Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation [1 ed.] 9781611224450, 9781608765782

Citation preview

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated, 2009.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

ENERGY SCIENCE, ENGINEERING AND TECHNOLOGY SERIES

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CORN STRAW AND BIOMASS BLENDS: COMBUSTION CHARACTERISTICS AND NO FORMATION

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 rendering legal, medical or any other professional services.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

ENERGY SCIENCE, ENGINEERING AND TECHNOLOGY SERIES Oil Shale Developments Ike S. Bussell (Editor) 2009. ISBN: 978-1-60741-475-9 Power Systems Applications of Graph Theory Jizhong Zhu 2009. ISBN: 978-1-60741-364-6 Bioethanol: Production, Benefits and Economics Jason B. Erbaum (Editor) 2009. ISBN: 978-1-60741-697-5

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Introduction to Power Generation Technologies Andreas Poullikkas 2009. ISBN: 978-1-60876-472-3 Handbook of Exergy, Hydrogen Energy and Hydropower Research Gaston Pélissier and Arthur Calvet (Editors) 2009. ISBN: 978-1-60741-715-6 Energy Costs, International Developments and New Directions Leszek Kowalczyk and Jakub Piotrowski (Editors) 2009. ISBN: 978-1-60741-700-2 Radial-Bias-Combustion and Central-Fuel-Rich Swirl Pulverized Coal Burners for Wall-Fired Boilers Zhengqi Li 2009. ISBN: 978-1-60876-455-6

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Syngas Generation from Hydrocarbons and Oxygenates with Structured Catalysts Vladislav Sadykov, L. Bobrova, S. Pavlova, V. Simagina, L. Makarshin, V. Julian, R. H. Ross, and Claude Mirodatos 2009 ISBN: 978-1-60876-323-8 (Softcover)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Corn Straw and Biomass Blends: Combustion Characteristics and NO Formation Zhengqi Li 2009. ISBN: 978-1-60876-578-2

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

ENERGY SCIENCE, ENGINEERING AND TECHNOLOGY SERIES

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

CORN STRAW AND BIOMASS BLENDS: COMBUSTION CHARACTERISTICS AND NO FORMATION

ZHENGQI LI

Nova Science Publishers, Inc. New York

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009 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. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works.

Copyright © 2009. 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 ISBN: 978-1-61122-445-0 (Ebook) Available upon request

Published by Nova Science Publishers, Inc.  New York

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

CONTENTS

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Preface

ix

Chapter 1

Introduction

1

Chapter 2

Kinetic Study of Corn Straw Pyrolysis

3

Chapter 3

Analysis of Coals and Biomass Pyrolysis

19

Chapter 4

Combustion Characteristics of Corn Straw and Biomass Blends and no Formation in a Fixed Bed

31

Combustion Characteristics and NO Formation for Biomass Blends in a 35–Ton-Per-Hour Travelling Grate Utility Boiler

67

Numerical Simulations of Biomass Combustion and NO Formation in a Fixed Bed

81

Chapter 5

Chapter 6

Acknowledgments

103

References

105

Index

109

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

PREFACE Corn is one of China's three major crops and its straw is rich in resources. The thermal decomposition of corn straw samples (corn stalks skins, corn stalks cores, corn bracts and corn leaves) were studied using thermogravimetric analysis. Assuming the addition of three independent parallel reactions, corresponding to three pseudocomponents linked to the hemicellulose, cellulose and lignin, two different three-pseudocomponent models were used to simulate the corn straw pyrolysis. Model parameters of pyrolysis were given. It was found that the threepseudocomponent model with n-order kinetics was more accurate than the model with first-order kinetics at most cases. It showed that the model with n-order kinetics was more accurate to describe the pyrolysis of the hemicellulose. The thermal decomposition of coals and biomass was studied using thermogravimetric analysis with the distributed activation energy model. Experiments were carried out on a one-dimensional bench combustion test rig. The bed temperature distribution and the mass loss of fuel and gas components such as O2, CO, CO2 and NO were measured in the bed. The influence of the parameters such as the primary air flow, corn stalk length, air preheating and fuel moisture on the combustion characteristics and NO formation is given. Measurements were taken for a 35-ton-per-hour biomass-fired travelling grate boiler. Local mean concentrations of O2, CO, SO2 and NO gas species and gas temperatures were determined in the region above the grate. For a 28-ton-per-hour load, the mass ratios of biomass fly ash and boiler slag were 42% and 58%, the boiler efficiency was 81.56%, and the concentrations of NOx and SO2 at 6% O2 were 257 and 84 mg/m3. For an 18-ton-per-hour load, the fuel burning zone was nearer to the inlet than it was for the 28-ton-per-hour load, and the contents of CO and NO in the fuel burning zone above the grate were lower. A numerical model of biomass combustion and NO formation is presented. Simulation results of bed

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

x

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

temperature, gas species concentration, combustion velocity and NO emission concentration agreed with experiment data for different primary air rates.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Chapter 1

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

INTRODUCTION Current energy consumption of biomass accounts for about 14% of total world energy consumption, second only to coal, oil, and natural gas, and developing countries account for 75% of the biomass utilization [1-3]. According to the statistics from 1998 to 2003, biomass resources in China were equivalent to 700 million tons of standard coal, of which about half was from straw [4]. Combustion is frequently used for energy conversion of biomass [5], and corn straw combustion in grate furnaces is widespread in China. Corn is one of China’s three major crops and its straw is rich in resources. Understanding the mechanism of corn straw combustion in a fixed bed will help the clean and efficient use of corn straw as a source of energy.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Chapter 2

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

KINETIC STUDY OF CORN STRAW PYROLYSIS The pyrolysis, basically a polymeric structure cracking process, converted the lignocellulosic materials into volatile fraction and char. The volatile fraction (gas or liquid, depending on its molecular weight) can be used as a fuel or as a chemical synthesis source. On the other hand, the solid fraction presented several applications, like as a domestic fuel, in the production of activated carbon, or as a reducing agent in metallurgy [6, 7]. Understanding pyrolysis kinetics was important for the effective design and operation of the thermochemical conversion units. It was always a fundamental step for these conversion processes. Thermoanalytical techniques, in particular thermogravimetric analysis (TGA) and derivative thermogravimetry (DTG), allowed this information to be obtained in a simple and straightforward manner. Several researchers [8-12] have intensively investigated the kinetics of biomass pyrolysis. The pyrolysis process of the biomass was interpreted in terms of a simple first-order reaction with Arrhenius kinetics by Aiman and Stubington, while Thurner and Mann have approximated the pyrolysis process by two competitive first-order reactions. Thermogravimetric studies showed that each kind of biomass had unique pyrolysis characteristics, by virtue of the specific proportions of the components present in it. There was no detectable interaction among the components during pyrolysis. Orfao et al. [13] assumed that the biomass components react independently and, therefore, that the global thermal behaviour reflected the individual behaviour of the components, weighed by the composition. Meszaros et al. [14] advised to use a three-pseudocomponent model to simulate biomass pyrolysis. The objective of this study was to investigate the pyrolysis of different parts of corn straw and to determine the decomposition kinetics by comparison of

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Zhengqi Li

4

experimental and modeled data. The data were analyzed using two different threepseudocomponent models to simulate corn straw pyrolysis.

2.1. Experimental 2.1.1. Materials Corn straw was collected from Harbin farm. The following four samples were prepared: (a) corn stalks skins (corn stalks without their cores); (b) corn stalks cores (corn stalks without their skins); (c) corn bracts; and (d) corn leaves. All samples were meshed to small particles with range from 0.60 to 1.0 mm. The average sizes of all samples closed to 0.77 mm. Table 1 gives the proximate and ultimate analysis of these samples. Four of the samples had similar carbon, hydrogen, oxygen, volatile and fixed carbon values but had differences in the content of ash, nitrogen and sulphur on a dry basis and on an ash-free basis [15]. Table 1. Proximate and ultimate analysis. Corn stalks skins

Corn stalks cores

Corn bracts

Corn leaves

10.00

11.60

11.00

11.00

83.61 16.39

84.95 15.05

84.78 15.22

83.68 16.32

2.44

1.80

2.73

6.99

C

49.38

49.54

48.02

49.28

H O N S

5.84 44.28 0.48 0.03

6.03 43.72 0.65 0.06

5.78 45.51 0.62 0.07

5.67 44.12 0.87 0.06

17.19

17.60

16.52

16.31

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Samples

Proximate analysis (wt. %) (as dry and ash free basis)

Ultimate analysis (wt. %) (as dry and ash-free basis)

Moisture (as received basis) Volatiles Fixed carbon Ash (as received basis)

Gross calorific value(MJ kg-1)

2.1.2. Equipment and Procedures The thermogravimetric experiments were carried out in a Mettler-Toledo TGA/STDA851. The inert gas used for the pyrolysis was nitrogen with a flow rate of 150 ml min-1. In this work, heating rates of 20, 50 and 100 K min-1 were selected for these samples. Three thermogravimetric experiments were carried out for each biomass sample in the entire range of biomass decomposition, 40 to

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Kinetic Study of Corn Straw Pyrolysis

5

about 900 ºC. A blank experiment was conducted to exclude a buoyancy effect. The initial mass of the samples used was in a range of 14–40 mg.

2.2. Analytical Methods DTG curves of biomasses frequently contain shoulders and tailing, as shown below in this work. This indicated that more than one reaction was involved and that the biomass consisted of components with different reactivity for pyrolysis. To consider this, the approach of Varhegyi et al. [16] was chosen to describe the overall decomposition by means of independent parallel reactions and thus by independent components. In this work, it was assumed that the biomass consists of three pseudocomponents, and the pyrolysis rate was then described by 3 3 ⎛ E ⎞ da da = ∑ ci i = ∑ ci Ai exp⎜⎜ − A,i ⎟⎟ f (ai ) dt i=1 dt i =1 ⎝ RT ⎠

(1)

The variable a is the degree of transformation. The subscripts i represent

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

the different pseudocomponent of the biomass. Parameters ci , E and A are the coefficient, express the contribution of the partial processes to the overall mass loss, the activation energy and the pre-exponential factor and R is ideal gas constant. This assumption of three pseudocomponents was consistent with the nature of most biomasses which were often composed of hemicellulose, cellulose and lignin. In fact, three pseudocomponents represented a pool of fractions of the main biomass components. This was the definition of pseudocomponent used in previous studies [16, 17] and applied again in this work. In this research, to explain analysis result more clearly, hemicellulose, cellulose and lignin were used to represent three pseudocomponents respectively. Depending on the reaction order n , the following kinetic equations were used for each pseudocomponent: ⎛ E ⎞ dai n = Ai exp⎜⎜ − A,i ⎟⎟(1 − ai ) i dt ⎝ RT ⎠ ⎧ A ⎛ E ⎞ = Ai exp⎜⎜ − A,i ⎟⎟ exp⎨− i RT ⎝ ⎠ ⎩ β

∫

T

T0

⎛ E ⎞ ⎫ exp⎜⎜ − A,i ⎟⎟dT ⎬ ⎝ RT ⎠ ⎭

ni = 1

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

(2)

6

Zhengqi Li ⎛ E ⎞ dai n = Ai exp⎜⎜ − A, i ⎟⎟(1 − ai ) i dt RT ⎝ ⎠ ⎛ E ⎞⎧⎪⎛ (n − 1)Ai = Ai exp⎜⎜ − A,i ⎟⎟⎨⎜⎜ i β ⎝ RT ⎠⎪⎩⎝

⎛ E A, i ⎞ ⎞ ⎫⎪ ∫T0 exp⎜⎜⎝ − RT ⎟⎟⎠dT ⎟⎟⎠ + 1⎬⎪ ⎭ T

ni / (1− n i )

ni ≠ 1 with ai =

(3)

(mio − mi ) (mi − mash,i )

(4)

where β is heating rate (K min-1), mi is the mass of the solid and the subscripts

io and ash, i refer to the initial and residual amounts, respectively. The unknown parameters of the model were determined by the evaluation of the experimental data by non-linear square fitting. The quality of the fit was calculated by Eq. (5), Parameter k is the amounts of the data points, calc (da / dt )exp k is the experimentally observed value measurement, and (da / dt )k

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

is the calculated value obtained by numerical solution of the equation with the given set of parameters. calc 2 ⎛ k ⎡ da exp ⎞ ⎛ da ⎞ ⎤ ⎛ ⎞ ⎜ ⎟ − k ⎜ ⎟ ⎥ ⎢⎜ ⎟ ⎜⎜ ∑ ⎟⎟ ⎝ dt ⎠ k ⎥⎦ i =0 ⎢ ⎣⎝ dt ⎠ k ⎝ ⎠ Fit = 100 exp (− da / dt )max

1/ 2

(5)

2.3. Results and Discussion 2.3.1. General Pyrolysis Behaviour of Corn Straw Samples The corn straw samples revealed large differences in their decomposition behaviour (Fig. 1). The samples of corn stalks skins and cores showed one peak at about Tpeak = 635 K with a very light shoulder at about 580 K in the DTG curves. The samples of corn bracts and leaves showed two clearly pronounced peaks, one at about 580 K and one at 640 K. Normally, DTG curves of biomass exhibited a peak at high temperature mainly due to the pyrolysis of the cellulose and a shoulder or a peak at lower temperatures that can be attributed to the pyrolysis of

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Kinetic Study of Corn Straw Pyrolysis

7

the hemicelluloses part like in the case of corn straw samples in the Fig. 1. Compared to the DTG curve of corn leaves, the main decomposition of corn stalks cores was finished about 78 K and that of corn stalks skins and corn bracts about of 34 K earlier. The beginning of the main decomposition regime (about 500 K) was very similar for the samples of corn stalks skins, cores and corn leaves and that of the corn bracts was about 12 K higher as compared to them. The maximum pyrolysis rate were 4.29×10-3, 3.84×10-3, 3.34×10-3 and 3.16× 10-3 s-1 and the temperature corresponding to the maximum pyrolysis rate of these

samples were 632, 634, 640 and 643 K, for corn stalks cores, corn stalks skins, corn bracts and corn leaves, respectively.

