 2401-2420. Van Swygenhoven, H. Science, 296, 2002, 66-67. Schiotz, J.; Di Tolla, F. D.; Jacobsen, K. W. Nature, 39, 1998, 561-563. Fischer-Cripps, A. C.; Karvankova, P.; Veprek, S. Surf. Coat. Technol.; 200, 2006, 5645-5654. Sundgren, J. E.; Hentzell, H. T. G.; J. Vac. Sci. Technol. A, 4, 1986, 2259-2279. Loffler, F. H. W. Surf. Coat. Technol., 68-69, 1994, 729-740. Kim, G.; Kim, B.; Lee, S.; Hahn, J. Surf. Coat. Technol., 200, 2005, 1669-1675. Paul, H. M.; Anders, H.; Lennart, K.; Jacob, S.; Tommy, L.; Christian, M.; Lars, H. Appl. Phys. Lett., 83, 2003, 2049-2051. Wen, J. G.; Mayrhofer, P. H.; Mitterer, C.; Greene, J. E.; Petrov, I. Microsc. Microanal., 12, 2006, 720-721. Mattox, D. M. J. Vac. Sci. Technol. A, 7, 1989, 1105-1114. Musil, J.; Kadlec, S.; Vyskocil, J.; Valvoda, V. Thin Solid Films, 167, 1988, 107-120. Mayrhofer, P. H.; Kunc, F.; Musil, J.; Mitterer, C. Thin Solid Films, 415, 2002, 151159. Zhang, S.; Wang, H. L.; Ong, S.; Sun, D.; Bui, X. L. Plasma Proces. Polymers, 4, 2007, 219-228. Bhagwat, M.J.; Celis, J.P.; Roos, J.R. Trans. IMF, 6, 1983, 72-79. Lyklema, J. ―Fundamentals of Interface and Colloid Science‖, NY and London, Academic Press, Vol.2, 1995. Rector, D. R.; Bunker, B. C.; Sedimentation Models, PNL-10754, September 1995. Erler, F.; Jakob, C.; Romanus, H.; Spiess, L.; Wielage, B.; Lampke, T.; Steinhauser, S. Electrochim. Acta 48, 2003, 3063-/3070. Landau, L.D.; Lifshitz, E.M.; Fluid Mechanics, Elsevier, 1987. Xiang, Y.; Zhang, J.; Jin, C. Plat. Surf. Finish., 21, 2001, 64. Chen, Y.C.; Ger, M.-D.; Hwu, W.H.; Kuo, S.L.; Hou, K.H. J. Chin. Colloid Interface Soc., 24, 2002, 89-95. Saffman, P.G.; Turner, T.S. J. Fluid Mech., 1, 1956, 16 – 30. Strenge, K.; Mühle, K. Coagulation and Flocculation, Ed. by B. Dobiáš, Surfactant Science Series 47, Marcel Dekker Inc., NY, 1993, pp. 265-320, 355-390. Probstein, R.F. Physicochemical Hydrodynamics, ed., J. Wiley & Sons, New York, 1994, pp.237-256, 277-302. Guglielmi, N. J. Electrochem. Soc. 119, 1972, 1009-1012. Celis, J. P.; Roos, J.R. J. Electrochem. Soc. 124, 1977, 1508-1511. Valdes, J.L. J. Electrochem. Soc. 134, 1987, 223C-225C. Fransaer, J.; Celis, J. P.; Roos, J. R. J. Electrochem. Soc., 139, 1992, 413-425.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

294

Subir Kumar Ghosh and Jean-Pierre Celis

Fransaer, J. Study of the behaviour of particles in the vicinity of electrodes, PhD thesis, Catholic University of Leuven (1994). [100] Hwang, J.B.; Hwang, C.S., J. Electrochem. Soc. 140, 1993, 979-987. [101] Wang, D.L.; Li, J.; Dai, CH. S.; Hu, X.G. J. Appl. Electrochem., 29, 1999, 437-444. [102] Vereecken, P.M., Shao, I.; Searson, P.C. J. Electrochem. Soc., 147, 2002, 2572-2575. [103] Bercot, P.; Pena-Munoz, E.; Pagetti, J. Surf. Coat. Technol. 157, 2002, 282-289. [104] Stappers, L.; Fransaer, J. J. Electrochem. Soc., 152, 2005, C-392-C398. [105] Hassan, S.F.; Gupta, M. Mater. Sci. Technol. 20, 2004, 1383–1388. [106] Broszeit, E.; Heinke, C.; Weigand, H. Metallurgica, 25, 1971, 470. [107] Verelst, M.; Bonino, J.P.; Brieu, M.; Rousset, A. Mater. Sci. Eng. A 91, 1995, 165-169. [108] Ruimi M.; Martinou, R. Proc. Interfinish ’88, Paris, 1998, p. 129. [109] Shrestha, N.K.; Masuko, M.; Saji, T. Wear, 254, 2003, 555-564. [110] Thoma, M. Leeds Lyon symposium, 1989 Session 1. [111] Kumar, K.; Chandramohan, R.; Kalyanaraman, D. Appl. Surf. Sci., 227, 2004, 383-386 [112] Dellacorte, C.; Edmonds, B. J. US Patent 5866518, 1999. [113] Wu, Y.; Gao, J.; Shen, B.; Liu, L.; Hu, W.; Ding, W. Plat. Surf. Finish. 92, 2005, 3842. [114] Liu, Y.; Zhao, Q. Plat. Surf. Finish., 91 (2004) 48-51. [115] Nishira, M.; Takano, O. Plat. Surf. Finish., 81, 1994, 48-50. [99]

[116] Shi, L.; Sun, C.; Liu, W. Appl. Surf. Sci., 254, 2008, 6880-6888.

[117] Zou, T.Z.; Tu, J. P.; Zhang, S.C.; Chen, L.M.; Wang, Q.; Zhang, L.L.; He, D.N. Mater. Sci. Eng. A, 426 (2006) 162-168. [118] Franser, J.; Roos, J.R.; Celis, J.P. Metal. Finish., 87, 1989, 107-109. [119] Zhu, L.; Feng, L.; He, Y.; Plat. Surf. Finish. 92, 2005, 49-53. [120] Foster, J. Trans. Inst. Metal. Finish., 63, 1985, 115-119.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[121] Malfatti, C.F.; Zoppas Ferreira, J.; Santos, C.B.; Souza, B.V.; Fallavena, E.P.; Vaillant, S.; Bonino, J. –P. Corr. Sci., 47, 2005, 567-580.

[122] Song, Y.W.; Shan, D.Y.; Chen, R.S.; Han, E.H. Surf. Eng., 23, 2007, 334-338. [123] Kazumi, N.; Yasuhika, M.; Hada, T. Proc. Symp. Corrosion Protection by Organic Coatings (Ed), M. W. Kending and H. Leidheiser Jr., 87(2) (1987) 140-152. [124] Kimoto, M.; Yakawa, A.; Tsuda, T.; Kammel, R. Metall. Wiss Tech. 44 (1990) 1148. [125] Abdel, A.Aal; Hassan, B.H.; Abdel Rahim, M.A. J. Electroanal. Chem., 619-20, 2008, 17-25. [126] Gierlotka, D.; E. Rowinski, A. Budniok and E. La¸Giewka, J. Appl. Electrochem., 27 (12) (1997) 1349-1354. [127] Nishikawa, J.; Nakamura, K.; Asadullah, M., Miyazawa, T.; Kunimori, K.; Tomishige, K.; Catalysis Today, 29 (2008) 146-155 [128] Fajardo, H.V.; Probst, L.F.D.; Carreño, N.L.V; Garcia, I.T.S.; Valentini, A. Catalysis Letters, 119, 2007, 228-236. [129] Shoeib, M.; Maamoun, M.A. Tribotest, 9, 2002, 49-56 [130] Verslet, M; Bonino, J. –P.; Rousset, A. Mater. Sci. Eng. A, 135 (1991) 51-57. [131] Ramesh Bapu, G. N. K.; Yusuf, M. M. Mater. Chem. Phys. 36, 1993, 134-138. [132] Suzuki, Y.; Asai, O. J. Electrochem. Soc., 134, 1987, 1905-1910. [133] Berkh, O.; Eskin, E.; Berner, S.; Zahavi, J. Plat. Surf. Finish., 82, 1995, 54-59. [134] Bozzini, B.; Giovanelli, G.; Cavallotti, P.L. J. Microscopy, 185, 1996, 283-291.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Plating of Nano-composites- Overview and Trends

295

[135] Bozzini, B.; Giovannelli, G.; Nobili, L.; Cavallotti, P.L. AIFM Galvanotecnica e Nuove Finiture, 5, 1995, 92. [136] Garcia, I.; Franser, J.; Celis, J. P. Surf. Coat. Technol., 148, 2001, 171-178 [137] Berkh, O.; Bodnevas, A.; Zahavi, J. Plat. Surf. Finish. 82, 1995, 62-66. [138] Pavlatou, E.; Stroumbouli, M.; Gyftou, P.; Spyrellis, N. J. Appl. Electrochem. 36, 2006, 385-394. [139] Maurin, G.; Lavanant, A. J. Appl. Electrochem., 25, 1995, 1113-1121. [140] Kim, S.K.; Yoo, H.J. Surf. Coat. Technol., 108-109, 1998, 564-569. [141] Shao, I.; Vereecken, P.M.; Cammarata, R.C.; Searson, P.C. J. Electrochem. Soc., 149, 2002, C610-C614. [142] Shao, I.; Vereecken, P.M.; Chien, C.L.; Searson, P.C; Cammarata, R.C. J. Mater. Res., 17, 2002, 1412-1418. [143] Lozano-Morales, A.; Podlaha, E.J. J. Electrochem. Soc., 151 (7) (2004) C478-C483. [144] Koch, C.C. Nanostructured Materials: Processes, Properties and Applications, Noyes Publications, NY, 2002. [145] Abdel Aal, A.; El-Sheikh, S.M.; Ahmed, Y.M.Z., Mater. Res. Bull., 44, 2009, 151–159. [146] Balaraju, J.N.; Millath Jahan, S.; Rajam, K.S. Surf. Coat. Technol., 201, 2006, 507-512. [147] Ramesh Bapu, G. K. N. Surf. Coat. Technol., 67 (1994) 105-110. [148] Ramesh Bapu, G. N. K.; Mohan, S. Plat. Surf. Finish., 82, 1995, 86-88. [149] Afshar, A.; Ghorbani, M.; Mazaheri, M. Surf. Coat. Technol., 187, 2004, 293-299. [150] Dedeloudis, C.; Kaisheva, M.K. ; Muleshkov, N.; Muleshkov, T.; Nowak, P.; Franser, J.; Celis, J. P. Plat. Surf. Finish., 86 (8), 1999, 57-60. [151] Wang, S.; Wei, W-C. Mater. Chem. Phys. 78, 2003, 574-580. [152] Fransaer, J.; Leunis, E., Hirato, T.; Celis, J.P. J. Appl. Electrochem., 32, 2002, 123-128. [153] Wu, G.; Li, N.; Zhou, D.; Mitsuo, K. Surf. Coat. Technol., 176, 2004, 157-164. [154] Roos, J. R.; Celis, J. P.; Franser J., Buelens, C. J. Met., 42, 1990, 60-63. [155] Lee, H-K.; Lee, H. Y.; Jeon, J. M. Surf. Coat. Technol., 201, 2007, 4711-4717. [156] Meguno, M.; Ushida, T.; Hiraoko, T.; Esumi, K. Bull. Chem. Soc. Jpn., 60, 1987, 8994. [157] Tomaszewski, T.W.; Clauss, R.J.; Brown, H. Proc.Am. Electroplaters Soc., 50, 1963, 169-174. [158] Greco, V.P. Bladauf, W. Plating, 55, 1968, 250-257. [159] Celis, J. P. Electrolytische deposite van koper-aluminiumoxyde deklagen, Phd-thesis, Katholieke Universiteit Leuven, (1976), [160] Brandes, E.A.; Goldthrope, D. Metallurgica, 76, 1967, 195-198. [161] Takahashi, A.; Miyoshi, Y.; Hada, T. J. Electrochhem. Soc., 141, 1994, 954-957. [162] Kariapper, A. M. J.; Foster, J. Trans. IMF, 52, 1974, 87-91. [163] Vidrich, G.; Castagnet, J-F.; Ferkel, H. J. Electrochem. Soc., 152, 2005, C294-C297. [164] Kuo, S.L.; Chen, Y.C.; Ger, M. D.; Hwu, W. H. Mater. Chem. Phys., 86, 2004, 5–10 [165] Roos, J. R.; Celis, J. P.; Helsen, J. Trans. IMF, 55, 1977, 113-117. [166] Sinha, P.K.; Dhananjayan, N.; Chakrabarti, H. K. Plating, 60, 1973, 55-59. [167] Chang, Y.; Chang, Y.; Lin, C. –I. Electrochim. Acta, 43, 1998, 315-324. [168] Wu, G.; Li, N.; Wang, D. L.; Zhou, D. R.; Xu, B. Q.; Mitsuo, K. Mater. Chem. Phys. 87, 2004, 411–419. [169] Rudnik, E. Appl. Surf. Sci., 255, 2008, 2613–2618. [170] Bund, A.; Thiemig, D. J. Appl. Electrochem., 37, 2007, 345–351