0.0045

Corn stalks skins Corn stalks cores Corn bracts Corn leaves

0.0040 0.0035

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

-1

da / dt [s ]

0.0030 0.0025 0.0020 0.0015 0.0010 0.0005 0.0000 400

500

600

700

800

900

1000

Temperature [K]

Figure 1. DTG curves of corn straw samples (20 K min-1).

The results of experiments with different heating rates are shown exemplarily for corn stalks skins in Fig. 2. It was clear from these DTG profiles that considerable different trends in the rates of mass losses took place when heating rate changed between 20 and 100 K min-1. The maximum pyrolysis rate increased with heating rate increasing and the temperature at the peak pyrolysis rate also increased from 635 K at 20 K min-1 to 654 K at 100 K min-1. The phenomenon

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

8

Zhengqi Li

related to this important change in the maximum pyrolysis rate can be interpreted by the fact that biomass has a heterogeneous structure and possess a number of constituents. These constituents gave their characteristic individual decomposition peaks in definite temperature ranges in the pyrolysis process. When heating rate was sufficiently low during pyrolysis, most of these peaks can be seen as small broken lines or vibrations. However, at high heating rates separate peaks did not arise because some of them were decomposed simultaneously, and several adjacent peaks were united to form overlapped broader and higher peaks [18].

0 .0 2 0 0 .0 1 8

β

= 1 0 0 K m in

β

= 5 0 K m in

-1

0 .0 1 6

0 .0 1 2

-1

da / dt [s ]

0 .0 1 4

0 .0 1 0 0 .0 0 8 0 .0 0 6

β

= 2 0 K m in

-1

-1

0 .0 0 4

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0 .0 0 2 0 .0 0 0 400

500

600

700

800

900

1000

T e m p e r a tu r e [K ]

Figure 2. DTG curves of corn stalks skins at different heating rates.

2.3.2. Models to Simulate Corn Straw Pyrolysis The TG experiments at different heating rates were used to calculate mechanism parameters. Assuming the addition of three independent parallel reactions, corresponding to three pseudocomponents linked to the hemicellulose, cellulose and lignin, two different three-pseudocomponent models to represent biomass pyrolysis were studied. The simplest form of the three-pseudocomponent model was the approximation of the three individual reaction steps by a first order reaction (denoted here as model I). For comparison, the simulation was also conducted by the three-pseudocomponent model with ni ( i = 1, 2, and 3) as an additional free fitting parameter (model II). The higher complexity of model II may be interpreted as a heterogeneous decomposition inside the particle, where the reaction rates vary with an increasing conversion. Values of n different from

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Kinetic Study of Corn Straw Pyrolysis

9

one can also be due to the presence of a heterogeneous material, to the variation of the reaction surface and to the influence of diffusion effects [17].

0.0040

C orn stalks skins

0.0035

-1

da / dt [s ]

0.0030

M odel Ⅱ

0.0025 0.0020

M odelⅠ

0.0015

Experim ental data 0.0010 0.0005 0.0000 400

500

600

700

800

900

1000

0.0040

C orn stalks cores

0.0035 0.0030 -1

da / dt [s ]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

T em perature [K ]

M odelⅡ

0.0025 0.0020

M odelⅠ

0.0015

E x perim ental data 0.0010 0.0005 0.0000 400

500

600

700

800

900

T em perature [K ]

Figure 3. Continued on next page.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

1000

10

Zhengqi Li

0.0040

C orn bracts

0.0035 0.0030 -1

da / dt [s ]

M odel Ⅱ 0.0025 0.0020

M odelⅠ

0.0015

Experim ental data 0.0010 0.0005 0.0000 400

500

600

700

800

900

1000

Tem perature [K]

0.0030

C orn leaves

0.0025

M odel Ⅱ

-1

da / dt [s ]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0.0035

0.0020

M odel Ⅰ 0.0015

Experim ental data 0.0010 0.0005 0.0000 400

500

600

700

800

900

1000

Tem perature [K]

Figure 3. Comparison of simulation of pyrolysis of corn straw by different models (20 K min-1).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Kinetic Study of Corn Straw Pyrolysis

11

Fig. 3 indicated that the pyrolysis of the investigated four corn straw samples were quite well described by models I and II, whereby a slightly better fit was obtained by model II where we had three more fitting parameters (values of the three reaction orders of the three individual components). Especially, from Fig. 3, it can see that the model II had most satisfied simulation result at hemicellulose decomposition part. Simply setting reaction scale of hemicellulose as one in model I was not precise to simulate pyrolysis process. The parameters used for the simulation are given in Table 2. The experiment data of corn stalks skins at different heating rates was taken as an example to compare the two models (see Fig. 4). In Fig. 4 not only the overall rate of pyrolysis was shown, but also the individual rates of the three pseudocomponents, which may be regarded as hemicellulose (component 1), cellulose (component 2), and lignin (component 3). For the second pseudocomponent (probably cellulose), a reaction order of one was also optimal for model II, which was in agreement with the results of Meszaroset et al. [14] and Hu et al. [17]. The activation energy of cellulose was in a range of 175-202 kJ mol-1 (model II) and 148-186 kJ mol-1 (model I), which was agreement with literature data (128-263 kJ mol-1 [17]). The formal reaction of lignin was the component responsible for the tailing. The reaction order varies in a relative wide range (1.3–3.7, Table 2). The reason was probably that during pyrolysis the surface with the active centers for decomposition changes. The activation energy of lignin was in a range of 30-52 kJ mol-1 (model II) and 27-33 kJ mol-1 (model I), which was good agreement with literature data (35-65 kJ mol-1 [16], 34-36 kJ mol-1 [19]). Hemicellulose (first peak) was usually thought to follow a first-order kinetic. In this study, a reaction order of hemicellulose in range of 1.1-2.1 was determined (Table 2). Meszaroset et al. [14] also used the n th order kinetics model to simulate the partial reactions. They limited the iteration to n < 2 in order to avoid unusual peak shapes at higher values of n . They found an optimal value for n of two for Hemicellulose. This value was similar to the values determined in this work for corn stalks cores (2.1). The other investigated samples had lower values (1.1-1.3). The activation energy of hemicellulose was in a range of 98-141 kJ mol1 (model II) and 96-136 kJ mol-1 (model I), which was agreement with literature data (105-110 kJ mol-1 [16], 154-165 kJ mol-1 [19]). Obviously, heterogeneous chemical and physical characteristics of hemicellulose in these samples affected the pyrolysis.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated, 2009. ProQuest

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 2. Model parameters of pyrolysis to corn straw samples.

Sample

β (K min-1)

20

Corn stalks skins

50

100

Model (n=1) Component Ci

A (s-1)

E (kJ mol-1)

1

0.17

1.0E+07

100

2

0.43

3.0E+13

186

3

0.34

1.6E+00

1

0.23

2

Model (n≠1) Fit (%)

Ci

A (s-1)

E (kJ mol-1)

N

0.19

3.5E+09

126

1.3

0.45

7.5E+14

202

1.1

32

0.33

11E+00

42

1.3

1.0E+07

98

0.13

4.1E+09

126

1.3

0.47

3.9E+13

184

0.43

7.4E+14

199

1.1

3

0.28

2.5E+00

30

0.42

21.5E+00

39

1.3

1

0.20

1.4E+07

98

0.19

3.4E+09

124

1.3

2

0.48

3.2E+13

183

0.46

7.8E+14

200

1.1

3

0.30

3.6E+00

29

0.33

3E+01

39

1.3

2.2

2.4

2.7

Fit (%)

1.8

2.4

2.2

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated, 2009. ProQuest

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 2. (Continued).

20

Corn stalks cores

50

100

20

Corn bracts

50

100

1

0.20

1.5E+07

98

0.34

2.0E+07

101

2.1

2

0.54

3.0E+11

161

0.47

8.0E+14

202

1.1

3

0.25

6.5E-01

28

0.19

6.5E+01

48

3.7

1

0.22

1.7E+07

96

0.40

1.2E+07

98

2.1

2

0.57

3.0E+11

157

0.42

3.6E+14

198

1.1

3

0.16

2.0E+00

31

0.18

1.6E+02

52

3.7

1 2

0.19 0.44

1.7E+07 1.3E+11

97 155

0.37 0.30

1.2E+07 7.4E+14

98 202

2.1 1.1

3

0.18

6.0E+00

33

0.26

2.2E+02

49

3.7

1

0.32

3.0E+10

136

0.33

8.0E+10

140

1.2

2

0.43

1.0E+13

181

0.46

1.9E+13

184

1.1

3

0.24

1.0E+00

32

0.19

0.7E+00

30

2.0

1

0.37

3.4E+10

133

0.39

8.3E+10

137

1.2

2

0.44

1.9E+13

180

0.45

1.9E+13

180

1.1

3

0.17

1.2E+00

29

0.16

3.0E+00

35

2.0

1

0.43

3.1E+10

136

0.46

8.9E+10

141

1.2

2

0.43

2.0E+13

184

0.45

1.9E+13

184

1.1

3

0.14

3.0E+00

31

0.09

5.0E+00

35

2.0

1.6

1.2

1.6

1.5

2.4

3.2

1.4

1.0

2.0

1.5

2.7

3.0

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated, 2009. ProQuest

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 2. (Continued).

20

Corn leaves

50

100

1

0.27

3.5E+09

127

0.28

6.8E+09

130

1.1

2

0.46

3.5E+10

152

0.49

6.4E+12

179

1.3

3

0.24

4.2E-01

27

0.23

9.0E+00

44

2.6

1

0.25

3.4E+09

123

0.29

6.4E+09

126

1.1

2

0.47

3.6E+10

148

0.47

6.7E+12

175

1.3

3

0.20

1.5E+00

31

0.22

4.0E+01

50

2.6

1

0.29

3.1E+09

126

0.35

5.7E+09

129

1.1

2

0.50

3.5E+10

151

0.52

6.0E+12

179

1.3

3

0.21

3.1E+00

32

0.13

5.8E+01

52

2.6

2.0

2.3

1.8

1.5

2.7

2.8

Kinetic Study of Corn Straw Pyrolysis β =20 K min

-1

ModelⅠ : Three pseudocomponent model With n=1

0.0040

Calculated data

0.0035

Component 2

0.0030

da / dt [s-1]

15

0.0025

Experimental data 0.0020 0.0015 Component 1

Component 3

0.0010 0.0005 0.0000 400

500

600

700

800

900

1000

β =20 K min

Calculated data

0.0035 0.0030 -1

-1

ModelⅡ : Three pseudocomponent model With n≠ 1 n1=1.3

0.0040

da / dt [s ]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Temperature [K]

n2=1.1 n3=1.3

0.0025

Component 2

0.0020

Experimental data 0.0015

Component 1 Component 3

0.0010 0.0005 0.0000 400

500

600

700

800

900

Temperature [K]

Figure 4. (Continued)

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

1000

16

Zhengqi Li β =5 0 K min

0.010

Experimental data

-1

ModelⅠ : Three pseudocomponent model With n=1

-1

da / dt [s ]

0.008

0.006

Calculated data

Component 1

Component 2

0.004

Component 3

0.002

0.000 400

500

600

700

800

900

1000

Temperature [K]

β =50 K min

Experimental data 0.008

ModelⅡ : Three pseudocomponent model With n≠ 1 n1=1.3 n2=1.1

-1

da / dt [s ]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0.010

-1

n3=1.3 0.006

Component 1

Calculated data Component 2

0.004

Component 3 0.002

0.000 400

500

600

700

800

900

Temperature [K]

Figure 4. (Continued)

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

1000

Kinetic Study of Corn Straw Pyrolysis β =100 K min

17

-1

0.020 0.018

Experimental data

da / dt [s-1]

0.016

ModelⅠ : Three pseudocomponent model With n=1

0.014 0.012 0.010

Calculated data Component 1 Component 2

0.008 0.006 0.004

Component 3

0.002 0.000 400

500

600

700

800

900

1000

Temperature [K]