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

296

Subir Kumar Ghosh and Jean-Pierre Celis

[171] Aslanidis, D.; Fransaer, J.; Celis, J. P. J. Electrochem. Soc., 144(7), 1997, 2352-2357. [172] Kalantary, M.R.; Holbrook, K.A.; Wells, P. B. Trans. Inst. Metal. Finish., 71(2), 1993, 55-61. [173] Necula, B. S.; Apachitei, I.; Fratila-Apachitei, L. E.; Teodosiu, C.; Duszczyk, J.; J. Colloid Interface Sci, 314(2), 2007, 514-22. [174] White, C.; Foster, J. Trans. IMF, 56, 1978, 92-96. [175] Tomaszewski, T.W.; Tomaszewski, L.C.; Brown, H. Plating 56, 1969, 1234–1239. [176] Lee, C. C.; Wan, C. C.; J. Electrochem. Soc., 135, 1988, 1930-1933. [177] Yin, K. –M.; Hong, H. J. Trans. Inst. Metal. Finish., 75, 1997, 35-38. [178] Shrestha, N. K.; Hamal, S. B.; Saji, T. Surf. Coat. Technol., 183, 2004, 247-253. [179] Yang, X.; Li, K.; Peng, X.; Wang, F. Trans. Nonferro. Metal Soc. 19, 2009, 119-124. [180] Wu, G.; Li, N.; Zhou, D.; Mitsuo, K. Surf. Coat. Technol. 176, 2004, 157–164. [181] Wang, L.; Gao, Y.; Liu, H.; Xue, Q.; Xu, T. Surf. Coat. Technol., 191, 2005, 1-6.

[182] Vidrine, A. B.; Podlaha, E. J. J. Appl. Electrochem. 31, 2001, 461-468. [183] Helle, K. Report AKZO Research, Arnhem, 1993. [184] Hu, X.; Dai, C.; Li, J.; Wang, D. Plat. Surf. Finish. 84, 1997, 51-53. [185] Kinoshita, H.; Yonezawa, S.; Kim, J.; Kawai, M.; Takashima, M.; Tsukatani, T. J. Fluorine Chem., 129, 2008, 416-423.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[186] Shrestha, N. K.; Kawai, M.; Saji, T. Surf. Coat. Technol. 21, 2005, 2414-2419.

[187] Terzieva, V.; Fransaer, J.; Celis, J.-P. J. Electrochem. Soc. 147 (2000) 198–202. [188] Shrestha, N. K.; Miwa, I.; Saji, T. J. Electrochem. Soc., 48, 2001, C106-C109. [189] Shrestha, N. K.; Takebe, T.; Saji, T. Diamond Relat. Mater., 15, 2006, 1570–1575 [190] Ohashi, K. Jp. Patent, JP H5-39360, 1993. [191] Oba, K. Jp. Patent, JP H7-224178, 199). [192] Iwata, T. Jp. Patent, JP H6-2156, 1994. [193] Tulsi, S.S. Trans IMF, 61, 1978, 147-149. [194] Zhao, Q.; Liu, Y. Surf. Coat. Technol, 200, 2005, 2510-2514. [195] Matsuda, H.; Nishira, M.; Kiyono, Y.; Takano, O. Trans. Inst. Metal Finish., 73, 1995, 16-18. [196] Henuset, Y. M.; Menini, R. Plat. Surf. Finish., 89, 2002, 63-67. [197] Alexandridou, S.; Kiparissides, C.; Fransaer, J.; Celis, J. P. Surf. Coat. Technol., 71, 1995, 267-276. [198] Hovestad, A.; Heesen, R. J. C. H.; Janssen, L. J. J. J. Appl. Electrochem. 29, 1999, 331338. [199] Ger, M-D. Mater. Chem. Phys., 87, 2004, 67-74 [200] Friedman, H.; Eidelman, O.; Feldman, Y.; Moshkovich, A.; Perfiliev, V.; Rapoport, L.; Cohen, H.; Yoffe, A.; Tenne, R. Nanotechnology, 18, 2007, 115703-8. [201] Katz, A.; Redlich, M.; Rapoport, L.; Wagner, H. D.; Tenne, R. Tribol. Letters, 21, 2006, 135-139. [202] Redlich, M.; Katz, A.; Rapoport, L.; Wagner, H. D.; Feldman, Y.; Tenne, R. Dental Mater., 24, 2008, 1640-46. [203] Guo, C.; Zuo, Y.; Zhao, X.; Zhao, J.; Xiong, J. Surf. Coat. Technol., 202, 2008, 33853390. [204] Xu, H.; Yang, Z.; Li, M. K.; Shi, Y. L.; Huang, Y.; Li, H. L. Surf. Coat. Technol., 191, 2005, 161– 65.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

297

[205] Zou, T.; Tu, J.; Huang, H.; Lai, D.; Zhang, L.; He, D. Adv. Eng. Mater., 8, 2006, 289293. [206] Wang, L. M. J. Appl. Electrochem., 38, 2008, 245–249 [207] Hazan, Y.; Werner, D.; Z‘graggen, M.; Groteklaes, M.; Graule, T. J. Coll. Interface. Sci. 328, 2008, 103-109. [208] Shi, L.; Sun, C.; Gao, P.; Zhou, F.; Liu, W. Appl. Surf. Sci. 252 (2006) 3591-3599. [209] Ramesh Bapu, G.K.N. Plat. Surf. Finish., 67, 1994, 105-110. [210] Ramesh Bapu, G.K.N. Plat. Surf. Finish., 82, 1995, 70-75. [211] Tian, B. R.; Cheng, Y. F. Electrochim. Acta, 53, 2007, 511-517. [212] Xu, B.; Wang, H-d.; Dong, S-y.; Jiang, B.; Tu, W-y.; Electrochem. Commu. 7, 2005, 572575. [213] Xu, B.S.; Dong, S.Y.; Liang, Z. J.; Hu, Z.F.; Jiang, B.; Wang, H. M. Patent, CN 02101196.6. [214] B.S. Xu, W. Hai-dou, D. Shi-yun, J. Bin, T. Wei-yi, Electrochem. Comm. 7 (2005) 572575. [215] Okumiya, M.; Takeuchi, H.; Tsunekawa, Y. J. Jpn. Inst. Met. 59, 1995, 640. [216] Chin, D.-T.; Hsueh, K.-L. Electrochim. Acta 31, 1986, 561-564. [217] Takeuchi, H.; Tsunekawa, Y.; Okumiya, M. Mater. Trans., 38, 1997, 43-46. [218] Thiemig, D.; Bund, A.; Talbot, J. B. Electrochim. Acta ., 54 2009, 2491–2498 [219] Gan, Y.; Lee, D.; Chen, X.; Kysar, J. W. J. Eng. Mater. Technol. 127 (4), 2005, 451– 456. [220] Martins, L.; Martins, J. I.; Romeira, A. S.; Costa, M.E.; Costa, J.; Bazzaoui, M.; Mater. Sci. Forum, 455–456, 2004, 844–848.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[221] Xia, Fa-f.; Wu, M-h.; Wang, F.; Jia, Z-y.; Wang, Ai-l. Current Appl. Phys. 9, 2009, 44-47.

[222] Roos, J. R.; Celis, J. P.; Kelchtermans, H.; Van Camp, M.; Buelens, C. Proceedings of 10th World Congress on Metal Finishing, (edited by S. Hawyama) The metal Finishing Society of Japan, Kyoto, 1980, p. 203 [223] Stojak, J. L.; Talbot, J. B. J. Electrochem. Soc., 146, 1999, 4504-4513. [224] Panda, A.; Podlaha, E. J. Electrochem. Solid-State Lett. 6, 2003, C149-C152. [225] Palumbo, G.; Gonzalez, F.; Tomantschger, K.; Erb, U.; Aust, K. T. Plat. Surf. Fin. 90, 2003, 36-45. [226] Ghosh, S.K.; Grover, A.K.; Dey, G.K.; Totlani, M.K. Surf. Coat. Technol., 126, 2000, 48-63. [227] Ghosh, S.K.; Dey, G.K.; Kulkarni, U.D.; Grover, A.K.; Dusane, R.O.; Suri, A.K.; Banerjee, S. J. Mater. Res., 21, 2006, 45-61. [228] Hovestad, A.; Ansink, R.; Janssen, L. J. J. J. Appl. Electrochem., 27, 1997, 756-761. [229] Yeh, S. H.; Wan, C. C. Mat. Sci. Technol,. 11, 1995, 589-593. [230] Stankovic, V. D.; Gojo, M. Surf. Coat. Technol. 81, 1996, 225-232. [231] Foster, J.; Cameron, B. P.; Carew, J. A. Trans. Inst. Met. Finish. 63, 1985, 115-119. [232] Spanou, S.; Pavlatou, E.A.; Spyrellis, N. Electrochem. Acta, 54, 2009, 2547-2555. [233] Guo, C.; Zuo, Y.; Zhao, X.; Zhao, J.; Xiong, J. Surf. Coat. Technol. 202, 2008, 33853390. [234] Chou, M.; Ger, M –D.; Ke, S.; Huang, Y.; Wu, S. Mater. Chem. Phys., 92, 2005, 146151. [235] Banovic, S.W.; Barmak, K.; Marder, A. R. J. Mater. Sci., 34, 1999, 3203-3211. [236] Podlaha, E. J.; Landolt, D. J. Electrochem. Soc. 144, 1997, L200-L202.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

298

Subir Kumar Ghosh and Jean-Pierre Celis

[237] Podlaha, E. J. Nano Lett. 1, 2001, 413-416. [238] Stroumbouli, M.; Gyftou, P.; Pavlatou, E.A.; Spyrellis, N. Surf. Coat. Technol. 195, 2005, 325-332. [239] Bahrololoom, M. E.; Sani, R. Surf. Coat. Technol., 192, 2005, 154-163. [240] Chen, L.; Wang, L.; Zeng, Z.; Xu, T. Surf. Coat. Technol., 201, 2006, 599-605. [241] Celis, J. P.; Roos, J. R.; Van Vooren, W.; Vanhumbeeck, J. Oberflache Surfcae, 6, 1987, 16-21. [242] Thiemig, D.; Lange, R.; Bund, A. Electrochim. Acta, 52, 2007, 7362-7371. [243] Wang, W., Hou, F.Y.; Wang, H.; Guo, H-T. Scrpita. Mater. 53, 2005, 613-618. [244] Qu, N. S.; Chan, K. C.; Zhu, D. Scr. Mater. 50, 2004, 1131-1134.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[245] Shrestha, N. K.; Sakurada, K.; Masuko, M.; Saji, T. Surf. Coat. Technol., 140, 2001, 175181.