0.020 0.018

Experimental data 0.016

da / dt [s-1]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

β =100 K min

-1

ModelⅡ : Three pseudocomponent model With n≠ 1 n1=1.3

0.014

n2=1.1

0.012

n3=1.3 Calculated data

0.010 0.008

Component 1

Component 2

0.006

Component 3

0.004 0.002 0.000 400

500

600

700

800

Temperature [K]

900

1000

Figure 4. Calculation results of two models for corn stalks skins at different heating rates.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

18

Zhengqi Li

Table 2 gives the simulation results and corresponding deviations of two models. About 120 data points were used and the deviation between the experimental and calculated curves, defined by Eq. (5), varied from 1.0% to 3.0% for corn straw samples. By introducing higher order reactions in the model II, corresponding to DTG curves with higher shape indexes, it obtained slightly better fits. Compared the deviation to the model I, it showed that model II had better simulation result at most cases in mathematic meaning. Reaction order of these three pseudocomponents with heating rates of 20, 50, and 100 K min-1 was the same value when it was used to simulate the partial reactions by the n th order kinetics model. Compared to mechanism parameters, the activation energy of these three pseudocomponents had only slightly changes at different heating rates while the values for parameter ci and the pre-exponential factor had clear changes. Obviously, the contribution of the partial processes to the overall mass loss was different at different heating rates. This can be interpreted by the fact that biomass had a heterogeneous structure and the process of heterogeneous decomposition was different at different heating rates.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2.4 Conclusions Important differences on the pyrolysis behavior of corn straw samples (corn stalks skins, corn stalks cores, corn bracts and corn leaves) were observed on the DTG profiles. The maximum pyrolysis rates increased with the heating rate increasing. Assuming the addition of three independent parallel reactions, two different three-pseudocomponent models to represent biomass pyrolysis were studied. It was found that the models I and II gave a good fit with the experimental data at most cases. It showed that model II was more accurate to describe the pyrolysis of the hemicellulose.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Chapter 3

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ANALYSIS OF COALS AND BIOMASS PYROLYSIS The distributed activation energy model (DAEM) is widely used in the pyrolysis of coal as it produces good results. Miura verified the model which assumed a Gaussian distributed function and his results showed that the assumption was reasonable for the 19 types of coal used [20]. The DAEM was recently applied to biomass pyrolysis and a Gaussian distribution was assumed [21]. This assumption, however, needs to be validated. Alkali catalytic effects have been found to be responsible for the variation of distributed functions during the pyrolysis of different biomass species [22]. Table 3. Sample proximate and ultimate analyses. Samples Moisture Volatiles Proximate analysis (wt. %) Fixed (as air dry basis) carbon Ash C H Ultimate analysis (wt. %) O (as air dry basis) N S Gross calorific value(MJ kg-1)

Datong

Jindongnan

Corn-stalk skins

2.12 26.95

1.26 11.04

9.14 73.91

60.53

50.32

14.49

10.40 72.51 4.33 8.95 1.22 0.47 29.66

37.38 50.06 2.43 7.05 0.73 1.09 19.16

2.46 43.65 5.16 39.14 0.42 0.03 17.19

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

20

Zhengqi Li

This part validates and compares the distribution functions of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins and also reports the obtained values of the activation energies and frequency factors (k0) for the corresponding conversions. Corn-stalk skins were collected from a Harbin farm. Two types of coal were obtained from Datong and Jindongnan, respectively. Table 3 gives the proximate and ultimate analysis of these samples.

3.1. Analytical Methods The DAEM assumes that a number of parallel, irreversible and first-order reactions with different activation energies occur simultaneously. All the reaction activation energies had the same k0 at the same conversion rate. The activation energy had a continuous distribution. The release of volatiles is given by: ∞

1 - V/V*= ∫ exp(−

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0

k0 a

T

∫

0

e−E/(RT)dT) f (E)dE

(6)

where V is the volatile content at temperature T, V* is the effective volatile content, f(E) is a distribution curve of the activation energy that represents the difference in the activation energies of the many first-order irreversible reactions. The distribution function value has an important role in determining the contribution rate of the total volatiles released. The k0 corresponds to the E value and a is the heating rate. Eq. (6) can be simplified to Eq. (7) [23]. ∞

Es

Es

0

V / V * = 1 − ∫ f (E)dE = ∫ f (E)dE

(7)

Most researchers assume that f(E) has a Gaussian distribution with a mean activation energy E0 and a standard deviation σ. The k0 is assumed to be a constant or related to the temperature [11, 24]. Eq. 8 relates k0 to E [25]:

k 0 = α e Eβ where α and β are constants dependent on the reaction system.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

(8)

Analysis of Coals and Biomass Pyrolysis

21

In this part, Miura’s method was used to estimate f(E) and k0 of the Datong bituminous coal, Jindongnan lean coal and corn-stalk skins. The difference in E and k0 between coal and biomass was compared. The α and β values were also obtained. The method used is as follows:Both f(E) and k0 are obtained from three experiments using different heating profiles without assuming any functional forms for f(E) and k0. The calculation is done using Eq. (9) [20]:

ln (

k R a E 1 ) = ln ( 0 ) + 0.6075 − 2 T E RT

(9)

The procedure used to estimate f(E) and k0 is summarized as:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

1) Measure V/V* vs. T using at least three different heating rates. 2) Calculate the values of ln(a/T2) and -1/(RT) at the same V/V*. 3) Plot ln(a/T2) and -1/(RT) at the selected V/V* ratio and then determine the activation energies E from the slopes and k0 from the intercept. 4) Plot V/V* and E and differentiate the V/V* vs. E relationship by E to obtain f(E). This calculation shows that temperature plays an important role in the DAEM. The conversion rates increased in accordance with the increase of pyrolysis temperature. The conversion rates always corresponded to the pyrolysis temperature at all heating rates. Eq. (9) indicates that the temperature is important in the calculation of the abscissa and the ordinates. Different temperatures affected the point coordinates and thus the slope and intercept of the linear fit. This in turn affected E and k0.

3.2. Result and Discussion 3.2.1. Wt Vs. T Figure 5 shows the thermogravimetric curves of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins [26]. An increase in heating rate resulted in the thermogravimetric curves of coal and biomass shifting to higher temperatures. When the thermogravimetric weight loss of the sample was 2% to 95% the main pyrolysis intervals of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins are 205 ºC - 930 ºC, 210 ºC - 1075 ºC and 230 ºC - 540

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

22

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

°C, respectively. The initial pyrolysis temperature of Datong bituminous coal and Jindongnan lean coal is lower than that of corn-stalk skins, but the final pyrolysis temperature of Datong bituminous coal and Jindongnan lean coal is higher than that of corn-stalk skins. The pyrolysis temperature intervals of biomass are included in the coal intervals, but it cannot reflect that the activation energy changed following with the change of conversion rate when only refer to the intervals. The activation energy can be obtained at different times by the thermogravimetric curves according to the DAEM.

Figure 5. wt vs. t relationships.

3.2.2. Ln(A/T2) Vs. -1/(RT) Figure 6 shows the relationship between ln(a/T2) and -1/(RT) of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins at the selected ratio of V/V*. The activation energy is determined as the slope of the linear fit. More than three heating-rate measurements are needed to minimize errors. Activation energies were increasing continuously with increased conversion rates, as shown in Fig. 7. The k0 are obtained from the intercept using Eq. (9).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Analysis of Coals and Biomass Pyrolysis

23

-1 2 .0 Da t o n g *

V/ V = 0 . 1

-1 3 .0

0. 2

0. 3

0. 4

-1 3 .5

0. 5

0. 6

0. 7

2

l n ( a / T ) / ( s K)

-1

-1 2 .5

-1 4 .0 -1 4 .5 -1 5 .0 -0 .2 0

-0 .1 8

-0 .1 6

-0 .1 4

-0 .1 2

- 1 / ( RT ) / J / mo l

-12.4

-1 2

l n ( a / T ) / ( s K)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

-12.8

J i ndongnan

-13.2

*

V/ V = 0 . 2

-13.6

0. 3

0. 4

0. 5

0. 55

-14.0 -14.4 -14.8 -15.2 -0.16

-0.15

-0.14 -0.13 - 1 / ( RT ) / J / mo l

Figure 6. Continued on next page.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

-0.12

24

Zhengqi Li -1 1 st al ks

ski ns

*

V/ V = 0 . 1 0. 2

0. 3

0. 4

-1 3

0. 6

0. 8

2

l n ( a / T ) / ( s K)

-1

Co r n

-1 2

-1 4

-1 5 - 0 .2 3

- 0 .2 2

- 0 .2 1

- 0 .2 0

- 0 .1 9

- 0 .1 8

- 0 .1 7

- 1 / ( R T ) / J / mo l

Figure 6. ln(a/T2) vs. -1/(RT) relationships.

0.9 0.8 Co r n s t a l k s s k i n s

0.6

V/ V*

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0.7

Da t o n g

0.5 0.4

J i ndongnan

0.3 0.2 0.1 0.0 50

100 150 200 250 300 350 400 450 500 E/ k J / mo l

Figure 7. V/V* vs. E relationships.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Analysis of Coals and Biomass Pyrolysis

25

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

3.2.3. V/V* Vs. E Figure 3 shows the relationship between V/V* and E for three test samples. The release rates of volatiles with Datong bituminous coal and Jindongnan lean coal are significantly different at similar activation energies. Jindongnan lean coal undergoes a longer geographic formation and thus has more difficulty releasing volatiles. This was also noted by Maki in analyzing experimental data [27]. The conversion rate of corn-stalk skins were 1% - 84% at activation energy of 62-169 kJ/mol. The activation energies of coals and biomass are distinct at the same conversion rates. For example, the activation energies of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins were 313 kJ/mol, 383 kJ/mol and 132 kJ/mol, respectively, at conversion of 0.4. The activation energies of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins increased with a larger degree of conversion. 3.2.4. F(E) Vs. E Figure 8 was obtained by differentiating the V/V* vs. E relationship by E and also differentiating the curve obtained by a Gaussian fit, which was obtained from f(E). Datong bituminous coal and Jindongnan lean coal had f(E) peaks at 0.0039 mol/kJ and 0.0035 mol/kJ and had corresponding activation energies at 299 kJ/mol and 338 kJ/mol respectively. The f(E) of Datong bituminous coal and Jindongnan lean coal had an approximate Gaussian distribution and this was consistent with the study of 19 coals by Miura [20]. The peak f(E) of corn-stalk skins was 0.098 mol/kJ at an activation energy of 134 kJ/mol. This was 25 times that of Datong bituminous coal and 28 times that of Jindongnan lean coal. The f(E) of corn-stalk skins also had an approximate Gaussian distribution. 3.2.5. Lnk0 Vs. E Figure 9 shows the linear relationship between lnk0 and E for Datong bituminous coal, Jindongnan lean coal and corn-stalk skins. This data fits to Eq. (8). The linear fit is applied to three sets of experimental data. The constants α and β are listed in Table 4 and r is a correlation coefficient. No trend for α and β was observed. The k0 of the coals was in the range e19.5-e59.0 s-1 and e13.0-e55.8 s-1 for Datong bituminous coal and Jindongnan lean coal respectively. The k0 of the corn-stalk skins was in the range e10.8-e26.5s-1.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

26

Zhengqi Li

Figure 8. f(E) vs. E relationships.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Analysis of Coals and Biomass Pyrolysis

27

Table 4. The α and β constants. Coal/Biomass

α

Datong bituminous coal

0.105

0.996

3.8

0.119

0.993

2.0

0.146

0.996

e

Corn-stalk skins

e

Da t o n g

E x p e r i me n t a l

dat a

40

l nk

0

/s

-1

50

r

e

Jindongnan lean coal

60

β

11.1

Ca l c u l a t e d

30

dat a

20 150

200

250

300

350

400

450

500

E / k J / mo l 60 Ji ndongnan

50

E x p e r i me n t a l

dat a

0

/s

-1

40

l nk

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

100

30

Ca l c u l a t e d

dat a

20 10 100

150

200

250

300

350

400

450

E / k J / mo l

Figure 9. Continued on next page.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

500

28

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 9. lnk0 vs. E relationships.

3.2.6. V/V* Vs. T The f(E) and k0 values are used to obtain calculated curves for Datong bituminous coal, Jindongnan lean coal and corn-stalk skins (Fig. 10). The experimental data and calculated curves were similar for Datong bituminous coal, Jindongnan lean coal and corn-stalk skins without making assumptions for f(E) and k0. The use of the integral method is thus suitable for Datong bituminous coal, Jindongnan lean coal and corn-stalk skins.