[246] Kruger, H.G.; Knote, A.; Schindler, U.; Kern, H.; Boccaccini, A. R. J. Mater. Sci. 39, 2004, 839-844. [247] M. Matsuda, Y. Kiyono, M. Nishira and O. Takano, Trans. Inst. Metal Finish., 72, 1994, 55-57., [248] Tang, J.; Zhou, S.; Xie, Y. Plat. Surf. Finish., 85, 1998, 241-244. [249] de Hazan, Y.; Werner, D.; Z‘graggen, M.; Groteklaes, M.; Graule, T. J. Collo. Inter. Sci., 323, 2008, 103-109. [250] de Hazan, Y.; Reuter, T.; Werner, D.; Clasen, R.; Graule, T. J. Collo. Inter. Sci., 328, 2008, 293-300. [251] Rudnik, E.; Kokoszka, K.; Lapsa, J. Surf. Coat. Technol. 202, 2008, 2584-2590. [252] Sarret, M.; Muller, C.; Amell, A. Surf. Coat. Technol., 201, 2006, 389-395. [253] Balaraju, J. N.; Seshadri, S.K. Trans. Inst. Metal. Finish, 77, 1999, 84-86. [254] Ebrahimian-Hosseinabadi, M.; Azari-Dorcheh, K.; Moonir Vaghefi, S.M. Wear, 260, 2006 123-127, [255] Ge, J-P.; Che, R. X.; Wang, X. Z. Plat. Surf. Finish. 85, 1998, 69-73, [256] Xiang, Y.; Zhang, J.; Jin, C. Plat. Surf. Finish., 88, 2001, 64-67. [257] Bin-Shi, X.; Hai-dou, W.; Shi-yun, D.; Bin, J.; Wei-yi, T. Electrochem. Commu., 7, 2005, 572-575. [258] Ramesh Babu, G.K.N.; Muralidharan, V.S.; Vasu, K. I. Plat. Surf. Finish., 78, 1991, 126-132. [259] Izzard, M.; Dennis, J. K. Trans. Inst. Metal Finish., 65, 1987, 85-89. [260] Chen, L.; Wang, L.; Zeng, Z.; Zhang, J. Mater. Sci. Eng. A, 434, 2006, 319-325. [261] Novakovic, J.; Vassiliou, P.; Samara, Kl.; Argyropoulos, Th. Surf. Coat. Technol., 201, 2006, 895-901. [262] Fawzy, M. H.; Ashour, M.M.; Abd El-Halim, A.M. Trans. Inst. Metal. Finish., 73, 1995, 132-138. [263] Fellner, P.; Cong, P.K. Surf. Coat. Technol., 82, 1996, 317-319. [264] Wan, Y.Z.; Wang, Y.L.; Tao, H.M.; Cheng, G.X.; Dong, X.H. Trans. IMF, 77, 1999, 52-54. [265] Dai, P-Q.; Xu, W-C. Huang, Q-Y. Mater. Sci. Eng. A, 483-484, 2008, 172-174. [266] Zimmerman, A.F.; Palumbo, G.; Aust, K.T.; Erb, U. Mater. Sci. Eng. A, 328, 2002, 137–146. [267] Xinmin, X.; Zongang, D. Trans. Inst. Met. Finish., 70 (2) (1992) 84-86, [268] Grahl, A.; Jobst, K., Warlimont, H. Plat. Surf. Finish., 86, 1999, 98-103.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Plating of Nano-composites- Overview and Trends

299

[269] Tang, J.; Huang, L.; Yao, S.; Zhou, S. Plat. Surf. Finish., 85, 1998, 84-86. [270] Wu, Y.; Gao, J.; Shen, B.; Liu, L.; Hu, W.; Ding, W. Plat. Surf. Finish., 92, 2005, 38. [271] Shrestha, N.K.; Saji, T. Surf. Coat. Technol, 186, 2004, 444-449. [272] Srivastava, M.; Willams Grips, V.K.; Rajam, K.S. Mater. Lett;, 62,2008, 3487-3489. [273] Klingenberg, M.L.; Brooman, E.W.; Naguy, T.A. Plat. Sruf. Finish., 92, 2005, 42-48. [274] Lukashev, E.A. Russ. J. Electrochem. Soc., 30, 1994, 83-86. [275] Zhu, L.; Feng, L.; He, Y. Plat Surf. Finish., 92, 2005, 49-53. [276] Arai, S.; Fujimori, A.; Murai, M.; Endo, M. Mater. Lett., 62, 2008, 3545-3548. [277] Gyftou, P.; Pavlatou, E.A.; Spyrellis, N. Appl. Surf. Sci., 254, 2008, 5910-5916. [278] Gyftou, P.; Stroumbouli, M.; Pavlatou, E.A.; Asimidis, P., Spyrellis, N. Electrochem. Acta, 50, 2005, 4544-4550. [279] Thiemig, D.; Bund, A. Appl. Surf. Sci., 255, 2009, 4164-4170. [280] Zhang, J. M.; Zhang, Y.; Xu, K.W. J. Cryst. Growth, 285, 2005, 427-430. [281] Eisenmann, E. T. J. Electrochem. Soc., 124, 1977, 1957. [282] Denise, F.; Leidheiser, H. J. Electrochem. Soc. 100, 1953, 490-495. [283] Evans, D. J. Trans. Far. Soc., 54, 1958, 1086-1091. [284] Knödler, A. Metalloberfläche 20 (1966) 52. [285] Solomaon, A.J.; Antler, M. Plating, August, 1970, 812-816. [286] Flühmann, W.F.; Reid, F.H.; Mausli, P.A.; Steinemann, G. Plat. Surf. Finish., 67, 1980, 62-65. [287] Withlaw, Souter, J. W.; Wright, I. S. Proc. 12th ICECP, 1984, 33-45.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[288] Hou, K. H.; Ger, M. -D. Wang, L. M.; Ke, S. T. Wear, 253, 2002, 994-1003.

[289] Fisher, J. R.; Gurland, J.; Metal Sci. 15, 1981, 185-192. [290] Clyne, T.W.; Withers, P.J. An Introduction to Metal Matrix Composites, Cambridge University Press, Cambridge, 1993, pp. 256–263. [291] Garcia-Lecina, E.; Garcia-Urrutia, I.; Diez, J. A.; Salvo, M.; Smeacetto, F.; Gautier, G.; Seddon, R.; Martin, R. Electrochem. Acta, 54, 2009, 2556-2562. [292] Aslanyan, I. R.; Bonino, J. -P.; Celis, J. P. Surf. Coat. Technol., 200, 2006, 2909-2916. [293] Aslanyan, I. R.; Bonino, J. -P.; Celis, J. P. Surf. Coat. Technol., 201, 2006, 581-589. [294] Yucheng, W.; Rong, R.; Fengtao, W.; Zaoshi, Y.; Tugen, W.; Xiaoye, H. Mater. Res. Bull., 43, 2008, 3425-3432. [295] Hou, F.; Wang, W.; Guo, H. Appl. Surf. Sci. 252, 2006, 3812–3817. [296] Rapoport, L.; Bilik, Y.; Feldman, Y.; Homyonfer, M.; Cohen, S. R.; Tenne, R. Nature, 387, 1997, 791-793. [297] Huang, H.D.; Tu, J. P.; Zou, T.Z.; Chen, L. M.; Wang, Q.; Zhang, L. L.; He, D.N. Mater. Sci. Eng. A, 426, 2006, 162-168. [298] Garcia, I.; Conde, A.; Langelaan, G.; Fransaer, J.; Celis, J. P. Corro. Sci., 45, 2003, 1173-1189. [299] Ghosh. S. K.; Grover, A.K.; Dusane, R. O. J. Alloys and Compds., 426, 2006, 235-243. [300] Mishra, R.; Balasubramanium, R. Corro. Sci., 46, 2004, 3019-3029. [301] Rofaga, R.; Langer, R.; Al-Sherik, A.M.; Erb, U.; Palumbo, G.; Aust, K.T. Script. Metal. Mater., 25, 1991, 2867-2872. [302] Cameron, B.P.; Foster, J.; Carew, J.A. Trans Inst. Metal. Finish., 57, 1979, 113-118. [303] Kimoto, M.; Yakawa, A.; Tsuda, T.; Kammel, R. Metall. 44, 1990, 1148-1152. [304] Rajiv, E. P.; Seshadri, S.K. Plat. Surf. Finish., 79(6), 1992, 85-88. [305] Periene, N.; Cesuniene, A.; Taicas, L. Plat. Surf. Finish., 80, 1993, 73-76.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

300

Subir Kumar Ghosh and Jean-Pierre Celis

[306] Wang, L.Y.; Tu, J.P.; Chen, W.X.; Wang, Y.C.; Liu, X.K.; Olk, C.; Cheng, D.H.; Zhang, X.B. Wear, 254, 2003, 1289-1293. [307] Pushpavanam, M.; Natarajan, S.R.; Balakrishnan, K. Plat. Surf. Finish., 84(6), 1997, 88-91. [308] Periene, N.; Cesuniene, A.; Matulionis, E. Plat. Surf. Finish., 81(10), 1994, 68-71. [309] Benea, L.; Ponthiaux, P.; Wenger, F.; Galland, J.; Hertz, D.; Malo, J.Y. Wear, 256(9– 10), 2004 948–953. [310] Lemaire, E.; Le Calvar, M. Wear, 249(5–6), 2001, 338–344. [311] Celis, J. P.; Ponthiaux, P.; Wenger, F. Wear, 261, 2006, 939–946. [312] Goodman, S.B. Compatibility of biomedical implants: P. Kovacs, N.S. Istephanous (Eds.), Proceedings of the Electrochemical Society 94-15, The Electrochemical Society Inc., Pennigton, NJ, 1994. [313] Meunier, L. Sedel, Actes des journe´ es francophones de tribologie-corrosion, in: Ph. Kapsa (Ed.), Socie´ te´ Tribologique de France, SIRPE, Paris, 1998, pp. 193–200. [314] Watson, S.W.; Friedersdorf, F.J.; Madsen, B.W.; Cramer, S.D. Wear, 181-183, 1995, 476–484. [315] Batchelor, A.W.; Stachowiak, G.W. Wear 123, 1988, 281–291. [316] Schumacher, W.J. Mater. Perform. 32, 1993, 50–53. [317] Jiang, X.; Li, S.Z.; Tao, D.D.; Yang, J.X. Corrosion, 49, 1993, 836–841. [318] Stemp, M.; Mischler, S.; Landolt, D. Wear, 255, 2003, 466–475. [319] Bratu, F.; Benea, L.; Celis, J.-P. Surf. Coat. Technol. 201, 2007, 6940-6946. [320] Benea, L.; Wenger, F.; Ponthiaux, P.; Celis, J.P. Wear, 266, 2009, 398-405. [321] Benea, L. J Appl Electrochem., DOI 10.1007/s10800-009-9859-5. [322] Stappers, L.; Fransaer, J. J. electrochem. Soc., 154, 2007, D598-D611. [323] Hirato, T.; Fransaer, J.; Celis, J.P. J. Electrochem. Soc., 148, 2001, C280-C283. [324] Stappers, L.; Fransaer, J. J. Electrochem. Soc., 153, 2006, C472-C482. [325] Nowak, P.; Socha, R.P.; Kaisheva, M.; Fransaer, J.; Celis, J. P.; Stoinov, Z. J. Appl. Electrochem., 30, 2000, 429-431. [326] Dedeloudis, C.; Fransaer, J.; Celis, J. P. J. Phys. Chem., 104, 2000, 2060–2066. [327] Takei, T. Bull. Chem. Soc. Jpn., 47, 1974, 257-264. [328] Sarabi, R. S.; Singh, V.B. Met. Finish., 96, 1998, 26-31.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: High Performance Coatings for Automotive and Aerospace… ISBN: 978-1-60876-579-9 Editor: Abdel Salam Hamdy Makhlouf, pp. 301-324 ©2010 Nova Science Publishers, Inc.