3.3. Conclusion 1) When the DAEM is applied, experiments showed that the f(E) of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins have an approximate Gaussian distribution. 2) lnk0 is linear with E for Datong bituminous coal, Jindongnan lean coal and corn-stalk skins. No trend for the constants α and β was observed. 3) The activation energy of corn-stalk skins is significantly smaller than for Datong bituminous coal and Jindongnan lean coal at the same degree of conversion. The activation energies of Datong bituminous coal, Jindongnan lean coal and corn-stalk skins increase with a greater degree of conversion.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Analysis of Coals and Biomass Pyrolysis

29

0.8 Ex p e r i me n t a l Ca l c u l a t e d

V/ V*

0.6

Da t o n g

dat a dat a

1 0 0 ℃ / mi n 7 5 ℃ / mi n

0.4

5 0 ℃ / mi n 3 5 ℃ / mi n 2 0 ℃ / mi n

0.2

0.0 200

300

400 500 t / ℃

600

700

0.5

Ex p e r i me n t a l Ca l c u l a t e d

dat a dat a

1 0 0 ℃ / mi n

J i ndongnan

7 5 ℃ / mi n 5 0 ℃ / mi n

0.4

V/ V*

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

0.6

3 5 ℃ / mi n

0.3

2 0 ℃ / mi n

0.2 0.1 350

400

450

500

550

600

650

t / ℃ Figure 10. Continued on next page.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

700

30

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 10. V/V* vs. t relationships.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Chapter 4

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

COMBUSTION CHARACTERISTICS OF CORN STRAW AND BIOMASS BLENDS AND NO FORMATION IN A FIXED BED Different types of biomass have very different physical and chemical characteristics. The fuel properties and process conditions, such as fuel type, particle size, air flow rate and fuel moisture, affect the combustion characteristics, altering the heat generation, heat transfer and reaction rates in a complicated manner. Combustion of biomass such as wood and wheat straw in a fixed bed has been well studied [28-35]. The primary air flow rate, which determines the amount of oxygen available and heat transfer by convection, is the key process parameter of solid fuel combustion in a fixed bed. The combustion process is divided into three successive regimes, depending on the primary air flow rate: (1) oxygen-limited; (2) reaction-limited; and (3) extinction by convection [36]. Both experimental and simulated results indicate that the ignition front propagation rate of wheat straw initially increases rapidly with increased air flow rate and then decreases [35]. The experimental results indicate that the burning rate in a fixed bed of wheat straw and wood is linearly proportional to the air flow rate, which suggests that the whole process of combustion is oxygen-limited [29, 37]. However, temperature, gas composition and mass loss curves identify two distinct stages as combustion progresses in the bed: (1) ignition front propagation, and (2) char oxidation [33, 35, 37]. The experimental results indicate that the concentration of NO in the exhaust gas initially reaches a maximum and then decreases towards a stable value after the bed is ignited, and the NO exhaust concentration is insensitive to the bed temperature [34].

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

32

Zhengqi Li

4.1. Experimental Rig Figure 11 shows a diagram of the one-dimensional bench combustion test rig [38]. The reactor is a vertical cylindrical combustion chamber suspended from a weighing scale, with a mass measurement error of ± 1%. The chamber is 1.3 m high with a diameter of 180 mm. The combustor is axis-symmetric, thermally insulated by a 50 mm thick refractory wall, and surrounded by a thick layer of insulating material and an external casing. The grate at the bottom of the chamber consists of a stainless steel perforated plate that can withstand a temperature of the order of 1000 ºC, with approximately 95 holes of 7 mm diameter, representing a 14.7% open area. A gas burner placed at 750 mm above the grate and angled at 45° toward the fuel bed is used to initiate the burning process and maintain the freeboard flue gas temperature.

Scale

Flue gas exhauster

Combustion chamber Refractory material Stainless steel Insulating material a b c flexible pipe

a Burner

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

b Temperature measuering point

Rotameter

Data logging system

Propane

c Compressed air

Grate Air heater Gas analyser

Figure 11. The one-dimensional bench combustion test rig.

Thermocouples with the tip at the centre of the chamber are used to monitor the bed temperature at different heights (T1–T11; see Table 5). The measurement error of these thermocouples is ± 2.5% of actual temperature. A water-cooled gas-sampling probe is inserted into the bed at 400 mm above the grate. A Testo-350 gas analyzer is used to measure the concentrations of the gas species of interest, O2, CO, CO2 and NO, with measurement errors of: O2, ±0.8 vol%; CO, ±5 vol%; CO2, ± 5 vol%; and NO, ± 5 vol%.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

33

Table 5. Positions of bed temperature measurement points T1–T11 in the combustion chamber. Measurement point T1 T2 T3 T4 T5 T6

Position above the grate (mm) 1240 970 750 550 480 390

Measurement point T7 T8 T9 T10 T11

Position above the grate (mm) 300 210 120 30 -90

4.2. Combustion Characteristics of Corn Straw and NO Formation

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

4.2.1. Combustion Characteristics of Different Parts of Corn Straw and NO Formation 1) Corn Straw Samples and Test Cases Corn straw stalks, leaves and bracts have different densities and shapes. The following five samples were prepared: (a) corn stalks (corn straw without leaves or bracts); (b) corn leaves; (c) corn bracts; (d) hollow corn stalks (corn stalks without their cores); and (e) flaked corn stalks (corn stalks without their cores and cut into small pieces). Table 6 gives the proximate and ultimate analysis of these samples. Four of the samples have similar carbon, hydrogen, oxygen, volatile and fixed carbon values but have differences in the content of ash, nitrogen and sulphur on a dry basis and on an ash-free basis. All corn stalks were cut to a length of 50(± 5) mm. Twenty samples of straw were selected at random for measurement of the average mass ratio of stalks/leaves/bracts, which was 0.47:0.25:0.28. The same values were found for sample A (a mixture of stalks, leaves and bracts) and sample B (a mixture of leaves and bracts). The initial height of the biomass materials packed in the bed was 560 mm. Table 7 gives the initial mass, bulk density and primary air flow rates of the biomass materials for cases 1–12.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

34

Zhengqi Li Table 6. Proximate and ultimate analysis. Samples

Moisture (as received Proximate analysis basis) (wt.%) Volatiles (as dry and ash free Fixed carbon basis) Ash (as received basis) C Ultimate analysis H (wt.%) O (as dry and ash free N basis) S Gross calorific value(MJkg-1)

Corn whole stalks

Corn leaves

Corn bracts

Flaked corn stalks

10.50

11.00

11.00

10.00

84.46 15.54

83.68 16.32

84.78 15.22

83.61 16.39

2.96

6.99

2.73

2.44

49.83 5.91 43.59 0.60 0.07 17.22

49.28 5.67 44.12 0.87 0.06 16.31

48.02 5.78 45.51 0.62 0.07 16.52

49.38 5.84 44.28 0.48 0.03 17.19

Table 7. Test cases. Case no.

1

2

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Sample Mass (kg) Bulk density (kgm-3) Primary air flow rates (10-3kgm-2s1 )

3

4

5

6

7

8

9 Hollow corn stalks

10 Flaked corn stalks

1.00

0.90

70

Whole corn stalks

18

29

37

44

53

63

77

92

11

12

Sample A

Sample B

0.96

0.42

0.26

63

67

29

18

18

18

29

29

2) Bed Temperature, Mass Loss and Gas Composition Figure 12 shows the temperatures within the bed, the gas concentration at the grate top (y = 400 mm), and the mass loss rate as a function of time for case 2. As the ignition front propagated downwards and reached the thermocouple, the bed temperature at each thermocouple increased rapidly to 640 ºC. After the ignition front left each thermocouple, the bed temperature decreased due to heat loss from the gas to the bed wall and fresh layers of fuel. At about 990 s, the ignition front reached the bottom of the bed. During this period, the gas at the grate top (y = 400 mm) was about 10–12% CO2, 2–3% O2 and 5–7% CO. The higher concentration

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

35

of CO was due to the fuel-rich conditions with the combustion stoichiometric ratio of 0.17. (a) 1100 1000

800

0

Temperature ( C)

900

700 600 500 400

T4

300

T10

200 100 0 0

500

1000

1500

2000

2500

3000

3500

(b) 500

500 200

400

Bed height (mm)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Time(s)

500

300

100 300

400 200

600

400

300

700 800

100

200

900

0 0

500

1000

1500

2000

2500

3000

Time (s)

Figure 12. Continued on next page.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

3500

36

Zhengqi Li (C )2 5 800

NO o2 600

15

co2 400 10

NO (ppmV,dry)

conc. (%v,dry)

20

co 200

5

0 0

500

1000

1500

2000

2500

0 3500

3000

T im e ( s )

(d ) 10 0

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Mass left on bed (%)

80

60

c h a r o x id a tio n

40

20

0 0

500

1 000

150 0

200 0

250 0

300 0

350 0

T im e ( s )

Figure 12. Bed temperature history at different measurement points (a), bed temperatures (b), gas concentrations in the bed (c) and mass loss rate as a function of time (d) for case 2.

After the ignition front reached the grate, a char burnout front could be detected moving in the opposite direction, i.e. from the grate upwards. This was in agreement with other results reported for bench combustion of wheat straw [29]. Thus, the time that the measurement point T10 reached its peak bed temperature was earlier than T9. However, the peak bed temperature at T10 was lower than that at T9, because T9 was located at the central zone of residual char. As the residual char gradually burned out, the bed temperature at each measurement point declined rapidly. During this period, the concentration of CO decreased to 3.5% at 1250 s, increased to 7.5% at 1500 s, decreased slowly to 5% at 2000 s, then decreased quickly, while the concentration of CO2 remained constant and then started to decrease at 1500 s. A higher CO concentration was due to the fuel-

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Combustion Characteristics of Corn Straw and Biomass Blends…

37

rich condition with the combustion stoichiometric ratio of 0.83. Fig. 12(c) indicates that the concentration of O2 began to increase at 1500 s. The bed temperatures measured at six points inside the bed are plotted in Fig. 12(b). No plateau at 100 ºC indicating the drying process was observed. This was due to the low moisture content and high heat flux from above the burning zone. There was no plateau at 300 ºC to indicate the pyrolysis process. The temperature gradient was steeper in the burning zone than that in the heating zone. At about 1750 s, a high bed temperature zone at a temperature of 900–956 ºC (see Fig. 12(a) and (b)) was formed 120 mm above the grate at the central zone of the residual char. Because more than 90% of NOx from the straw combustion process was NO, only NO was measured. As the bed combustion temperature was below 1100 °C, fuel-NO was the main source of NO. Thermal NOx was neglected. The NO formation of the corn straw was similar to that observed in previous batch wheat straw firing [34]. Two zones of high exhaust NO concentration remained after the wheat and corn straw was ignited. The exhaust NO concentration reached a peak shortly after the straw at the gas-sampling port was ignited (at 540 s). This was due to the small amount of volatile formation from straw devolatilization and because the combustion occurred under the fuel-lean condition. The concentration then decreased (at 680 s) due to the combustion condition being changed from the fuel-lean to the fuel-rich condition, and the amount of volatiles from straw devolatilization increased. There was another peak of the NO exhaust concentration at 1030 s. The measured values at the gas sampling port indicated the flue gas composition resulting from the fuel combustion between the gas sampling port and the ignition front. As the ignition front went downward, the amount of burning fuel increased, causing the concentration of NO to increase. After the volatiles had gone and only char burned, a steep decrease in the NO concentration was detected, after which it decreased steadily to zero. During the ignition front propagation, the mass left on the bed decreased at a uniform rate. The fuel burning rate was 0.031 kg m-2 s-1. During the char oxidation, the mass loss slowed and the reaction time was about 2/3 of the whole combustion time. Generally, two main features at bed level were responsible for the inside bed temperature history. One was the sharp increase of bed temperature upon the arrival of the ignition front; another was the rise of bed temperature when combustion had proceeded for a significant period. The first stage was defined as the ignition front and the second was defined as the char oxidation front [35]. Since the heterogeneous char oxidation was relatively slow and oxygen was consumed first by the volatiles from the particles, carbonized particles remained

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

38

Zhengqi Li

above the ignition front. Therefore, the reactions of volatiles with air were dominant before the ignition front reached the grate. Once the ignition front reached the grate, the oxidation of the remaining char took place. The period before the ignition front reached the grate was defined as the ignition front propagation period, and the next period was defined as the char oxidation period. The ignition front propagation velocity can be derived from the experimental values of the distance between the thermocouples and the time for the ignition front to travel between two thermocouples. The time that the ignition front reached the grate can be obtained from the initial height of the biomass materials divided by the velocity of the ignition front. The bed temperature was defined as the mean value of the peak temperature in the front measured by the different thermocouples (T4 to T10). The burning rate was defined as the rate of mass loss of the bed per unit area and unit time. The finish of the combustion process in the bed was judged by no change in the mass of bed and the oxygen concentration up to 20%.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

3) Effect of the Primary Air Flow Rate Figure 13 shows the fuel burning rates during the ignition front propagation and during the char oxidation, the average burning rate and the bed temperature. The average burning rate ν av was defined as follows:

ν av = aν 1 + bν 2

(10)

where a is the ratio of the fuel mass burned during the ignition front propagation, b is the ratio of the fuel mass burned during the char oxidation , ν 1 and ν 2 are the fuel burning rates during the ignition front propagation and during the char oxidation, respectively.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends… 1000

0.044

950

0.040 0.036

900

Bed temperature

0.028

Average burning rate

o

0.032

850 800

Burning rate of char oxidation period Burning rate of ignition propagation period

0.024 0.020

Bed temperature ( C)

2

Burning rate (kg/(m s))

39

750

0.016

700

0.012 650 0.008 0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

600 0.10

2

Primary air flow rate(Kg/(m s))

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 13. The burning rate and bed temperature as a function of the primary air flow rate.

The experimental result indicated that the average burning rate and the bed temperature initially increased with increased air flow rate when the primary air flow rate was less than 0.069 kg m-2 s-1, and then decreased. It was similar to the results for whole wheat straw reported by Zhou et al. [35]. These cases were in fuel-rich conditions with the combustion stoichiometric ratio of 0.15-0.50. There are three sources of heat in the reaction front: the heat generated by combustion of volatiles or char oxidation; transfer of heat to unburned fuel; and the transfer of heat to the primary air by convection. With the low primary air flow rate, the amount of heat transferred to the primary air by convection was less than the heat generated by fuel combustion. The fuel beds were in a fuel-rich combustion condition; i.e. the amount of oxygen in the primary air was not sufficient to convert all of the combustible gas species and char. More volatiles such as tar, CO, etc. and char were consumed by oxygen with increased primary air flow rate. The heat produced by fuel combustion increased, which caused the average burning rate and the bed temperature to increase rapidly. However, once the primary air flow rate was greater than 0.069 kg m-2 s-1, the amount of heat transferred by convection to the primary air was larger than the amount of heat generated by fuel combustion. The average burning rate was reached and the bed temperature decreased.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

40

Zhengqi Li 1000

Ig n itio n p r o p a g a t io n p e r io d C h a r o x id a tio n p e r io d O v e r a ll a v e r a g e N O c o n c e n tr a tio n

2

NO Concentration, mg/m (%6O )

900

3

800 700 600 500 400 300 200 0 .0 1

0 .0 2

0 .0 3

0 .0 4

0 .0 5

0 .0 6

0 .0 7

0 .0 8

0 .0 9

0 .1 0

2

P r im a r y a ir f lo w r a t e ( k g / ( m s ) )

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 14. Variations of NO concentration with the primary air flow rate.

Figure 14 shows the mean concentration of NO during the ignition front propagation and during the char oxidation, and the overall average concentration of NO. The mean concentrations of NO during the ignition front propagation [NO]av,1 and during the char oxidation [NO]av, 2 and the overall average concentration of NO [NO ]av are defined as follows:

∫ ν [NO] t1

[NO]av,1

t0

=

O 2 = 6%

∫

t1

t0

∫ ν [NO] t1

O 2 = 6%

∫

t2

t1

∫ ν [NO] t0

O 2 = 6%

∫

t2

t0

dt

(12)

νdt

t2

[NO]av =

(11)

νdt

t2

[NO]av, 2 =

dt

dt

νdt

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

(13)

Combustion Characteristics of Corn Straw and Biomass Blends…

41

where t 0 is the start time of the ignition front propagation, t1 is the time when the ignition front reaches the bottom of the bed, and it is the start time of char oxidation, t 2 is the end time of char oxidation. ν is the fuel burning rate, and

[NO]O = 6% is the concentration of NO formed via O2 = 6%.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

2

During the ignition front propagation, the mean concentration of NO decreased with increased air flow rate until the primary air flow rate was 0.058 kg m-2 s-1 and then increased. The test cases were with combustion stoichiometric ratio of 0.11-0.45 during this period. The average burning rate increased with an increasing air flow rate when the primary air flow rate was less than 0.058 kg m-2 s-1 (see Fig. 13). Then, the volatiles were released more quickly and fuel combustion was under fuel-rich conditions. Thus, the concentration of NO decreased with increased primary air flow rate. When the primary air flow rate was grater than 0.058 kg m-2 s-1, the formation of NO increased because the fuel combustion was under relatively fuel-lean conditions. During the char oxidation, the mean concentration of NO decreased rapidly with increased air flow rate when the primary air flow rate was less than 0.058 kg m-2 s-1 and then increased. The fuel burning rate during the ignition front propagation increased with increased air flow rate when the primary air flow rate was less than 0.058 kg m-2s-1. Thus, the mass ratio of residual char to the raw materials decreased. Char N was less and the rate of formation of NO decreased rapidly with increased air flow rate. As the primary air flow rate increased from 0.058 kg m-2 s-1 to 0.092 kg m-2 s-1, the combustion stoichiometric ratio increased from 0.90 to 1.36, and the combustion condition changed from the fuel-rich to the fuel-lean condition. The mass ratio of residual char to raw materials increased to the extent that the amount of char-N became greater. Thus, the formation of NO increased with further increase of the air flow rate. These results showed that a higher average burning rate and a lower NOx emission can be obtained when the primary air flow rate is 0.058 kg m-2 s-1. 5) The Combustion Characteristics of Different Parts of Corn Straw Figure 15 shows the mass left on the bed and the average burning rate for whole corn stalks, hollow corn stalks, and flaked corn stalks at an air flow rate 0.018 kg m-2 s-1. During the ignition front propagation, the mass loss rate was largest for whole corn stalks and smallest for flaked corn stalks. During the char oxidation, the mass loss rate was the largest for flaked corn stalks. The average burning rate was largest for whole corn stalks. The reason is as follows: on the one hand, the experimental results from the TGA experiments (heating rate = 20 °C min-1) indicated that the activation energy and temperature at the peak

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

42

Zhengqi Li

degradation rate of whole corn stalks and samples of hollow and flaked corn stalks were similar to each other. On the other hand, the initial masses of whole corn stalks, hollow corn stalks, and flaked corn stalks were 1.00, 0.90, and 0.96 kg, respectively. The initial height of all samples packed in the bed was 560 mm. The bed porosity of flaked corn stalks was lower than that of hollow corn stalks, and thus the average burning rate for hollow corn stalks was higher. The heat capacity of the cores was lower than that of the skins of corn stalks. The heat capacity of whole corn stalks was less than that of hollow or flaked corn stalks. The lower the fuel heat capacity, the less energy needed to be transferred to the preheat zone to preheat fresh fuel. The temperature distribution in the bed of these three samples was determined. It showed that the time at which the drying front reached the bottom of the bed for whole corn stalks, hollow corn stalks and flaked corn stalks was 800 s, 700 s and 500 s, respectively. The time required for the drying front to reach the bed bottom was affected by the ignition front propagation velocity. The experimental results showed that the ignition front propagation velocity of flaked corn stalks was the highest. However, the burning rate for flaked corn stalks was the lowest. This can be explained by the channel effect becoming significant for these cases, which is described later. Thus, the time required for the drying front to reach the bed bottom for flaked corn stalks was less than that of whole corn stalks and hollow corn stalks. A high bed temperature zone was formed for these samples. The temperatures of these high bed temperature zones were 900–935 ºC, 800–850 ºC and 700–743 ºC, respectively. The high-temperature zone of whole corn stalks was a larger region at 120 mm above the grate and at about 1500–2000 s. For hollow corn stalks, it was a smaller region at 110 mm above the grate and at about 1600–1900 s. For flaked corn stalks, it was a smaller region at 100 mm above the grate and at about 1250–1750 s. This can be attributed to the Channel effect and different burning rate of these samples. Figure 16 shows the NO concentration profiles. The peak value of NO concentration for whole corn stalks，hollow corn stalks, and flaked corn stalks were 725 ppmv (at 1020 s), 1287 ppmv (at 930 s), and 2730 ppmv (at 730 s), respectively. This difference in the peak values of the NO concentrations can be attributed to the combustion stoichiometric ratio differences for these samples. At the primary air flow rate 0.018 kg m-2 s-1, the combustion stoichiometric ratios of whole corn stalks, hollow corn stalks, and flaked corn stalks were 0.14, 0.15, and 0.18, respectively. The NO concentration of flaked corn stalks had another peak (at 1880 s), but the other samples had only one peak. The reason is as follows: the mass ratios of the amount of burning fuel during the char oxidation to the total amount of burning fuel were 0.26, 0.24, and 0.44 for whole corn stalks, hollow

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

43

corn stalks, and flaked corn stalks, respectively. Thus, the largest amount of burning fuel was observed for flaked corn stalks. (a ) 100

c a s e 9 (h o llo w c o rn s ta lk s ) Mass left on bed (%)

80

c a s e 1 (w h o le c o rn s ta lk s ) 60

c a s e 1 0 (fla k e d c o rn s ta lk s )

40

20

0 0

500

1000

1500

2000

2500

3000

3500

4000

T im e (s )

Ig n itio n P ro p a g a tio n P e rio d A v e ra g e B u rn in g R a te C h a r O x id a tio n P e rio d

0 .0 3 5 0 .0 3 0 2

Burning rate (kg/ (m s))

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

(b ) 0 .0 4 0

0 .0 2 5 0 .0 2 0 0 .0 1 5 0 .0 1 0 0 .0 0 5 0 .0 0 0 case1

case9

(w h o le c o rn s ta lk s )

(h o llo w c o rn s ta lk s )

case10

(fla k e d c o r n s ta lk s )

Figure 15. Mass loss rate (a) and burning rate (b) as a function of time for cases 1, 9 and 10.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

44

Zhengqi Li 3000 2800 2600 2400

case 10(flaked corn stalks)

NO (ppmv,dry)

2200 2000 1800 1600

case 9(hollow corn stalks)

1400 1200

case 1(w hole corn stalks)

1000 800 600 400 200 0 0

500

1000

1500

2000

2500

3000

3500

4000

T im e (s)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 16. Variations of NO concentration for cases 1, 9 and 10.

6) Effect of the Sample Properties Corn stalks, sample A and sample B were used in an experiment with a primary air flow rate of 0.029 kg m-2 s-1. The results indicated that the mass of sample A and sample B changed more rapidly than that of whole corn stalks during the ignition front propagation. The burning rates of these three samples were 0.031 kg m-2 s-1, 0.034 kg m-2 s-1 and 0.037 kg m-2 s-1, respectively, during the ignition front propagation. The initial masses of these samples were 1.00, 0.42, and 0.26 kg, respectively, and all of them had an initial height of 560 mm. The bulk density of sample B was the lowest. At the same primary air flow rate, sample B and air mixed the best. Thus, the low bulk density of sample B may have enhanced its ignition beyond the level observed in whole corn stalks and sample A. On the other hand, the experimental results from the TGA experiments (heating rate = 20 ºC min-1) indicated that the activation energy of the leaves of corn straw was about 10 kJ mol-1 lower than that of whole corn stalks. This indicates that the leaves of corn straw devolatilized more easily than whole corn stalks. The mass ratios of leaves for whole corn stalks, sample A, and sample B were 0%, 25%, and 47%, respectively. Moreover, there were no corn stalks in sample B. Thus, there was a 9% increase in the ignition rate of sample B compared with whole corn stalks.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

45

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

The experimental results indicated also that the concentration of CO2 in the exhaust gases increased rapidly and was at about 10% when the samples A and B were ignited. The concentration of CO2 of whole corn stalks was zero until the ignition front reached the gas sample port (at about 450 s). Then, the CO2 concentration rose quickly and was about 11% during ignition front propagation. This was due to the ignition front propagation velocities for samples A and B were higher than that for whole corn stalks. The concentrations of CO for samples A and B were less than that of whole corn stalks. This may be attributed to channel formation in the fuel beds of sample A and sample B. Channel caused most of the air or gas flow to pass through passages with higher local void fraction or ‘short-cuts’ inside the bed [35, 37](Zhou et al., 2005; Changkook et al., 2006). The NO concentration distributions of whole corn stalks and sample A had two peaks, while that of sample B had only one peak. This can be attributed to the less initial mass and the higher burning rate of sample B. 7) Conclusions Corn straw combustion occurred in two stages: ignition front propagation and char oxidation. During ignition front propagation, the mass left on the bed decreased at a uniform rate. A high bed temperature zone was formed near the grate due to the large amount of residual char oxidation. The average burning rates increased rapidly with increased air flow rate when the primary air flow rate was less than 0.069 kg m-2 s-1, and then decreased. The mean concentration of NO decreased with increased air flow rate when the primary air flow rate was 0.058 kg m-2 s-1, and then increased. A higher average burning rate and lower NOx emission can be obtained when the primary air flow rate is 0.058 kg m-2 s-1. The average burning rate for whole corn stalks was the fastest. The temperatures in the high bed temperature zones of whole corn stalks, hollow corn stalks and flaked corn stalks were 900–935 ºC, 800–850 ºC and 700–743 ºC, respectively. The time taken for the drying front to reach the bottom of the bed for these samples was 800 s, 700 s and 500 s, respectively. The maximum concentrations of NO of whole corn stalks, hollow corn stalks and flaked corn stalks were 725 ppmv, 1287 ppmv and 2730 ppmv, respectively. The fuel burning rates of whole corn stalks, sample A (a mixture of corn stalks, leaves and bracts) and sample B (a mixture of corn leaves and bracts) were 0.031 kg m-2 s-1, 0.034 kg m-2 s-1 and 0.037 kg m-2 s-1, respectively. The maximum concentrations of CO and NO were clearly different for the three samples, and there was only one peak in the NO concentration distribution for sample B.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

46

Zhengqi Li

4.2.2. Effect of Corn Stalk Length The corn stalks were cut into five different lengths: 20, 30, 40, 50, and 70 mm [39]. The average diameter of corn stalk was 17 mm. Table 8 gives the proximate and ultimate analysis of the fuel. The initial mass was 1.1 kg and the bulk density of the fuel material in the packed state was around 80 kg m-3. The initial height of the fuel materials packed in the bed was around 540 mm. The primary air flow rate was 0.053 kg m-2 s-1 for all cases. Primary air was fed from the bottom of the reactor through the grate without preheating (20 ºC). Table 8. Proximate and ultimate analysis of corn stalk fuel.