Chapter 9

IMPROVEMENT OF THE REINFORCEMENTS DISTRIBUTION IN THE COMPOSITE MATRICES USING POWDER COATING PROCESS Walid M. Daoush1*, Byung K. Lim2, Hee S. Park2, Sayed F. Moustafa,1 and Soon H. Hong2 Department of Powder Technology, Central Metallurgical Research & Development Institute, P.O.87 Helwan, Cairo, Egypt1 Department of Materials Science and Engineering, Korea Advanced Institute of Science and Technology, 373-1, Guseong-dong, Yuseonggu, Daejeon, 305-701, Korea2

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Composite material is a tailor-made material which provides extraordinary flexibility to design the required properties to suit an application. Density and homogeneity of the composites are very important factors in the engineering applications because inhomogeneity and residual pores are harmful to the mechanical and physical properties of the composites. However, when using the conventional powder metallurgy route, the interfacial bonding of matrix and reinforcements proves to be weak due to their mutual insolubility and/or non-wetability, which results in low density, high porosity content, and segregation of reinforcements. Several alternative processing techniques have been researched to develop composites through hot isostatic pressing and liquid phase sintering. Although these techniques could overcome the problem of low density, they failed to solve the problem of in-homogeneity. Composite coatings improved properties of composites heavily dependent on the nature and content of reinforcements in the coatings. The most important point is to obtain continuous, uniformly-distributed and dense-coated metal layer. Otherwise, the layer with voids or gaps may weaken, even destroying the integration between the reinforcement and the matrix. There are many advantages of the coating technique of the powders such as improvement the homogeneity and better distribution of reinforcement materials within matrix, increasing the density and good interfacial bonding, capability of *

Corresponding Author: [email protected]

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

302

Walid M. Daoush, Byung K. Lim, Hee S. Park et al using fine powders, possibility of using small alloying addition in a more uniform manner (e.g. grain growth inhibitors and micro-alloying), enhancing wetability between reinforcement and molten or semi fusion matrix, and finally adjusting the densities between blended powders to enhance mixing and powder distributions. There are only a few methods that can be used for coating (encapsulating) reinforcement powders such as electrolytic deposition, electroless deposition, molten salts method, and mechanical alloying. This work aims at improving the interfacial bonding between the metal matrix and the reinforcements phase. Several kinds of reinforcements were used as metallic or ceramic powders in the form of particles metallic or non-metallic, fibers, short fibers, and carbon nano-tube were coated by metallic or ceramic layers using the electroless deposition, electrolytic deposition, molten salts method, and mechanical alloying processes to enhance the bonding between the matrix and the reinforcement phases as well as to increase the distributions and the homogeneity of the reinforcements in the matrix. The achievement of this work are the enhancement and improvement of the properties of the products produced by traditional powder technology routes such as gears, brushes for applying to in the automotive industry, heat sink materials, and reinforced polymers for applying to the electronic industry.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION Coating techniques have been used to prepare the composite materials by powder technology. As the advancement of powder coatings, in which particles are used in the reinforcing phase, the coated metal or ceramic was a matrix. But up to now, it is described rarely about nanometer composite coatings. Electroless deposition with catalytic metals was an effective way for necessary surface treatments, after which the coated layers can serve as medium for adhesion and transferring loads. Previous studies proceeded with coating metals and ceramics by metals with sensitization by stannous chloride followed by activations with palladium chloride and after that coating of the activated surface with metal layer, there followed by mixing these composite powders with metal powder and underwent powder technology processing of these composites. On the other hand electro-deposition of metals on the surface of particles or fibers is a unique method to improve the contact between the core particle or fiber and the outer metallic layer. In addition, deposition techniques involving metallic materials are generally used to improve the surface properties of ceramic materials. To date, techniques such as PVD, CVD, and metallic powder sintering have been developed. The molten salt reaction is a new, simple and low cost method, whereby a metallic layer is joined with ceramics very well. It is expected that the metal matrix composites reinforced by particles, short fibers, long fibers, or carbon nano-tubes would have better properties than the pure metals depends on the volume fraction of each reinforcement in the matrix. In general, the mechanical properties of composites are often affected by microstructures in the composites. In the following chapter the authors summarized and discussed their previous research work related to different coating methods for different reinforcements with different applications, especially in the automotive industry like friction materials, cutting abrasive materials, magnetic materials, heat sink materials, metallic filters for oil purifications, and

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Improvement of the Reinforcements Distribution in the Composite Matrices …

303

electromagnetic shielding materials. Our prepared materials were classified according to the type of reinforcement and the matrix for each material to particle reinforcements in ceramic matrix, particle reinforcements in metal matrix, fiber reinforcements in metal matrix, fiber reinforcements in polymer matrix, CNT reinforcements in metal matrix and lamellar composites, and sandwich panel composites. The following chapter includes these different types of composite materials according to their applications.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

2. NICKEL-ALUMINUM INTER-METALLICS Nickel-Aluminum (NiAl) inter-metallics are heat-resistant materials based on nickel and are generally used at temperatures above about 800oC. These highly microstructurallycomplex materials exhibit a combination of mechanical strength and resistance to surface degradation [1]. NiAl inter-metallics have been used in aircraft, industrial, and marine gas turbines; nuclear reactors; aircraft skins; space craft structures; petrochemical production; and environmental protection applications. Although developed for high-temperature applications, some are used at cryogenic temperatures. Applications continue to expand, although aerospace remains the predominant application [2]. Researchers from the Central Metallurgy have several techniques for synthesizing inter-metallics, including conventional melting and solidification [3]; reactive solidification, where a stoichiometrically-weighed high melting point metal is immersed or submerged into a carefully prepared and weighted molten pool of the lower melting point metal to initiate the exothermic reaction and, thus, the fabrication of the desired inter-metallics [4]; reactive sintering of constituent powders, where an exothermic reaction is initiated in a mixture of various constituent powders and the resulting heat is used for the sintering process [5]; and explosive bonding, which is based on the chemicallyinduced shock consolidation of constituent powders [6]. Other methods include selfpropagating high temperature synthesis (SHS), where an exothermic reaction of dissimilar mixed powders is initiated at one end of the powder perform and the resulting heat is used to induce reaction within the unreacted powders in a continual wave form [7]; dry mixing of the constituents followed by hot isostatic pressing [8]; mechanical alloying in several times [9]; and conventional melting followed by cold rolling to decrease the number of pores [10]. In our work, an electroless method was used for encapsulation of aluminum powder by nickel to prepare NiAl composite powder which underwent cold compaction and sintering at different temperatures to identify each of the inter-metallic phases. Electrical and magnetic properties were measured to identify the inter-metallics. Figure 1 shows a schematic diagram for the electroless deposition bath. Aluminum powders of 95%

CNP

Properties Length: 1-3µm Diameter: 50-70nm Length: 30nm, Purity > 95%

2.2. Effect of Solvents on the Viscosity Because of the high viscosity of AV138M (100,000 - 300,000 Cp at 25 ºC), the dispersion of nanofillers in the matrix was not possible without decreasing the viscosity to the processable range. Two solvents, Di-Methyl Formamide (DMF) and N-Methyl Pyrrolidone (NMP) were selected [9, 10] to reduce the viscosity of AV138M. The properties of the solvents are presented in Table 2. The viscosity measurements were carried out using programmable viscometer at room temperature. The solvent added to MWCNT/CNP and sonicated allows the agglomerates to be separated [11] due to the vibration energy [5] and the solid amorphous carbon to settle to the bottom of the solution [12]. Table 2. Properties of Solvents Density (g/cm3) 0.944 1.028

Boiling Point (ºC) 153 202

Freezing Point (ºC) - 61 -24

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Solvent DMF NMP

Figure 1. Effect of DMF and NMP solvent on the viscosity of AV138M. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

328

M S Vinod and V J Sundaram

The wt % of NMP added to the resin was higher than the wt % of DMF to bring the viscosity of AV138M to the processable (castable) range. The trend of reduction in viscosity with the addition of solvents is clearly depicted in Fig.1. The boiling point and density of NMP was higher than that of DMF. So, DMF was selected as the solvent for AV138M.

2.3. Nano Composite Preparation MWCNTs were ultrasonicated with DMF for 1h to achieve good dispersion. AV138M was mixed with DMF separately and added to the above suspension and sonicated again for 1h. Hardener HV998-Solvent mixture were added into this suspension (Resin : Hardener – 100 : 40 by wt) and poured in to Teflon mould of size 260mm x 130mm x 3mm, kept under vacuum for 1h. It was then cured for 24 h at room temperature and finally post cured for 1h at 60ºC. The same procedure was adopted for dispersing CNPs into AV138M. Adding the hardener and further processing the composite according to the manufacturer‘s regulations, the final level of dispersion was achieved.

2.4. Electrical Resistivity Measurements

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The DC electrical resistivity values of neat cast (NC)/resin and MWCNT/CNP nanocomposite specimens (100mm diameter and 3mm thick) were measured as per ASTM D257, using Keithley 6517 Electrometer (maximum resistance = 1017  and minimum resistance = 50  ). Silver paints were used at the electrode point for ohmic contact. A constant DC voltage of 500V was applied and a two-probe method was chosen for the test, because of high resistance. The resistances were measured and resistivity in -cm was computed.

2.5. Mechanical Property Measurements Tensile tests of nano-composites were studied using a 10T UTM according to ASTM 638 with specimen dimensions as 216 mm x 19mm x 3mm.

2.6. Morphological Characterization Scanning Electron Microscope (SEM) was used to study the dispersion of the fillers in the epoxy composites. All the samples were investigated with gold – sputtering.

2.7. Differential Scanning Calorimeter (DSC) Measurements Glass Transition Temperature (Tg) of AV138M / CNP and AV138M / MWCNT nanocomposites specimens were obtained using DSC. 5mg sample was sealed in hermetic Aluminium crucible. To obtain the curing heat flow pattern of the composite, dynamic

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Mechanical and Electrical Characterization of MWCNT/CNP …

329

scanning experiment was conducted with a ramp of 20oC/min from room temperature to 150oC.

3. RESULTS AND DISCUSSIONS 3.1. Volume Resistivity Results

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The neat resin casting was electrically insulting (1.8 x 1017  -cm) and the electrical resistivity decreased with an increase in the concentration of MWCNT and CNP as shown in Fig. 2. Both the resistivity curves exhibited similar pattern of behavior as evidenced by the slopes of the curves; steep decrease followed by gradual decrease. 0.1wt% specimens showed a drastic drop in resistivity values to 1.2 x1014  -cm for MWCNT based composites and 1.94 x1015  -cm for CNP based nano-composites. But in both cases further addition of reinforcement showed nominal changes in resistivity. Though the resistivity drop up to1011  -cm was achieved in both cases, the loading levels of the fillers were different.

Figure 2. Effect of MWCNT and CNP on Volume Resistivity of the nano-specimens.

Typically for electrostatic charge dissipation to occur the resistivity should be between 10 - 1010  -cm. An increase of conductivity by a factor of 100% was achieved. MWCNTs are generally good conductors and typically have a very high aspect ratio. Due to improvement in dispersion of nano tubes in the epoxy, aggregated phases formed a conductive 3D network throughout the samples. Such composites show characteristic percolation behavior. Both the percolation threshold and the maximum composite conductivity appear to depend on the type of carbon tubes and the degree of dispersion [13], i.e., the high aspect ratio of MWCNT leads to percolation at a lower volume fraction and thus increases the electrical conductivity of the polymer at a lower volume fraction [14]. CNT 17

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

330

M S Vinod and V J Sundaram

based nano-composites have higher conductivities and lower percolation thresholds than either Carbon Black (CB) or CNP [4, 15].

3.2 Mechanical Properties

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Fig. 3 shows the Ultimate Tensile Strength (UTS) vs. wt% of MWCNT and CNP addition into epoxy nano-composites. Nano-composite samples prepared using MWCNT showed better improvement in UTS as compared to nano-composite samples filled by CNP. This may be due to the poor dispersion, solvent effect, or solvent residual in the polymer matrix. However, the MWCNT/Epoxy composites are found to possess 30-45% higher UTS improvement than the CNP/Epoxy composites at the same loading. But, there was an overall increase of UTS by 160% (NC to 2wt% nano-filler addition). The results demonstrate that composites using DMF as solvent, can withstand higher loads compared to that of NMP based composites. There are relatively large holes seen in the NMP solvent based neat resin as well as nano-composites indicating that the load transfer is not allowed from the matrix to the nano particles due to poor adhesion between them. But in case of MWCNT based polymer composites, the load has been transferred through interfacial shear stress between polymers and the sidewalls of CNTs [18]. The strong interaction between the MWCNT and the matrix may be due to the mechanical interlocking of the coil with the matrix, which makes CNT better anchored in the embedding matrix [16]. The intriguing feature is that, in contrast to traditional carbon fibers, nanotubes remain curved and interwoven in the composite, suggesting extreme flexibility [17]. Nanotube reinforcements, however, will increase the toughness of the composites by absorbing energy because of their highly flexible elastic behavior during loading. In nano tube-filled adhesive polymer films, adhesion between nano tube-filled polymer surfaces is greatly enhanced due to the attractive van der Waal‘s forces between the nano-tubes present on each of the surfaces [1].

Figure 3. Effect of CNP and MWCNT on Ultimate Tensile Strength of Nano-Composites Specimens. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Mechanical and Electrical Characterization of MWCNT/CNP …

331

3.3. Morphological Characterization

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Fig. 4(a) and Fig. 4(b) show the morphology of neat resin prepared by using 5% DMF and 7% NMP solvent. Fig. 4(b) shows the presence of very minute blow holes (porosity - P) present in the neat resin with addition of 7% NMP. Fig. 4(a) shows the absence of blow holes with the addition of 5% DMF into the neat resin. Fig. 4(c) shows good dispersion of 1wt% MWCNT into AV138M with 5% DMF solvent, and Fig. 4(d) shows the dispersion of 1wt% of MWCNT into AV138M with 7% NMP solvent. Presence of MWCNT aggregates (A) and porosity (P) are more promoted in the NMP based nano-composites samples as compared to that in DMF samples shown. In the well-dispersed sample (using DMF), none or little porosity and relatively homogeneous distribution of MWCNTs was observed, though some places showed a little thicker. The investigation via SEM showed that the agglomeration could be reduced by the introduction of DMF leading to a better dispersion of the carbon nano tubes. The presence of DMF enabled better dispersion of CNTs and so reduced the amount of impurities. Hence, DMF has been preferred to NMP.