Proximate analysis (wt. %) (as dry and ash free basis)

Ultimate analysis (wt. %) (as dry and ash free basis)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Gross calorific value (MJ kg-1)

Moisture (as received basis)

Volatiles

Fixed carbon

Ash (as received basis)

11.50

81.74

18.26

2.58

C

H

O

N

S

49.64

5.73

44.15

0.44

0.04

17.64

There was a significant difference in the combustion behaviours of different particle length fuels. Under the operating condition of the current investigation, the initial bed porosity was artificially changed and kept at a constant value for shorter and longer particles. However, more uniform packing conditions can be obtained in the fuel bed, and the void space between the particles was relatively small for shorter particles. The bed absorption of radiation flux was affected by the packing conditions. The radiation flux was more absorbed by a packing of shorter particles and hence penetrated a short distance in the bed; this can produce a thinner ignition front in the bed and affect the temperature and gas concentration profiles as a consequence. On the other hand, the smaller void space in the fuel bed for shorter particles can enhance the mixing rate of the volatile gaseous fuel with air flowing from under the grate, and shorter particles produce a higher burning rate (with the other conditions being the same) and hence higher combustion intensity in the gas phase [40]. Figure 17 shows the fuel length effect on the fuel burning rates during ignition front propagation and during char oxidation, the average burning rate (see Eq. (10)).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

47

0.030

2

Burning rate (kg/(m s))

0.035

0.025

0.020

0.015

0.010

0.005 20

30

40

50

60

70

Particle length (m m )

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 17. Effect of particle length on burning rate. Legend: (□), burning rate of ignition propagation period; (▲), average burning rate; (×), burning rate of char oxidation period.

The experimental result indicates that the average burning rate decreased with an increasing particle length and the 70 mm fuel produced the lowest average burning rate of 0.0258 kg m-2 s-1, compared to 0.0289 kg m-2 s-1 for the 20 mm fuel. Figure 18 shows the fuel length effect on the ignition front propagation velocity and average bed temperature. The ignition front propagation velocity may be calculated from the measured bed temperature. Ignition front propagation velocity Vf =S/T where S is the distance from the temperature measuring port T4 to T10 and T is the time taken for the ignition front to move from the temperature measuring port T4 to T10.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

48

Zhengqi Li 890 880 870 860 850

1.70

840

0

1.75

Temperature ( C)

Ignition front propagation velocity (m/hr)

1.80

830

1.65

820 810

1.60

800 20

30

40

50

Particle length (mm)

60

70

1100 1000

o

Temperature gradient ( C/min or C/ cm)

900 800

o

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 18. Effect of particle length on ignition front propagation velocity (□) and bed temperature (×).

700 600 500 400 300 200 100 20

30

40

50

Particle length (mm)

60

70

Figure 19. Transient and spatial temperature gradients at the ignition front. Legend: (□), ΔT/Δt (ºC/min); (×), Min/MaxΔT/Δt; (▲), ΔT/Δy (ºC/cm).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Combustion Characteristics of Corn Straw and Biomass Blends…

49

Ignition front propagation was controlled by the fuel burning rate and heat transfer to and from the flame zone. Thermal equilibrium, where the rate of heat loss equals the rate of heat generation by combustion, was a key factor to maintaining the flame zone propagation in the bed with approximately constant velocity. The experimental result indicates that the ignition front propagation velocity decreased with an increasing particle length and the 70 mm fuel produced the lowest ignition front propagation velocity of 1.60 m h-1 compared to 1.78 m h1 for the 20 mm fuel. Depending upon the particle length, the average bed temperature ranged form 814 to 881 ºC and it was seen that a fuel with a longer length produced a lower combustion temperature. Longer particles have slower devolatilization rates and more distributed heat transfers to nearby particles. The temperature profile provides additional information on the effect of particle length on the fuel burning rate. When the ignition front passed each thermocouple, it resulted in a sharp rise in temperature. The transient temperature gradient (ΔT/Δt) between 200 and 600 ºC at each thermocouple is shown in Figure 8. The spatial temperature gradient (ΔT/Δy) is obtained by dividing by the ignition front propagation velocities. This can be used as an indicator of overall downward heat flux at the ignition front. The two gradients decreased with increasing particle lengths, which suggested that the heat influx to the cold particles below the hot particles was more intense for shorter particles. It took about 39 s for the 20 mm particles to reach a temperature of 600 ºC from 200 ºC and about 69 s for 70 mm length particles. Figure 20 shows the fuel length effect on the CO concentration in the flue gases. A general trend was the declining CO concentration as the particle length increased. Shorter particles have a higher level of CO because of the high burning rates and resulted in fuel-rich combustion. Figure 21 shows the variation in NO concentration with particle length. It is seen the NO concentration increased with an increasing fuel length, from 88 mg m-3 (6% O2) for 20 mm to 115 mg m-3 (6% O2) for 70 mm. This was because the oxidation rate of the fuel N increased with longer particles lengths, which were under relatively fuel-lean conditions.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

50

Zhengqi Li

CO concentration, (%v,dry)

3.4

3.2

3.0

2.8

2.6

2.4

2.2

2.0 20

30

40

50

60

70

Particle length (mm)

3

NO Concentration, mg/m (%6O2)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 20. Variation in CO concentration with particle length.

115

110

105

100

95

90

85 20

30

40

50

Particle length (mm)

60

70

Figure 21. Variation in NO concentration with particle length.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Combustion Characteristics of Corn Straw and Biomass Blends…

51

The experiments demonstrate that shorter particles tend to produce an earlier ignition of the bed than longer fuel particles. This was consistent with the experimental results of different sizes of wood by Friberg and Blasiak [30] and Yang et al. [40]. The measured ignition front propagation velocity versus particle length (see Figure 18) was consistent with the findings of Friberg and Blasiak [30] and Yang et al. [40]. They demonstrated that the ignition front propagation velocity decreased as particle size increased. However, the current results were contrary to the findings of Gort [41]. He reported the ignition front propagation velocity for a larger particle increased in the size range of 10-30 mm. The measured bed temperature versus particle length (see Figure 18) was contrary to the finding of Yang et al. [40]. He demonstrated the bed temperature increased as particle size increased. Based on the above experiments, we can get following conclusions. The particle length has a significant influence on corn stalk combustion in a fixed bed. In general, the following conclusions can be drawn from the current study in terms of the fuel length effect: (1) The total burning time was shorter for shorter particles than that for longer particles under the same operating conditions. (2) Shorter fuel particles produced a higher flame temperature in the bed. The high-temperature zone for shorter particles was closer to the grate than that for longer fuel particles. (3) The variation with time of the flue gas O2, CO and CO2 concentrations was less intensive for longer particles during the ignition front propagation period and the char oxidation period. (4) In the packed bed, the heat influx to the cold particles below the hot particles was more intense for shorter particles. (5) Shorter fuel particles resulted in fuel-rich combustion and higher CO concentration, but the NO concentration increased with an increasing fuel length. (6) The average burning rate and ignition front propagation velocity decreased with an increasing fuel particle length.

4.2.3. Effect of Air Preheating and Fuel Moisture The corn stalks were cut to a length of 50 (± 5) mm. Table 8 gives the proximate and ultimate analysis of the fuel. To change the moisture level in the fuel, water was added to the corn straw samples to increase the moisture level to as high as 41.88%. The initial height of the fuel materials packed in the bed was around 540 mm. The primary air flow rate was 0.053 kg m-2 s-1 at room temperature for all cases. The primary air preheating temperature used in the experiments was 20, 60, 100 and 130 ºC respectively [42].

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

52

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

1) Effect of Air Preheating Figure 22 shows the effect of air preheating on the fuel burning rates during the ignition front propagation and during the char oxidation, the average burning rate. The experimental results indicated that the average burning rate increased with an increasing the primary air preheating temperature and the fuel produced the lowest average burning rate of 0.0259 kg m-2 s-1 with no preheated primary air (Tpa = 20 ºC), compared to 0.0319 kg m-2 s-1 with Tpa = 130 ºC.

Figure 22. Effect of air preheating on the fuel burning rate.

Figure 23 shows the air preheating effect on ignition front propagation velocity and the bed temperature. Ignition front propagation was controlled by the fuel burning rate and heat transfer to and from the flame zone. Thermal equilibrium where the rate of heat loss equals the rate of heat generation by combustion was a key factor to maintain the flame zone propagated in bed with an approximately constant velocity. The experimental results indicated that the ignition front propagation velocity increased with an increasing the primary air preheating temperature and the fuel produced the lowest ignition front propagation velocity of 1.63 m h-1 with no preheated primary air (Tpa = 20 ºC), compared to 2.84 m h-1 with Tpa = 130 ºC.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

53

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 23. Effect of air preheating on ignition front propagation velocity and bed temperature.

The bed temperature was the mean value of the peak temperature in the front measured by the different thermocouples (T4 to T10). Depending on the primary air preheating temperature, the bed temperature ranged from 847 to 864 ºC and it was seen that fuel with a higher primary air preheating temperature produced a lower peak flame temperature. 2) Effect of Fuel Moisture Figure 24 shows the effect of fuel moisture on the fuel burning rates. The effect of moisture was obvious. At a fixed primary airflow rate, drier fuel had higher average burning rate. For example, at 11.50% moisture the fuel average burning rate was 0.0259 kg m-2 s-1 while at 41.88% moisture it was only 0.0092 m-2 s-1, nearly 2.82 times lower. The reason is as follows: moisture evaporation consumed a great deal of heat and the evaporation time lengthened during the process of combustion. It resulted in the decay of ignition and lower chemical reaction rate.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

54

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 24. Effect of fuel moisture content on burning rate.

Figure 25 shows the residual mass loss rate, ignition front propagation velocity and the bed temperature at different fuel moistures. As fuel moisture was lower than 30.71%, a wetter fuel, surprisingly, had a lower residual mass loss rate and a higher bed temperature than that of a drier fuel at a fixed primary air flow rate. When the fuel moisture further increased to 41.88%, the residual mass loss rate underwent no further decrease and the bed temperature underwent no further increase, however. The experimental results also indicated that the ignition front propagation velocity decreased with an increasing the fuel moisture. The effect of moisture in the fuel can be explained in the following way. The char oxidation depended on three factors: the amount of formed char, the O2 availability and the temperature. A wetter fuel reduced the devolatilization rate and hence the O2 consumption by the burning of volatile gases. This made more O2 from the air supply available to the char oxidation in such a way that overweighed the effect caused by decreasing char formation, so the net effect was an increase in the char oxidation rate and decrease in the residual mass loss rate compared to a drier fuel. Some aspects should be taken into account to explain the variation of the ignition front flame temperature at a fixed primary air flow rate. On one hand, moisture in fuel tended to reduce the ignition flame front temperature due to the latent heat of evaporation. On the other hand, the combustion was in fuel-rich conditions. With the decrease of the ignition front propagation rate, more oxygen was available per volumetric mass of fuel and the wetter fuels had a higher char oxidation rate which can increase the flame temperature. When the fuel moisture further increased to 41.88%, the flame temperature was reduced by the latent heat

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

55

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

of evaporation in the ignition front. Therefore, the char oxidation rate was decrease and the residual mass loss rate was increase. This was similar to the results reported by Yang et al. [32].

Figure 25. Mass loss rate, ignition front propagation velocity and bed temperature as a function of moisture level in the fuel.

Wetter fuels have slow devolatilisation rate. The temperature profile provided additional information for the effect of fuel moisture on the fuel burning rate. When the ignition front passed each thermocouple, it resulted in a sharp rise of temperature. The transient temperature gradient (≥T/≥t) between 200 and 600 ºC at each thermocouple is shown in Fig. 26. This can be used as indicator of overall downward heat flux at the ignition front. The gradient suggested that the heat influx to the cold particles below the hot particles was more intense for drier fuel. It took about 48 s for 11.50% moisture to reach a temperature of 600 ºC from 200 ºC and about 245 s for 41.88% moisture. Figure 27 presents the CO and NO concentration as a function of moisture level in the fuel. CO concentration generally declined as the moisture level in the fuel increased. Drier fuels had a higher CO due to the high burning rates. The reason was that drier fuels resulted in fuel-rich combustion. The NO concentration decreased with an increasing the moisture level, from 102 mg m-3 (O2 = 6%) at 11.50% moisture to 50 mg m-3 (O2 = 6%) at 41.88% moisture.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

56

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 26. Transient temperature gradient at the ignition front.