Figure 4. SEM Micrographs, (a) SEM of Epoxy neat resin with DMF as the solvent, (b) SEM of Epoxy neat resin with NMP as the solvent, (c) 1 wt% MWCNT dispersed Epoxy nano-composites with DMF as the solvent, (d) 1 wt% MWCNT dispersed Epoxy nano-composites with NMP as the solvent.

3.4. DSC Results (Glass Transition Behavior) DSC was used to determine the Tg, using the measurement of heat flow depending on the change in temperature. The polymer has a higher specific heat above the Tg, because at the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

332

M S Vinod and V J Sundaram

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

glassy stage the chemical chains are more mobile and has more effective degrees of freedom. At high temperatures the barriers to rotation about chemical bonds are low enough for the chains to be constantly changing their conformations, and they can easily respond to an applied stress to change the shape of the polymer. The DSC graph (Fig. 5) shows increase in glass transition temperature with the addition of MWCNT and CNP. The neat resin AV138M/HV998 exhibited Tg at 79.01ºC, whereas the addition of solvents (DMF and NMP - 5 wt %) resulted in slight reduction of Tg to the extent of 0.847 %. The addition of MWCNT yielded high value of Tg than CNP [19]. Continuous increase in Tg is observed w. r. t the addition of MWCNT up to 1 wt %. So the addition of nanofillers enhanced the thermo stability of the composites, may be due to the reduction in the mobility of the epoxy polymer chains around the nanofillers by strong interfacial interactions [20-22]. The increase in Tg is an indicator of the improvement in the mechanical properties of the composites. Greater increase in Tg in case of MWCNT is attributed to its greater thermal stability and superior thermal properties than that of CB/CNP.

Figure 5. Glass Transition Temperatures of the nano-composites.

CONCLUSIONS A procedure has been developed to disperse the nanofillers, namely CNP and MWCNT into AV138M using a suitable solvent. The solvent selected, is based on its quantity addition to reduce the viscosity of the resin system to castable range and its boiling point. SEM images showed the presence of nano-filler aggregates and porosity being more promoted in the NMP based nano-composite samples as compared to that in DMF. Thus, DMF is preferred to NMP because it provides good interlock between MWCNT/CNP and the Epoxy.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Mechanical and Electrical Characterization of MWCNT/CNP …

333

The mechanical and electrical properties of MWCNT/Epoxy and CNP/Epoxy nanocomposites fabricated using DMF solvents were investigated with loading up to 2 wt%. The nano-composites showed reduction in electrical resistivity from 1017 - 1011  -cm, which means, there was an increase of conductivity by a factor of 100%. And AV138M/MWCNT specimens proved superior to AV138M/CNP since the loading levels in the former has been less. The MWCNT/Epoxy composites were found to possess 30-45 % higher UTS improvement than the CNP/Epoxy composites at the same loading. Also, there was an overall increase of UTS by 160% (NC to 2wt% nano-filler addition). Hence, the specimens have been characterized for its improved mechanical properties as well. So, MWCNT containing composites exhibit higher UTS and electrical conductivity than that of CNP based composites. And CNP is used as a reference in order to appraise the effect of the particleshape on the achieved improvements in various properties and to obtain the information on the real potential of CNT. Thus, this area of research will clearly lead to fruitful commercial applications, with significant economic effect, driven by materials with new combination of properties.

ACKNOWLEDGMENTS The authors acknowledge the Indian National Academy of Engineering (INAE), New Delhi for granting the Research Fellowship under the ‗Mentoring of Engineering Students by INAE Fellows‘ scheme for the year 2009. (INAE/405/RS dated 25-02-2009)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

REFERENCES [1]

[2]

[3]

[4]

[5]

[6]

Ajayan, P. M., Schadler, L. S., Giannaris, S. C., and Rubio, A., ―Single-Walled Carbon Nanotube-Polymer Composites: Strength and Weakness‖, Advanced Materials, Vol. 12, No. 10, pp. 750-753, 2000. Sandler, J. K. W., Kinloch, I. A., Shaffer, M. S. P. and Windle, A. H., ―Ultralow Electrical Percolation Threshold in Carbon Nanotube Epoxy Composites‖, Polymer, Vol. 44, No. 19, pp. 5893-5899, 2003. Nan, C. W., Shi, Z. and Lin, Y., ―A Simple Model for Thermal Conductivity of Carbon Nanotube based Composites‖, Chemical Physics Letters, Vol. 375, No. 5, pp. 666-669, 2003. Coleman, J. N., Dalton, A. B., Curran, S., Rubio, A., Davey, A. P., Drury, A., McCarthy, B., Lahr, B., Ajayan, P. M., Roth, S., Barklie, R. C., Blau, W. J., ―Phase Separation of Carbon Nanotubes and Turbostratic Graphite Using a Functional Organic Polymer‖, Advanced Materials, Vol. 12, No. 3, pp. 213-216, 2000. Bodo Fiedler, F., Malte, H. G., Wichmann, H. G. Mathias, A. M. and Schulte, K., ―Fundamental Aspects of Nano-reinforced Composites‖, Composite Science and Technology, Vol. 66, No. 11, pp. 3115-3125, 2006. Ajayan, P. M., Schadler, L. and Braun, P., ―Nanocomposite Science and Technology‖, VCH/Wiley Publishers, 2003.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

334 [7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[15]

[16]

[17]

[18] [19]

[20]

[21]

M S Vinod and V J Sundaram Allaoui, A., Bai, S., Cheng, H. M. and Bai, J. B., ―Mechanical and Electrical Properties of a MWCNT/ Epoxy Composite‖, Composites Science and Technology, Vol. 62, No. 15, pp. 1993-1998, 2002. Lau, K. T. and Shi, S. Q., ―Failure Mechanisms of Carbon Nanotube/Epoxy Composites Pretreated in Different Temperature Environments‖, Carbon, Vol. 40, No. 15, pp. 2961-2968, 2002. Song, Y. and Youn, J. R., ―Influence of Dispersion States of Carbon Nanotubes on Physical Properties of Epoxy Nano-composites‖, Carbon, Vol. 43, No. 7, pp. 16051606, 2005. Ausman, K. D., Piner, R., Lourie. O., Ruoff, R. S., and Korobov, M., ―Organic Solvent Dispersions of Single-Walled Carbon Nanotubes: Toward Solutions of Pristine Nanotubes‖, Journal of Physical Chemistry B, Vol. 104, No. 38, pp. 8911-8915, 2000. Carotenuto, G., Her, Y.S., and Matijevic, E., "Preparation and Characterization of Nanocomposite Thin Films for Optical Devices", Industrial & Engineering Chemistry Research, Vol. 35, No. 9, pp. 2929-2932, 1996. Kausala, M., and Zhang, L. C., ―Fabrication and Application of Polymer Composites Comprising Carbon Nanotubes‖, Recent Patents on Nanotechnology, Vol. 1, No.1, pp. 59-65, 2007. Gojny, F. H., Malte, H. G. W., Bodo Fiedler, I. A. K., Bauhofer, W., Windle, A. and Schulte, K., ―Evaluation and Identification of Electrical and Thermal Conduction Mechanisms in Carbon Nanotube/Epoxy Composites‖, Polymer, Vol. 47, No. 6, pp. 110, 2006. Xiao, P., Xiao, M., and Gong, K., ―Preparation of Exfoliated Graphite/Polystyrene composite by Polymerization-filling technique‖, Polymer, Vol. 42, No. 11, pp. 48134816, 2001. Xu, Y., Ray, G., and Abdel-Magid, B., ―Thermal Behavior of SWCNT Polymer-Matrix Composites‖, Composites Part A: Applied Science and Manufacturing, Vol. 37, No. 1, pp. 114-121, 2006. Kin-tak Lau, Mei Lu, and Kin Liao, ―Improved Mechanical Properties of Coiled Carbon Nanotubes Reinforced Epoxy Nano-composites‖, Composites Part A: Applied Science and Manufacturing, Vol. 37, No. 10, pp. 1837-1840, 2006. Sungjin Park, Seung Woong Yoon, Heechol Choi, Joon Sung Lee, Woo Kyung Cho, Jinhee Kim, Hyung Ju Park, Wan Soo Yun, Cheol Ho Choi, Youngkyu Do, and Insung S. Choi, ―Pristine Multiwalled Carbon Nanotube/Polyethylene Nano-composites by Immobilized Catalysts‖, Chemistry of Materials, Vol. 20, No.14, pp. 4588-4594, 2008. Schadler, L. S., Giannaris, S. C., and Ajayan, P. M., ―Load transfer in carbon nanotube epoxy composites‖, Applied Physics Letters, Vol. 73, No. 26, pp. 3842-3844, 1998. Vinod, M. S., Alok Shankar, Jeena, J. K., Sreejith, M., Murthy, H. N. N., and Krishna, M., ―Effect of Dispersing Nano-materials into Structural Adhesive on the Electrical and Mechanical Properties‖, Innovations in Composites for the New Century - Proceedings of INCCOM-6 Conference, pp. 1016-1021, 2008. Wang, Z., and Pinnavaia, T. J., ―Hybrid Organic−Inorganic Nanocomposites: Exfoliation of Magadiite Nanolayers in an Elastomeric Epoxy Polymer‖, Chemistry of Materials, Vol. 10, No. 7, pp. 1820-1826, 1998. Chin, I. J., Thurn-Albrecht, T., Kim, H. C., Russel, T. P., and Wang, J., ―On Exfoliation of montmorillonite in Epoxy‖, Polymer, Vol. 42, No. 13, pp. 5947-5952, 2001.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Mechanical and Electrical Characterization of MWCNT/CNP …

335

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[22] Agag, T., Koga, T., and Takeichi, T., ―Studies on Thermal and Mechanical properties of Polyimide-clay Nanocomposites‖, Polymer, Vol. 42, No. 8, pp. 3399-3408, 2001.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: High Performance Coatings for Automotive and Aerospace… ISBN: 978-1-60876-579-9 Editor: Abdel Salam Hamdy Makhlouf, pp. 337-351 ©2010 Nova Science Publishers, Inc.

Chapter 11

CHARACTERIZING COATINGS OF CAR BODY SHEETS BY GLOW DISCHARGE OPTICAL EMISSION SPECTROMETRY (GD-OES) Tamás I. Török1, Gábor Lévai2, Mária Szabó3, and József Pallósi4

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Department of Metallurgy and Foundry Engineering, University of Miskolc, 3515 Miskolc-Egyetemváros, Hungary1 Department of Metallurgy and Foundry Engineering, University of Miskolc, 3515 Miskolc-Egyetemváros, Hungary2 ISD DUNAFERR Co. Ltd. Material Testing and Calibration Laboratories Directorate, Dunaújváros, Hungary3 ISD DUNAFERR Co. Ltd. Material Testing and Calibration Laboratories Directorate, Dunaújváros, Hungary44

ABSTRACT The major characteristics of today‘s automotive body paints and coatings are briefly reviewed. A typical system consists of several and different organic layers developed on successive metallic coatings deposited on a rigid substrate (steel or aluminium). Such a composite type set of layered materials, i.e. finished car body sheet panels, was tested using a fast analytical technique (GD-OES) in complement to some additional and more traditional materials testing techniques (Scanning Electron Microscopy /SEM/ and X-ray microprobe). By means of the radio frequency (rf) glow discharge sputtering, depth profiles of the samples could be determined down into the substrate metal (steel panels). Analytical results were obtained and correlated with the information gathered from the SEM images and those of microprobe analysis done on the cross sections of the samples. In this way the capabilities and benefits of GD-OES could be demonstrated in the field of testing such relatively complex systems.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

338

Tamás I. Török, Gábor Lévai, Mária Szabó et al

SURFACE TREATMENT OF CAR BODY SHEETS The history of automotive paint is nearly as old as the history of automotive. In the early 1900‘s the process of painting the car could in many cases took as long as 40 days. These products were not colourful, the automotive body paint was varnish-based and applied by brush. At that time Henry Ford always said, ―You can have a car any colour you like as long as it is ‘black‖. This system was used until the mid of 1920‘s [1]. In this early period painting and drying was a serious bottle-neck to the mass-production process. The answer to this prolonged paint-drying problem was the removal of timber from the bodywork allowing the car bodies to be heated. The Austrian Joe Ledwinka, together with the American Edward Budd developed methods for making welded steel bodies. The so-called ‘Buddism‘ enabled genuine mass-production of automobiles, and the development of all-steel body technology was parallel to the developments in paint chemistry [2].