Figure 27. Continued on next page.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

57

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 27. CO and NO emissions from the bed top as a function of moisture level in the fuel. (a) CO emission; (b) NO emission.

3) Conclusions Experiments have been carried out for corn straw in a one-dimensional bench combustion test rig and the effects of air preheating and fuel moisture was investigated. In general, the following conclusions can be drawn from current study /1)The total burning time was shorter under the higher primary air preheating temperature and the higher primary air preheating temperature produced a lower ignition front flame temperature in the bed. (2) The average burning rate and ignition front propagation velocity increased with an increasing primary air preheating temperature level. (3) The variation of the flue gas O2, CO and CO2 concentrations with time was more intensive at a higher primary air preheating temperature during the ignition front propagation period and the char oxidation period. (4) An increase in the moisture level in the fuel produced lower average burning rate and ignition front propagation velocity. (5) In the packed bed, the heat influx to the cold particles below the hot particles was more intense for the drier fuel. (6) As the fuel moisture was lower than 30.71%, with the increase of the fuel moisture, residual mass loss rate decreased and ignition front flame temperature increased at a fixed air flow rate. (7) Drier fuels resulted in fuel-rich combustion and higher CO concentration. The NO concentration decreased with an increasing the moisture level in the fuel.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

58

Zhengqi Li

4.3. Combustion Characteristics of Biomass Blends and NO Formation Biomass fuel (a mixture of bark and waste wood cubes) were cut to a length of 50 (± 5) mm. Table 9 gives the proximate and ultimate analysis results for the fuel. The initial mass was 3.3 kg and the bulk density of the fuel material in a packed state was around 260 kg m–3. The initial height of the fuel materials packed in the bed was around 500 mm. Primary air was fed from the bottom of the reactor through the grate with preheating (at a temperature of Tpa = 150°C). The primary air flow rates used in the experiments were 0.14, 0.17, 0.20, 0.23 and 0.26 kg m–2 s–1. Table 9. Proximate and ultimate analysis results for biomass blends.

Proximate analysis (wt. %) (as received basis)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Ultimate analysis (wt. %) (as received basis) Gross calorific value (MJ kg-1)

Moisture

Volatiles (as dry and ash free basis)

Fixed carbon

Ash

40.00

80.68

11.33

1.34

C

H

O

N

S

31.08

3.44

23.93

0.18

0.03

10.47

1) Mass Loss History Figure 28 shows the mass loss rate as a function of time for the biomass blend samples. After switching on the gas burner to initiate the burning process (t = 0 s), there was an initial ignition period before a steady combustion stage was reached. At a primary air flow rate of 0.14 kg m–2 s–1, the initial mass loss was slow, reaching a steady state at t = 480 s. At the primary air flow rate of 0.17 kg m–2 s–1, the initial mass loss was slightly quicker, reaching a steady state at t = 450 s. At the primary air flow rate of 0.20 kg m–2 s–1, the mass loss rate reached a steady state after t = 330 s. At the primary air flow rate of 0.23 kg m–2 s–1, the fuel had the shortest ignition period (390 s) before a fully steady combustion stage was reached. However, when the primary air flow rate was increased to 0.26 kg m–2 s– 1 , the initial mass loss was slightly slower, reaching a steady state at t = 450 s.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

59

100

Mass left on bed (%)

-2

-1

0.14 kg m s -2 -1 0.17 kg m s -2 -1 0.20 kg m s -2 -1 0.23 kg m s -2 -1 0.26 kg m s

80

60

40

20

0 0

1000

2000

3000

4000

5000

Time (s)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 28. Mass loss rates as a function of time for different primary air flow rates.

After the initial ignition period, a linear decrease in the total mass of the bed material was observed. This constant combustion period decreased from 2760 to 1890 s with an increased air flow rate from 0.14 to 0.23 kg m–2 s–1, and then increased to 2760 s at the primary air flow rate of 0.26 kg m–2 s–1. After the steady combustion stage, as the reaction time increased, the bed mass loss accelerated. The accelerated combustion period was 960, 720, 540, 570, and 510 s when the primary air flow rate was 0.14, 0.17, 0.20 0.23, and 0.26 kg m–2 s–1 respectively. After the accelerating combustion stage, the bed mass loss slowed as the reaction time increased. This final period of combustion lasted for about 790–1000 s before combustion was complete in these test cases. It was clear that the total burning time decreased with an increasing air flow rate when the primary air flow rate was less than 0.23 kg m–2 s–1, and then increased. 2) Temperature Profile in the Bed Figure 29 shows the bed temperature profile at different measurement points and the bed temperatures for a primary air flow rate of 0.17 kg m–2 s–1. Once the fuel was ignited by the gas burner, the ignition front propagated downwards and reached thermocouple T4 (Figure 29a), and the bed temperature at each thermocouple increased rapidly from room level to a peak value as high as 660°C.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

60

Zhengqi Li (a) 1300 1200

0

Temperature ( C)

1100 1000 900 800 700 600 500

T4

400 300 200

T10

100 0

1000

2000

3000

4000

Time (s)

500 500

500

400

600 Bed height (mm)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

(b)

5000

300

400

200 500

400

100

600 400

300

200

300 700

800 900 1000

100

0

1000

2000

3000

Time (s)

4000

Figure 29. Bed temperature history at different measurement points and bed temperature distribution.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Combustion Characteristics of Corn Straw and Biomass Blends…

61

After the ignition front left each thermocouple, the bed temperature decreased owing to heat loss from the gas to the bed wall and fresh layers of fuel. At about 3330 s, the ignition front reached thermocouple T6, while the bed temperature in the zone of thermocouple T10 sharply increased to 820°C. This meant that biomass fuel in this zone ignited. The reason for this is as follows. Primary air was fed from the bottom of the reactor through the grate with preheating (Tpa = 150°C). Biomass fuel in the zone of T10 dried little by little. The temperature of the fuel bed in this zone increased after the completion of drying. The volatile fraction was then released from the drying fuel. Furthermore, a channel often formed in the biomass fuel bed. Once the released volatile fraction in the zone of thermocouple T10 passed through passages with higher local void fractions or ‘short-cuts’ inside the bed and ignited in the zone of thermocouple T6, the ignition front sharply propagated downward and reached the zone of thermocouple T10. Once the fuel in this zone ignited, the temperatures at thermocouples T10, T9, T8, and T7 sharply increased (see Fig. 29a). The flame front temperature for T10 was about 360°C higher than the flame front temperatures for T9, T8 and T7. The reason for this was that volatiles at thermocouple T10 released quickly and burned intensively with sufficient oxygen flowing through the grate, which consumed most O2 in the primary air and left little for the burning of biomass blend fuel above thermocouple T10. After the ignition front reached the grate, there was another rise in the bed temperature in the accelerating combustion period. The bed temperature then decreased during the decay of the final char combustion reaction. The temperatures measured at six points inside the bed for the case of a primary air flow rate of 0.17 kg m–2 s–1 are plotted in Figure 3b. A plateau at 100°C indicates a drying process in the zone above the grate. This was due to the drying of the extensively preheated primary air for wet biomass blend fuel in this zone. There was no plateau at 300°C to indicate pyrolysis. The temperature gradient in the burning zone was steeper than that in the heating zone. After the ignition front reached the grate, there was a high-temperature zone having a temperature of 1100–1190°C (see parts a and b of Figure 29) about 30 mm above the grate in the central zone of the residual char. Generally, three main stages at bed level were responsible for the interior bed temperature history. The first stage was the sharp increase in bed temperature upon the arrival of the ignition front; the second stage was the rise in bed temperature during the accelerating combustion period; the third stage was the rise in bed temperature after the accelerating combustion period. The first stage is referred to as the steady ignition front propagation period, the second as the accelerating combustion period, and the third as the char oxidation period.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

62

Zhengqi Li

3) Gas Composition in the Bed Figure 30 shows the measured gas concentration at the probe position 400 mm above the grate for the case of a primary air flow rate of 0.17 kg m–2 s–1. The O2 concentration quickly dropped from the ambient level of 21% to about 2.6% as soon as the ignition front propagated downward and reached the gas-sampling port (y = 400 mm). There was an initial sharp increase in CO and CO2 concentrations, followed by a period of relatively steady CO and CO2 concentrations. The measured CO and CO2 concentrations were 8.65 and 14.50% respectively. The O2 concentration dropped from 2.6 to 1.34%. The higher concentration of CO was due to the fuel-rich conditions. Later at about 3990 s, the CO and CO2 concentrations decreased quickly and the O2 concentration began to increase to the ambient level during the decay of the final char combustion reaction. 500

22

NO

20

O2

400

16

CO2

350

14 300 12 250 10

NO (ppmV,dry)

Conc. (%v,dry)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

18

450

200

8

150

6

CO

4

100 50

2 0 0

1000

2000

3000

4000

0 5000

Time (s)

Figure 30. Gas compositions in the bed vs. reaction time.

Two zones of high exhaust NO concentration remained after the corn stalk and biomass blends ignited. The exhaust NO concentration peaked shortly after the biomass blends at the gas-sampling port ignited (at 3540 s). This was due to the small amount of volatile formation from biomass devolatilization and the combustion occurring under the fuel-lean condition. The NO concentration then

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

63

decreased owing to the combustion condition changing from the fuel-lean to the fuel-rich condition, and the amount of volatiles from biomass devolatilization increased. There was another peak of the NO exhaust concentration at 3930 s. The concentrations measured at the gas sampling port indicated the flue gas composition resulting from the fuel combustion between the gas sampling port and the ignition front. As the ignition front shifted downward, the amount of burning fuel increased, causing the concentration of NO to increase. After the volatiles had gone and only char burned, there was a rapid decrease in the NO concentration, after which the concentration decreased steadily to zero. 4) Effect of the primary air flow rate Figure 31 shows the fuel burning rates during the steady ignition front propagation, during accelerating combustion and during the char oxidation, the average burning rate and the bed temperature (the mean value of the peak temperature in the front measured by different thermocouples (T4–T10)). The average burning rate ν av was defined as

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

ν av = aν 1 + bν 2 + cν 3 ,

(14)

where a is the ratio of the fuel mass burned during the steady ignition front propagation, b is the ratio of the fuel mass burned during the accelerating combustion, c is the ratio of the fuel mass burned during the char oxidation, and ν 1 , ν 2 and ν 3 are the fuel burning rates during the ignition front propagation, during the accelerating combustion and during the char oxidation respectively. The experiment result indicated that the average burning rate increased with an increasing air flow rate when the primary air flow rate was less than 0.23 kg m– 2 –1 s , and then decreased. These cases were for fuel-rich conditions with a combustion stoichiometric ratio of 0.78–0.99. However, once the primary air flow rate was greater than 0.23 kg m–2 s–1, the average burning rate decreased. The experiment results show that the bed temperature increased with an increase in the primary air flow rate. Figure 32 shows the variation in the relative NO concentration with the primary air flow rate. The yield of NO was determined not only from the NO exhaust concentration but also from the average burning rate (i.e., the burning time), the amount of initial fuel-N in the fuel bed, which was proportional to the packed bulk density of the biomass blends, and the dilution effect of the primary

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

64

Zhengqi Li

air flow rate. Following Zhou et al. [34], a relative concentration ( FNO , rel ) was defined to compare the net amount of exhaust NO in different runs: 1600

0.16

Burning rate of steady ignition front propagation period Burning rate of accelerated combustion period Burning rate of char oxidation period Average burning rate Bed temperature

0.10

1200 1000

0.08

800

0.06

600

0.04

400

0.02

200

o

0.12

1400

Bed temperature ( C)

2

Burning rate (kg/(m s))

0.14

0

0.00 0.14

0.16

0.18

0.20

0.22

0.24

0.26

2

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Primary air flow rate(Kg/(m s))

Figure 31. Primary air flow rate vs. burning rate and bed temperature.

FNO,rel =

t bur m f ,ref ρ s ,ref C NO , t ref m f ρ s C NO,ref

where t is the burning time, m is the primary air flow rate,

(15)

ρ s is the bulk

density of packed biomass blends, C NO is the measured NO exhaust concentration, and the subscript ref refers to a primary air flow rate of 0.17 kg m–2 s–1 Under the operating condition of the current investigation, the initial bed porosity was artificially changed to a set value for all test cases ( ρ s ,ref = ρ s ).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics of Corn Straw and Biomass Blends…

65

2.2 2.0 1.8

FNO,rel

1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.14

0.16

0.18

0.20

0.22

2

0.24

0.26

Primary air flow rate(Kg/(m s))

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 32. Relative NO concentration vs. primary air flow rate.