Clear coat Colour coat Primer surfacer Primer electrocoat Zinc layer/deposit

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Steel substrate

Figure 1. Schematic representation of the typical coating system of today used on steel car body panels.

Later on, the effect of corrosion on vehicles has driven the automotive producers to incorporate anti-corrosion measures in the vehicle structures and body sheets applied for assembling modern cars. Today most car makers use pre-primed electrogalvanized, hot dip galvanized and galvannealed steel sheets for automotive bodies [3]. A typical coating system today has the following structure [4]: Table 1. Tendency for automotive powder coating Applied scope

State of art

Tendency

Requirements and driving force

Under chassis / underbody

Epoxy

Epoxy / Polyester with low curing temperature

Low price

Interior decoration Exterior decoration Primer / top coating for car body Clear coating for car body

Epoxy / polyester, polyester Polyester / polyurethan Epoxy / polyester, acrylics, polyester, polyester / acrylics Acrylics

Acrylics Polyester / polyurethan

Gloss and color controlling to be improved Enchancing UV resistance

Acrylics / polyester hybrid

Enchancing chip resistance and smoothness

GMA acrylics

Enhancing smoothness, chip and acid resistance

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission … 339 Today the automotive finishing technology has the highest level of surface treatment technology and auto coatings are among the best quality ones in the coating industry. Traditional finishing process includes solvent-based surface primer + base coat + clear coat, which features high quality, easy application, etc, but generates large VOCs (volatile organic compound) emission. However, waterborne systems can reduce the VOCs, and powder coating technology can eliminate the use of solvents. Waterborne and powder coatings are already widely used in Europe and North America. Table 1, after Chen Muzu [4], for example, summarizes the recent tendency for automotive powder coatings. Recently most of the modern automotive coatings consist of three typical organic layers applied on the metallic coatings covering the substrate: an electrocoat, a primer and a topcoat. The electrocoat, also called basecoat, is the first layer applied in automotive coatings, then the second layer is the primer surfacer and the final layer is the topcoat (or clearcoat). Basecoats and topcoats are polyester based whereas clearcoats are usually acrylic based. Automotive coatings are made of different resins. The choice of the resin used is totally dependent on the application. For the different layers, water soluble special resins, fillers and additives can be used. Automotive coatings offer a wide range of performances such as: Appearance and Aesthetics; Gloss; Resistance to heat, cold, ice and snow; Weatherability; UV resistance; Durability; Impact resistance.

PIGMENTS IN BODY SHEETS’ FINISH COATS

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The automotive industry also uses colourful paints in its painting lines partially due to the great variety of pigments available today. For red colour, for instance, several inorganic and organic red pigments are available of which a few examples [5] are given in Table 3. Table 2. Examples of common red pigments used in paints Inorganic red pigments Lead chromate Cadmium red Red iron oxide Lead molybdate

Organic red pigments

Anthraquinone

Disazo condensation pigments Quinacridone Perylene Benzimidazolone

Dibromanthrone Diketopyrrolo-pyrrole pigments (DPP)

Usage in vehicle body paints

In automotive metallics, often in combination with transparent iron oxides, and formerly in solid shades with molybdate reds. Can also be used in industrial and vehicle refinishing paints and as the basis of tinting system. Used in automotive OEM paints, vehicle refinishing. Used in OEM automotive finishes. Can also be used in metallic automotive and vehicle refinishing paints In automotive OEM and vehicle refinishing paints, coil coatings, and occasionally in tinting systems. In automotive OEM paints (often in combination with other more economic pigments)

The organic pigments often have complicated chemical structures like antraquinone shown below. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

340

Tamás I. Török, Gábor Lévai, Mária Szabó et al

Properties of anthraquinone Property

Anthraquinone

Color

bright, strong

Solvent resistance

good

Heat Stability

good

Light fastness

moderate-good

Figure 2. Structure and some major properties of an organic red pigment.

THE ”EFFECT” PIGMENTS Car manufacturers are continually seeking to increase choice and range of colours and even the so-called special finishes may well soon be commercially more widely available. Such special finishes include pearlescent and mica finishes which are now appearing on commercial models. This type of pigmented layers often contain aluminium flakes and even Electrostatic/Hand Spray Paint micronized and surface coated particles (Fig. 2) the latter ones make the paint appear to flip from a certain color to another one Cross-section (‘Hue shifting‘) depending on the prevailing light and viewing angle [6]. Aluminum flake

Hand spray

2nd or top layer

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Xirallic Pigment Cross-Section

Chromaflair Cross-section

1st or bottom layer

ColorStream Cross-section

Electrostatic spray METAL OXIDE (Ti or Fe)

MAGNESIUM FLUORIDE

METAL OXIDE (Ti or Fe)

ALUMINUM OXIDE

ALUMINUM

SILICON DIOXIDE

METAL OXIDE (Ti or Fe)

MAGNESIUM FLUORIDE

METAL OXIDE (Ti or Fe)

Xirallic

Chromaflair

Color Stream

Figure 3. Examples of a metallic pigment (aluminium developed in the colour coat by two different means of spraying technologies), and three special effect pigments (Xirallic, Chromaflair, Color Stream) used to produce interference colors or the hue shifting effect.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission … 341 The new ‘pearlescent‘-type pigment Xirallic is made along the same lines as the ‘tradidional‘ metal oxide coated mica pigments, but the base particle is an aluminium oxide produced synthetically to get more uniform particle sizes. The aluminum oxide particles are coated with a metal oxide (titanium or iron) to produce interference colors much like the traditional ‗pearlescent‘ pigments. This pigment is more flashy, less satiny than the mica based pearlescent pigments. Though the ‗hue shifting‘ pigments /e.g. Chromaflair and Color Stream/ are not cheap, and Chronaflair is mostly used for the inks of the new paper money, some car manufacturers like GM, for instance, have also started using them in some silver gray vehicles in low level [6].

GLOW DISCHARGE OPTICAL EMISSION SPECTROMETRY (GD-OES)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The phenomena of cathode sputtering is known for long time, but the use of glow discharge in spectrometry started when Grimm built his first glow dicharge spectrometer with an hollow anode source. Glow discharge spectrometry has become a widely used tool for surface and interface analysis with the Grimm type glow discharge source. The Grimm glow discharge source is a flat type source, which consists of an anode tube and a flat sample playing the role of cathode. There is a spacer (ceramic cathode block and O-ring) between the sample surface and the anode tube to maintain a fixed distance (d = 0.1 – 0.3 mm) and assure the vacuum tightness. [8] The volume in front of the flat sample is pumped down to the vacuum ~0.1-1 Pa (0.010.001 Torr) then filled up by high purity Ar gas up to 300-1300 Pa (2-10 Torr). (See Figure 4.)

Figure 4. Grimm-type glow discharge source. (With the permission of Horiba Jobin Yvon SAS).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

342

Tamás I. Török, Gábor Lévai, Mária Szabó et al

The anode is held at ground potential and the dc or rf voltage – which varies between 400 – 1200 volts – is applied to the back of the sample. When the plasma is ignited, free electrons and ions are generated. During the measurements this applied potential accelerates the electrons and the highenergy electrons collide with argon atoms causing the ionisation of the atoms and the creation of a plasma. The positive Ar ions are accelerated by the negative bias and collide with the sample surface. These collisions cause the sputtering of the surface and the sputtered excited atoms of the sample move away from the surface and are excited in the plasma where they emit photons with characteristic wavelengths (emission lines). [9] This cathodic sputtering forms a crater on the surface of the sample facing to the anode. The typical sputtering rate is between 2 – 5 m/min, the commonly used diameter of the anode tube is 4 mm. (See Figure 5.)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

TYPICAL RF CRATER (ON STEEL)

Figure 5. 2D Profile and view of a crater caused by rf glow discharge source (With the permission of Horiba Jobin Yvon SAS).

The emitted light –containing all emission lines of the atoms sputtered from the sample surface – pass through the lens into an optical spectrometer with many fixed channels (polychromator). During the sputtering process we measure the characteristic lines intensities of each element with photomultiplier tubes. (Figure 6.) The intensity of the light is proportional to the concentration of the lighting element. The elements emitting light came from different depths of the sample during the sputtering. If we plot the light intensities of the different elements versus the measurement time, we obtain a qualitative depth profile. After calibration between the ligth intensity and concentration of each element and a special calibration between the measuring time and the crater depth, we can built up the quantitative depth profile. (See Figure 7).

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission … 343

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 6. Typical layout of the glow discharge optical emission spectrometer (GD-OES) (With the permission of Horiba Jobin Yvon SAS).

Figure 7. Quantitative depth profile of a galvanised steel sheet(Zn on Fe).

Using a dc ( = direct current) discharge for the sputtering only conducting samples can be analysed, using rf ( = radio frequency) glow discharge then both conducting and nonconducting materials can be analysed. For characterizing the coatings of car body sheets we have been using a rf excited glow discharge optical emission spectrometer type JY 10000 rf equipped with 4 mm anode.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

344

Tamás I. Török, Gábor Lévai, Mária Szabó et al

ORIGIN AND COLLECTION OF SAMPLES For our present experimental study, exterior and finish coated steel body sheets were collected from several car repair shops with the primary aim of looking at the structure (number of layers, thicknesses, materials, etc.) and also at some elementary compositional properties of zinc coated steel sheets finished by the different paints used by various car makers in the last 20 years or so. The coated steel body sheet samples are grouped into three categories according to the size of vehicles (Table 3) among which most are relatively new models both from Europe /denoted by E/ and outside of Europe /NE/. The specific car makers are marked by numbers from 01 to 12. Table 3. Coated vehicle body sheet samples collected and tested Small Category

Middle Category

Premium Category

E – 04 / 2000

E – 01 / 2004

E – 07 / 2006

E – 06 / 2003

E – 06 / 2006

E – 11 / 2006

E – 06 / 2008

E – 06 / 2008

E – 12 / 2007

NE – 03 / 1995

E – 08 / 2004

NE – 01 / 2005 E = European Constructor NE = Non European

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

01-02-03… = Makercode / XXXX = Modell Year

All the samples were taken from the outside coated panels of the different vehicles.

GENERATION OF DEPTH PROFIES WITH THE GD-OES APPARATUS Most of the depth profile experiments were done using the above mentioned glow discharge optical emission spectrometer (GD-OES) type JY 10000 rf installed in the testing laboratory /ISD Dunaferr Co, Dunaujvaros/. The instrument is essentially dedicated to serve the special analytical needs of the iron and steel works in Dunaujvaros, therefore its measuring capabilities were originally adjusted for the analysis of metals and alloys, like aluminium, copper, nickel and tin. Its most important parameters are summarized in the Table 4.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission … 345 Table 4. Main technical parameters of the GD-OES equipment type JY 10000 rf

Jobin Yvon JY 10000 GD-OES Equipment POLYCHROMATOR Optics Channels

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Focal length Diffraction grating Resolution Spectral range Atmosphere in the optic‘s cabinet

Paschen-Runge polychromator with 29 channels H, B, C, N, O, Mg, Al, Si, P, S, Cl, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Zr, Nb, Mo, Cd, Sn,W, Pb 1m 3000 mm 0,3 nm / mm 120-520 nm Nitrogen

MONOCHROMATOR Focal length Atmosphere in the optic‘s cabinet

Czerny -Turner monochromator 1m Nitrogen

Detectable light elements Plasma generation Diameter of anode Maximal detectable depth Resolution of depth profile Relativ precision of depthprofile Detectable concentration interval Relative precision of layer-content

N, O, Cl, H, C RF 4 mm (standard) 0,1 mm ~ 5 nm 5% 100 % - 0,0001 % (dependent of element) 6%

Calibration of the equipment was done by using the Quantum IQ software package with standards and reference samples selected in view of the original analytical tasks for metallic materials. Hence, a reliable quantification of the non-metallic elements like carbon /C/, hydrogen /H/ and oxygen /O/, was not possible in the upper layers /clear coat, lacquer/ of the samples. Additionally some elements of possible interest (Ba etc) were not present amongst the channels. The measurement of the outermost organic layers of coated car body panels is very time consuming as these layers are heat sensitive and require soft operating conditions, Therefore it was decided to slightly scratch (with emery paper grade 320) all the samples before mounting them on the holder of the GD-OES equipment. Such scratches can still be seen in the red zones away from the craters shown in Figure 8. Doing that way more standard operating conditions could be used

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

346

Tamás I. Török, Gábor Lévai, Mária Szabó et al

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 8. Craters formed on sample E-06 / 2008 during rf plasma sputtering with the GD-OES apparatus. In case A the plasma discharge was stopped halfway while in case B the depth profile analysis was completed well into the steel substrate. The various layers are visible on the edge of the craters.

A set of four GD-OES depth profiles are displayed in Figure 9 to show some important other characteristics of this analytical tool when applied to painted car bodies. The different elements are displayed in arbitrary colours. 11 to 14 chemical elements were measured simultaneously during the depth profile analysis. In a GD depth profile the measurement is done from the surface (left of X axis) down to the base material (right of X axis). The zinc peak appears before the iron showing the presence of Zn layers on all the four samples presented in Figure 9. Other elements like sulphur, magnesium, silicon, and aluminium appear before zinc and can be related to different inorganic pigments and fillers of the lower organic layers /base coat and colour coat/. The magnitudes (i.e. the percentage ratios) of the latter ones are only semi-quantitative values as the organic components /C, H, O, N/ of these layers could not be determined with high accuracy due to the lack of proper standards and reference materials for the time being. The same reasoning applies to an even more pronounced extent to the first portions of the depth profiles, where the non-pigmented organic layers /clear coats/ are essentially containing C/N /H, O/. In addition the top surface was roughened by the prior mechanical abrading pretreatment applied as described above.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission … 347

Figure 9. A set of four GD-OES depth profiles displayed together to show some important common characteristics of the given GD-OES analytical tool applied to car body panels.

CHARACTERISATION OF THE COATED CAR BODY PANELS BY THE GD-OES TECHNIQUE An important advantage of the GD-OES technique is that it is quite fast; one can easily get depth profiles of many different coated steel sheets of the same kind in almost a few minutes.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

348

Tamás I. Török, Gábor Lévai, Mária Szabó et al

Features, benefits but also limitations of the GD-OES technique in such a relatively new and complex application field have been described above, therefore it was decided to complement the depth profiles with some additional experimental testings. For that purpose cross sections of the car body panels were looked upon by means of scanning electron microscopy /SEM, type 18301 Amray/ and X-ray microprobe analysis (equipment type EDAX DX4 (EDS). The given SEM images of the cross section could therefore be shown in parallel to the depth profiles as in Figure 10.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

Figure 10. Top: GD-OES depth profile of a ‘graphite-metal‘ coloured car body panel (Sample E-01 / 2004) and below: cross section‘s SEM image showing the zinc and other organic coating layers.

In Figure 10 the elements sulphur /S/, silicon /Si/ and magnesium /Mg/ are elementary components of the chemical compounds used as fillers, while the element aluminium /Al/ refers to the use of Al flakes in order to generate the so-called ‘metal‘ effect of the organic coating. Though the element barium /Ba/ was absent from the list of accessible channels inside the given GD-OES apparatus, its presence in the coating sublayer next to the zinc was identified by X-ray microprobe (Figure 11), with its most probable chemical form being a barium sulphate, common compound used as filler in such organic coats. [10]

Figure 11. Results of X-ray microprobe analysis of the DIFFERENT ORGANIC LAYERS (Clear Coat, Pigmented Layer, Base Coat) and of the ZINC COATING on the steel body sheet of Sample E01/2004. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission … 349

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

The total thickness of the coating layers of the higher category cars are often well above 100 µm, as for instance the Premium Category models E 07 and E 12 (See Table 3). Cross sections prepared from their finished body sheets were tested by scanning electron microscopy (Figure 12). The SEM images clearly show that the thicknesses of the outermost layers of clear coats are relatively high. Its principal reason is to provide also some kind of scratch resistance and protection of the lower organic layers from any type of degradation during normal use of the cars. The zinc layers are also relatively thick as compared to the recent tendency of applying thinner layers of zinc whenever possible for weight and cost reduction. Nevertheless, the zinc galvanizers are also trying to improve the corrosion resistance of zinc by proper alloying with for example aluminium, nickel and most recently with magnesium. The GD-OES technique has the capability to measure such elements in the zinc layers even at small amounts, and indeed the depth profiles recorded on the same samples of E 07 and E 12 show the presence of aluminium in the zinc (See Figure 13).

Figure 12. SEM images of cross sections prepared from samples of finished body sheets of Premium Category Cars, E 07 and E 12. Thicknesses shown from bottom to top refer to the zinc layers (15 µm and 30 µm, respectively) and the additional three different organic layers.

It is also clear from Figure 13 that the car body panel coated with the „Beige-metal‖ finishing paint contains aluminium flakes /modell E-12 / 2007/, as the large aluminium peak near the start of the depth profile reveals. The other coated body panel /sample of modell E-07 / 2006/ is not painted by a metaltype finish (there is no aluminium peak at the beginning), but it feratures a large sulphur peak referring to the pigmented layer with high filler content. In these 2 profiles the Al or the S peak appear relatively close to the beginning of the depth profile because the top layer was nearly completely removed by the mechanical polishing. Of course, such a preliminary sample preparation step somehow increases the uncertainty of determining the actual total depth of crater, but it saves a lot of time and should not affect the accuracy of the analysis of the lower layers containing less organic and more inorganic and metallic contents.

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

350

Tamás I. Török, Gábor Lévai, Mária Szabó et al

% 90 80 70 60 50 40 30 20 10 0 10

20

30

40

50

60

70

80

90

% 90 80 70 60 50

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

40 30 20 10 0

0

10

20

30

40

50

60

70

80

90

100

110

Figure 13. GD-OES depth profile of the coated car body sheets of models E-07 / 2006 and E-12 / 2007, for which the SEM images of cross sections are shown in Figure 12 above.

CONCLUSION It was demonstarted in this Chapter that Glow Discharge Optical Emission Spectrometry (GD-OES), which is a relatively novel and rather fast analytical technique, can be a useful tool in determining the major elementary components present in the coatings of car body sheets and identify the source of possible corrosion problems. Radio frequency exitation permits to create a plasma even with insulating materials. However, the outermost transparent organic layers, the so-called clear coats, which can be rather thick (> 50 µm) in the modern vehicles of today, remain challenging as they require the

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Characterizing Coatings of Car Body Sheets by Glow Discharge Optical Emission … 351 use of very soft operating conditions (consuming a lot of time). The risk alternatively if standard conditions are used is burn the layer creating instabilities in the profiles. This difficulty, however, can be overcome by simply rubbing down a substantial part of the top coat permitting one afterwards to test easily the other layers (organic and metallic down to the steel panel). If the GD-OES apparatus is equipped with all the relevant channels of the elements which are commonly present in the automotive paints, this technique could provide a sort of fingerprint of the coating system applied onto a given car body panel. By combining GD-OES technique, Scanning Electron Microscopy with X-ray microprobe, the primary results of the depth profile analysis can be more easily quantified. As it is the case at any spectroscopic technique, the elementary composition of the samples, e.g. in mass or atomic percentages, featuring wide concentration ranges and steady variations from layer to layer can be determined with high accuracy only if the necessary calibration samples and reference materials are available in the laboratory.

REFERENCES

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

[1] [2]

A brief History and Trends in Vehicle Paints, http://www.protectall.com/artpaints.htm Andrews, D.; Nieuwenhuis, P.; Ewing, P. D.: Black and beyond – colour and the massproduced motor car Optics & Laser Technology 2006, 38, 377-391. [3] Santos, D.; Raminhos, H.; Costa, M. R.; Diamantiono, T.; Goodwin, F.: Performance of finish coated galvanized steel sheets for automotive bodies Progress in Organic Coatings 2008, 62, 265-273. [4] Chen Muzu: Present Status and Future Development of Automotive Finishing Materials http://www.asiacoat.com/eng/emarkets/emarkets3.htm [5] http://www.specialchem4coatings.com/tc/color-handbook/index.aspx?id=red [6] What’s New in Automotive Paint Technology, Leanora Brun-Conti, Bureau of Alcohol, Tobacco, Firearms and Explosives [7] Grimm, W. Glimmentladungsröhre. German patent DE1589389, 1967. [8] Nelis, T.; Pallosi, J.: Glow Discharge as a Tool for Surface and Interface Analysis, Applied Spectroscopy Reviews 2006, 41, 227-258. [9] Payling, R.; Jones, D.; Bengtson, A.: Glow Discharge Optical Emission Spectrometry, Baffins Lane, Chichester, 1997 [10] Lindsay, J. H.: Coatings, and coating process for metals, ASM International, 1998, USA Reviewed by Patrick Chapon, GD Product Manager, HORIBA Jobin Yvon S.A.S. 16-18, rue du Canal - 91165 Longjumeau cédex - France

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved. High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

In: High Performance Coatings for Automotive and Aerospace… ISBN: 978-1-60876-579-9 Editor: Abdel Salam Hamdy Makhlouf, pp. 353-386 ©2010 Nova Science Publishers, Inc.

Chapter 12

CORROSION MONITORING USING IMPEDANCE DATA D. M. Bastidas,1* E. Cano,1 E. M. Mora,2 and J. M. Bastidas1 CENIM-National Centre for Metallurgical Research, CSIC, Avda. Gregorio del Amo 8, 28040 Madrid, Spain1 School of Naval Engineering, Polytechnic University of Madrid, UPM, Avda. Arco del Triunfo s/n, 28040 Madrid, Spain2

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT This chapter describes the electrochemical impedance method to study the corrosion behavior of coatings applied on metal substrates. The type of information that can be extracted from impedance measurements for three classic corrosion systems— copper/citric acid, stainless steel/NaCl solution, and carbon steel/concrete—are discussed. The transmission lines most frequently used in corrosion studies are analyzed— cylindrical and irregularly-shaped pores. A ladder network is used to model diffusion processes. A de-convolution method combined with a complex non-linear least squares (CNLS) program is showed to determine the individual components of the equivalent circuit analogue. Use of the Kramers-Kronig (KK) relationships as a tool for evaluating impedance data is described for both satisfying and non-satisfying KK rules. Stability analysis is performed using a pole-zero approach. Methods based on the transformation of transient data from the time domain into the frequency domain are described.

INTRODUCTION The electrochemical impedance spectroscopy (EIS) method has its origin in electrical engineering [1]. For this reason it is interesting to use electrical equivalent circuits (EEqC) constituted exclusively by resistors, capacitors, and inductors, not including electrical components defined by empirical functions of admittance.

*

Corresponding author‘s e-mail Address: [email protected] (D. M. Bastidas)

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

354

D. M. Bastidas, E. Cano, E. M. Mora et al.

EIS is a powerful method that allows one to quantify the three parameters defining a corrosion process: (i) the corrosion rate, through the charge transfer resistance (RCT) ( cm2), and, using Faraday‘s law, the penetration of the attack is estimated (m year1); (ii) the mass transport processes (diffusion) defined by the parameter (W) ( cm2 s1/2); and (iii) the electrochemical double layer capacitance at the metal/solution interface (Cdl) (F cm2). The EIS method is a minimally-invasive technique, having low cost and ease to use, and is frequently used in biomedical/biological applications to measure relative permittivity and conductivity of nerves and muscular tissues and to find the location of nerves under the human skin [2-4]. Other applications of the EIS technique are in implantable devices for recording/simulation and blood sensor devices [5]. The EIS method has been widely used to characterize the metal/solution interface, providing information about corrosion, chemical reactions, mass transport, adsorptiondesorption processes, and capacitance of the interfacial region. Impedance measurements have also been used to characterize aspects of materials such as their dielectric properties [6]. By applying a low-amplitude sine-wave voltage signal across a test system, v=Vmsin(t), it is possible to measure the frequency (f), f 

 where  is the angular 2

frequency (rad s1) and t the time. The phase shift or angle () and the amplitude of the resulting sine-wave current density (Im) is given by: i=Imsin(t+). If a metal/solution test system is time-invariant, then the impedance (Z) measured is timeindependent. Z is a vector defined as magnitude or modulus: Z 

Vm and  containing both Im

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

resistive (R) and reactive (C and/or L) components, both of which must be determined. Z is represented in the complex plane as: Z  Z' jZ" , where Z' and Z" are the real and imaginary parts, respectively, and j2=(1). Another complex formalism is admittance (Y), which is defined as: Y=Z1, and in the same way, Y  Y' jY". The procedure for interpreting corrosion impedance measurements is by using a mathematical model, or using an empirical EEqC analogue. The parameters can be estimated and compared with the experimental data [7-8]. This chapter describes the main uses of EIS method in corrosion studies (solid-state literature, aqueous, or polymer bibliography). Special attention will be paid to assessing the validity of impedance data through the use of Kramers-Kronig (KK) relationships.

FITTING PROCEDURE For analysis of the impedance data, a complex non-linear least squares (CNLS) procedure centered on minimizing the objective function (OF) with respect to the parameters Q of the model function, Z(,Q), is frequently used: N

 







' ' " " i , Q   Wi I Z exp i , Q  OF   Wi R Z exp  Z sim  Z sim i 1

2

2

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(1)

Corrosion Monitoring using Impedance Data

355

where N is the number of data points; WiR and WiI are the values of the weights associated with the real and imaginary part of the impedance, respectively, of the i-th experimental '

"

' value; Zexp and Zsim are the real experimental and fitted impedance data, respectively; Zexp

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

and Z"sim are the imaginary impedance; i is the i-th angular frequency data point; and Q is the set of M free and/or fixed parameters present in the simulated model [9-14]. Because use Eq. 1 is non-lineal, the fitting of such models requires an iterative approach, which may either fail to converge or converges to a local minimum. It is always desirable to carry out the fitting with two or more separate and quite differentiated initial parameter value sets. If two or more of them converge to the same final parameter set, it is reasonable to assume that the least squares solution has been obtained. The Levenberg-Marquardt (LM) algorithm [15-16], has been widely used to minimize OF, Eq. 1. The LM method is a compromise between the Gauss-Newton (GN) method and the steepest gradient descent method and is most useful when the parameter estimates are highly correlated, as is the case in the analysis of impedance data [11,17]. For analysis of complex circuits the starting values should be quite close to the optimal values, in order to avoid convergence to a secondary minimum, or worse. Because of this limitation, some researchers have preferred the more robust fit routine based on the Nelder and Mead (NM) algorithm [18], also known as the Simplex routine [19-20], which does not put restrictions on the initial parameter values. The main problem with the Simplex routine is that often a local minimum is encountered. Hence frequent restarts of the procedure are needed to find the absolute minimum value in OF, see Eq. 1. The Genetic Algorithm (GA) alleviates the secondary minimum problem [17,21-22]. Yang et al. have shown that the GA procedure is highly efficient in automatically establishing the EEqC parameters [17]. The major drawback is that the EEqC, or complete transfer function, must be known beforehand. Using the EEqC approach, this can be achieved by the de-convolution/subtraction procedure [23].

COPPER/CITRIC ACID SYSTEM In corrosion processes, it is often observed experimentally that there is not a single relaxation time (), but instead a distribution of relaxation times centered around a most probable value o. Plots of the  Z" vs. Z' of the metal/solution interface produce depressed semicircles whose centers lie below the real axis, which may be originated by surface roughness [24] and the presence of a porous corrosion product layer [25]. Commercial copper of 99.9% purity was used. A 0.5 M citric acid was used as the electrolyte component. The EIS method was used in the frequency range from 5.5 kHz to 1 mHz, using a logarithmic sweeping frequency of 5 steps/decade. Data was generated at the corrosion potential (Ecorr) (50 mV vs. SCE). EIS involved the imposition of a 10 mV amplitude sine-wave. A Solartron frequency response analyzer, model 1250, connected to an EG&G PARC potentiostat, model 273A, was used. The instrument was driven by EG&G 388 software. A conventional three-electrode cell configuration was used. The working electrode (WE) surface area was 1 cm2. The counter electrode (CE) was an AISI 316L stainless steel (SS) wire of large area, with a saturated calomel (SCE) reference electrode (RE). Fig. 1 shows

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

356

D. M. Bastidas, E. Cano, E. M. Mora et al.

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

an ‗ideal‘ behavior, not producing a depressed semicircle, and consequently it is possible to use the concepts of passive (concentrated) elements defining the equivalent circuit analogue (see inset). In corrosion experiments, the capacitive behavior is traditionally plotted in the first quadrant of the Nyquist plot. RS is the ohmic (frequency-independent) part of the impedance, i.e. the electrolyte resistance between the working (WE) and the reference (RE) electrode, which can be obtained from the abscissa axis intercepts of the semicircle at , RS36  cm2. RP can be obtained from the extrapolation of the diameter of the semicircle (broken line, Fig. 1) [26], RP4780  cm2, and may be associated with the copper corrosion rate. It may be assumed, as an approximation, that parameter RP is inversely proportional to the corrosion current density (icorr), according to the Stern-Geary formula:

Figure 1. Nyquist plot for copper in a 0.5 M citric acid solution, at 35 ºC after 96 h immersion. (o) Experimental. () Simulated. The values of the fitted parameters are: RS=36  cm2, Rp=4774  cm2, Cdl=24.12106 F cm2, and W=150.02  cm2 s1/2.

i corr 

where B is

B RP

(2)

ac , a and c are the anodic and cathodic Tafel slopes, respectively, 2.3 a   c 

generated by the direct current (DC) potentiodynamic polarisation curves experiment [27]. RP determination allows monitoring of the copper corrosion rate. Finally, using Faraday‘s law, the loss of copper per square decimeter can be calculated for the tested 96 h: 69 mg dm2 (3.5103 mg m2 day1, mmd). A value of 0.029 V was used for the B constant [28]. The Cdl parameter can be estimated from the relationship: Cdl=(RPtop)1, where top is the angular frequency, top=2ftop, ftop is the applied frequency pertaining to the ‗top‘ of the semicircle, where  Z" is a maximum (ftop=1.37 Hz, see Fig. 1), Cdl24106 F cm2. The

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

Corrosion Monitoring using Impedance Data

357

relaxation time or time constant of the corrosion process at high frequencies (HF) can be estimated from: HF=(top)1, HF0.12 s. HF can also be estimated from the formula: HF=RPCdl, HF0.12 s. The slope value of unity of the tail at low frequencies (Fig. 1) means that there is a diffusion process through the insoluble copper-citrate compound originated on the metallic substrate [28]. The time constant at low frequencies (LF) is given using the expression:

LF

 R  RP     s  W 2 

2

(3)

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

where W, the Warburg coefficient, is obtained from the Randles plot (see Fig. 2), then LF515 s.

Figure 2. Randles plot for copper in a 0.5 M citric acid solution at 35 ºC after 96 h immersion. (o) Experimental. () Simulated.

1 ) allows to obtain the Warburg coefficient (see  1 Fig. 2) [29]. The slope of the straight line yields W. Using a Z' vs. format W150   1 cm2 s1/2, and using a  Z" vs. plot W150  cm2 s1/2. The W parameter is defined  The Randles plot ( Z' or  Z" vs.

as:

W 

RT  1    n F2 A 2  C D  2

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(4)

358

D. M. Bastidas, E. Cano, E. M. Mora et al.

where R is the gas constant (8.314 J mol1 K1), T absolute temperature, n electrons per oxidized molecule, F the Faraday constant (9.649104 C mol1), D the diffusion coefficient for the diffusion controlling species, C its bulk concentration (mol cm3), and A is the electrode surface area. If C remains constant, then W becomes an exclusive function of D. Under this assumption, the greater the value of W, the greater will be the hindrance of the monitored diffusion process. It should be noted that the Sluyters-Rehbach and Sluyters plot

 vs. Y'

 , also allows to calculate the RP and W parameters [29]. Z' or  Z" vs.

1 

format also supplies information about the electrode reaction mechanism [30-32].

TRANSMISSION LINE MODELS

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

It is known that on rough and porous electrodes, on coplanar electrodes, on electrodes of a larger area than the counter electrode, and in general in crevice corrosion processes, a nonuniform distribution of the alternating current (AC) takes place on the electrode. In general, an EEqC that includes a finite number of resistors, capacitors, and inductors cannot adequately model experimental impedance data with the desired approximation. For this reason, different distributed electrical elements defined by empirical functions of impedance have been proposed in the literature (see below). For instance, the EEqC in Fig. 3 includes Warburg impedance (W) [29], which is a distributed electrical element described by the empirical function:

Figure 3. Electrical equivalent circuit (EEqC) with two distributed elements, CPE and ZW.

   W   w  2  j  The term

1 j

in Eq. 5 can also be written as

(5)

1 j , and consequently it is possible to 2

obtain the standardized expression of the Warburg impedance (W):

  W   w 1  j  

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

(6)

Corrosion Monitoring using Impedance Data

359

The other parameters in Fig. 3 are the electrolyte resistance (RS), constant phase element (CPE) and charge transfer resistance (RCT). The W in Eq. 5 includes a one-dimensional diffusional process through an infinite-length region, which is an unrealistic physical situation [33-34]. For this reason, Llopis and Colom introduced the concept of finite thickness layers [35], and the expression that models the impedance of the one-dimensional diffusional process (ZD) in a finite-length region is the distributed electrical element given by the following expression:

   Z D  2  w tgh j D  j 



where  D 



(7)

2 is the time constant of the diffusion process,  the diffusion layer thickness D

and D the diffusion coefficient. On the other hand, the CPE parameter in Fig. 3 is another distributed electrical element defined by an empirical function of admittance, given by the following expression:

YCPE  YP  j

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.



(8)

where YP is a real frequency-independent constant (F cm2 s(1), or 1 cm2 s). The dimensionless fractional  exponent (11) is related to the width of distribution of relaxation time [36-37]. On rough surfaces it is referred to as ‗capacitance dispersion‘. The impedance is not purely capacitive, but has a functional form as if the double-layer capacitance (Cdl) were frequency-dependent: Cdl()  Yp(j)1 [38]. When =0, the CPE parameter is a resistor, R=1/YP; when =1 it is a capacitor, Cdl=YP; and when =(1), it is an inductor, L=1/YP. Finally, if =0.5, Eq. 8 can be written as: YCPE  YP

j , the CPE is the

Warburg admittance [39]. In this case, the relationship between the YP parameter, Eq. 8, and the Warburg coefficient, Eq. 5, is given by the expression:

w 

1

(9)

YP 2

It should be said that different approaches have been made in the literature concerning the conversion of the CPE parameter (YP) into a capacitance (Cdl) [40-41]:

Cdl  YP "m 

 1

(10)

where "m is the angular frequency at which the imaginary part of the impedance ( Z" ) is maximum in a Nyquist plot. Other authors have proposed [42-47]:

High Performance Coatings for Automotive and Aerospace Industries, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,

360

D. M. Bastidas, E. Cano, E. M. Mora et al.

Cdl   YP R CT 

1

.

(11)

FRACTAL MODEL FOR THE COPPER/HYDROCHLORIC ACID SYSTEM It is not unusual to find impedance data for a corroding metal where the Nyquist plot is not a perfect semicircle at high frequencies and the diffusion process defined at low frequencies differs from a typical Warburg behaviour. Deviations of this kind at high frequencies, often referred to as frequency dispersions, have been attributed to multiple or coupled reaction sequences, to roughening of the electrode [48], and to frequency-dependent ohmic resistance caused by a non-uniform charging of the double layer. Several attempts have been made in the past to explain the physical meaning of the deviation in the low frequency tail. The scarce information available in literature associates this phenomenon with fractal models [49-50]. Wang discussed the possibility of using a discretized ladder network as indicated in Fig. 4 and proposed by Schrama in an unpublished paper [51]. Schrama, according to Wang [51], demonstrated that by adequately choosing the parameters of Fig. 4, it is possible to obtain a network that has the behaviour of a constant phase with the frequency and models the generalized Warburg behavior (‗tail‘ angle from the real axis different from 45º). The choice of such parameters is as follows:

Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.

R1

R2

C1

Rn-1

C2

Rn

Cn-1

Cn

Figure 4. Ladder network to model an irregular pore.

R k  2h 

1    k     h   ko   k  1   

C k  2k  1h 1

  k  1    1    k  1   

(12)

(13)

where Rk and Ck are the parameters of Fig. 4;  denotes the gamma function;  is the Kronecker delta; k=0, 1,…,n; h is an arbitrary small number; and  has been defined above. For >0.5 when k increases, it is found that Rk increases and Ck decreases. In contrast, for 