It is seen that the relative NO concentration decreased with an increasing air flow rate until the primary air flow rate was 0. 23 kg m–2 s–1 and then increased. These results were for fuel-rich conditions with a combustion stoichiometric ratio of 0.78–0.99. The average burning rate increased with an increasing air flow rate when the primary air flow rate was less than 0.23 kg m–2 s–1 (see Fig. 31). The volatiles were then released more quickly and fuel combustion was under fuelrich conditions. Thus, the concentration of NO decreased with an increasing primary air flow rate. When the primary air flow rate was greater than 0.23 kg m–2 s–1, there was greater formation of NO because the fuel combustion was under relatively fuel-lean conditions. A higher average burning rate and lower NO emission can be obtained when the primary air flow rate is 0.23 kg m–2 s–1. 5) Conclusions The combustion of biomass blends occurred in three stages: steady ignition front propagation, accelerating combustion and char oxidation. The average burning rates increased rapidly with an increasing air flow rate when the primary air flow rate was less than 0.23 kg m–2 s–1, and then decreased. The mean

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

66

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

concentration of NO decreased with an increasing air flow rate when the primary air flow rate was 0.23 kg m–2 s–1, and then increased. A higher average burning rate and lower NO emission can be obtained when the primary air flow rate is 0.23 kg m–2 s–1.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Chapter 5

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

COMBUSTION CHARACTERISTICS AND NO FORMATION FOR BIOMASS BLENDS IN A 35–TON-PER-HOUR TRAVELLING GRATE UTILITY BOILER A 35-ton-per-hour (tph) coal-fired travelling grate boiler in the Lanxi coalfired power plant has been retrofitted to co-fire biomass. In this study, data were recorded for local mean concentrations of O2, CO, and NO, and gas temperatures in the region above the grate. The method for calculating the mass ratio of the biomass fly ash and boiler slag is presented. The NOx and O2 contents were measured in the flue gas and the boiler efficiency determined.

5.1. Utility Boiler A 35-tph travelling grate coal-fired boiler in the Lanxi power plant is used for electricity and heat supply. The boiler has operated since 1986. No flue gas cleaning method exists, except for using a multitubular cyclone. The main parameters of the boiler are shown in Table 10. The bulk density of biomass fuel blends (forestry waste, such as cortex and wood blocks) was 265 kg/m3, and the angle of repose was 46°. Table 9 shows the proximate and ultimate analyses of the biomass blends. The moisture content (as received basis) was 40%. The volatility of the biomass blends was quite high. The gross calorific value was low and only amounted to half the designed coal gross calorific value. Nitrogen and sulphur contents were low.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

68

Zhengqi Li

After biomass blends were burned, the fuel mass and volume both increased. Thus when the coal-fired boiler was retrofitted, the front collecting header was raised so that the height of the fuel inlet increased. Biomass cut and feed systems were added, and the hopper in front of the furnace was enlarged. The furnace and the tail-heating surfaces were unchanged. In this retrofitting, we successfully used the original travelling grate instead of an expensive and complex water-cooled vibrating grate.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 10. Main design parameters of the coal-fired 35-tph boiler and main run parameters of the biomass-fired boiler. Fuel

Coal

Biomass blends

Biomass blends

Boiler load

35 tph

28 tph

18 tph

Discharge pressure of superheated steam

3.82 MPa

3.40 MPa

3.4 MPa

Discharge temperature of superheated steam

450°C

440°C

440°C

Feed water temperature

150°C

143°C

143°C

Hot air temperature

151°C

175°C

154°C

Flue gas temperature

159°C

155°C

130°C

NOx concentration at 6% O2

/

257 mg/m3

198 mg/m3

SO2 concentration at 6% O2

/

84 mg/m3

62 mg/m3

/

53.64%

46.88%

/

44.45%

21.50%

Distance from the end of the flame to the end of the grate

0 mm

1350 mm

2000 mm

Boiler efficiency

80.68%

81.56%

86.37%

Carbon in fly ash

C cfa

Carbon in boiler slag

Cbsc

With 100% biomass blends fired, the boiler operated stably, with parameters such as steam pressure and steam temperature being within normal range. The moisture content of the biomass blends used reached 40%. With the 35-tph load, 78901 m3/h of flue gas was produced with coal-fired and 91462 m3/h with

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics and NO Formation for Biomass Blends…

69

biomass blends fired. When biomass blends were used, the amount of flue gas exceeded the normal load of the suction fan. Therefore, with the old suction fan unchanged, the normal load of the biomass blends-fired boiler decreased to 28 tph. Experiments were conducted using 28- and 18-tph loads. The parameters measured were the gas temperature and concentrations of components in the region above the grate, the gas temperature of the furnace, the flue gas temperature and component concentrations and the fly ash concentration [43].

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

5.2. Data Acquisition Techniques Parameters were measured as follows. (1) The gas temperature and gas species mean concentration in the region above the grate were measured with a nickel–chromium nickel–silicon thermocouple and a water-cooled stainless steel probe inserted through monitoring ports (at z = 515 mm, where z = 0 is the position of the grate face) in the side walls. (2) The same thermocouple was inserted through monitoring ports near the front and rear walls (z = 4795 mm) and the monitoring ports (z = 9045 mm) in side walls to measure the gas temperature distribution. (3) The flue gas temperature and component and fly ash concentrations were measured in the back-end duct. The water-cooled stainless steel probe included a water-in pipe, water-out pipe, sampling pipe, and outer pipe. The samples passed through filtrating devices into a Testo 350 M gas analyzer to be analyzed. The fly ash was obtained with a sampling system that consisted of a sampling pipe, cyclone separator, ash collector, flow meter and air pump.

5.3. Results and Discussion 5.3.1. Combustion Characteristics The combustion of biomass blends on the grate had three zones. The first zone was the heating zone. In this zone, biomass blends received radiation heat from high-temperature gas and the boiler front arch after they were carried into the furnace through the inlet. The moisture in the biomass blends then vaporized rapidly so that we could see white vapor in the test. The second zone was the fuelburning zone, where the volatile escaped rapidly and burned as the temperature of the biomass blends increased, and some of the char in the fuel burned also. The combustion of biomass blends was semi-suspension combustion with a bright

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

70

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

flame. The third zone was the burnout zone for the large-size biomass blends and residual char. According to the results of the one-dimensional bench combustion test and biomass pyrolysis tests [26, 38, 39, 42], the primary air flow rate has a great effect on combustion characteristics and NO formation. Corresponding to the above three combustion zones, the dampers of six stoker air compartments were 0% open , 50% open, 100% open, 0% open, 0% open and 0% open, from the front damper to the last damper. There was a high-temperature gas recirculation zone between the front arch and the grate. This enhanced the heating of the biomass blends from the inlet.

5.3.2. Gas Temperature and Component Concentrations in the Region above the Grate The grate was 7180 mm long and 4520 mm wide. The gas temperature distributions in the region above the grate with the 28- and 18-tph loads are shown in Fig. 33, where x is the distance from the measurement points to the fuel inlet along the grate length, y is the distance from the measurement points to the inner face of the right wall along the grate width, and z is the distance from the measurement points to the grate along the furnace height. The experiment results for the 28-tph load show that biomass blends had different combustion zones along the grate length. There was an 850–950°C gas temperature zone in the region of x = 1700–2650 mm. Because the biomass blends distributed unevenly along the grate width, and there was a channel formation in the fuel bed that caused most of the air or gas flow to pass through passages with higher local void fractions or ‘short-cuts’ inside the bed, gas temperatures distributed asymmetrically along the grate width. The gas temperature near the wall was about 50–100°C lower than that near the centre of the grate. This was due to the presence of channel between the grate and the wall causing most of the air flow. For the 28-tph load, the distance from the position of the high-temperature zone above the grate to the fuel inlet was 500 mm less than that for the 18-tph load. Figures 34-36 show the gas component distributions in the region above the grate. The test results for the 28-tph load show the O2, CO and NO contents varied greatly along the grate length. At x = 1000–1700 mm, the fuel volatile began to escape and fire, and the O2 content was higher in the gas. In the main fuel burning zone (x = 1700–2650 mm), much fuel volatile escaped and fired.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics and NO Formation for Biomass Blends…

28 ton per hour 1050 1000

Temperature (0C)

950 900 850 800 750 700 650 600 550

y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

500 1000 1500 2000 2500 3000 3500 4000 4500 5000

x (mm)

18 ton per hour

1000 950

Temperature (0C)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

1050

900 850 800 750 700 650 600 550

y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

500 1000 1500 2000 2500 3000 3500 4000 4500 5000

x (mm) Figure 33. Gas temperature distribution in the region above the grate.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

71

72

Zhengqi Li

28 ton per hour 20

O2 (%)

15

10

y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

5

0 1000

1500

2000

2500

3000

3500

4000

4500

5000

x (mm)

18 ton per hour

.

15

O2 (%)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

20

10

y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

5

0 1000 1500 2000 2500 3000 3500 4000 4500 5000

x (mm) Figure 34. O2 distribution at the grate.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics and NO Formation for Biomass Blends…

28 ton per hour 90000

y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

80000

CO (ppmv)

70000 60000 50000 40000 30000 20000 10000 0 1000

1500

2000

2500

3000

3500

4000

4500

5000

x (mm)

18 ton per hour 80000

y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

70000

CO (ppmv)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

90000

60000 50000 40000 30000 20000 10000 0

1000 1500 2000 2500 3000 3500 4000 4500 5000

x (mm) Figure 35. CO distribution at the grate.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

73

74

Zhengqi Li

28 ton per hour 400

y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

350

NO (ppmv)

300 250 200 150 100 50 0 1000

1500

2000

2500

3000

3500

4000

4500

5000

x (mm)

18 ton per hour y=0 mm y=753 mm y=1506 mm y=2260 mm y=2636 mm

350 300

NO (ppmv)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

400

250 200 150 100 50 0 1000

1500

2000

2500

3000

3500

4000

4500

5000

x (mm) Figure 36. NO distribution at the grate.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics and NO Formation for Biomass Blends…

75

The O2 content decreased quickly and became approximately zero. The contents of CO and NO increased sharply and reached 85780 ppmv and 330 ppmv. Behind the main fuel burning zone, a little fuel was left to burn out. The O2 content increased gradually, and the contents of CO and NO decreased gradually. Because the biomass blends distributed unevenly along the grate width and there was channel formation in the fuel bed, the gas contents distributed asymmetrically along the grate width. For the 18-tph load, the fuel burning zone was nearer to the inlet than the zone for the 28-tph load was, and the contents of CO and NO were lower than the contents for the 28-tph load.

5.3.3. Gas Temperature in the Furnace Figure 37 shows the gas temperature in the furnace. Figures 33 and 37 show that the gas temperature increased after the gas left the grate, and then decreased when the gas reached the furnace exit. The gas temperature in the central zone of the furnace was higher than that near the side wall. The temperature at the measurement point near the front wall was higher than that near the rear wall at z = 4795 mm. At z = 4795 and 9045 mm, the gas temperature for the 28-tph load was higher than that for the 18-tph load.

1050 1000

Temperature (0C)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

28 ton per hour 1100

950 900 850 800 750 700

near front wall (z=4795mm) near rear wall (z=4795mm) z=9045mm

650 600 550 0

500

1000

1500

2000

2500

3000

y (mm) Figure 37. Continued on next page.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

76

Zhengqi Li

18 ton per hour 1100 1050

Temperature (0C)

1000 950 900 850 800 750 700 650

near front wall (z=4795mm) near rear wall (z=4795mm) z=9045mm

600 550 0

500

1000

1500

2000

2500

3000

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

y (mm) Figure 37. Gas temperature distribution in the furnace.

5.3.4. The Mass Ratio of Fly-Ash and Boiler Slag There has been no report in the literature on a method of calculating the mass ratio of fly ash and boiler slag in a biomass-fired boiler. Let us assume the mass ratio of fly ash is α fa ; then the mass ratio of boiler slag α bs = 1 – α fa . The bar gap of the grate was very small and it was observed that few biomass blends could leak through the gap. Therefore, the heat loss resulting from leaking biomass blends can be ignored. The heat loss due to unburned carbon in fly ash and boiler slag q 4 is

q4 = q4fa + q4bs .

(16) fa

bs

The heat losses due to unburned carbon in fly ash q4 and boiler slag q4 are

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics and NO Formation for Biomass Blends…

C cfa 327 Aar q = × α fa × Qnet ,ar 100 − C cfa fa 4

,

327 Aar Cbsc , q = × α bs × Qnet ,ar 100 − Cbsc bs 4

c

77

(17)

(18)

c

where Aar , C fa , Cbs and Qnet ,ar are the contents of fuel ash (as received basis), unburned combustible in flue dust and unburned combustible in slag and the fuel gross calorific value, respectively. The value of 32700 kJ/kg is the carbon calorific value. The boiler efficiency η can be determined as η = 1 – ( q2 + q3 + q4 + q5 + q6 ),

(19)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

where q2 , q3 , q5 and q6 are the heat loss due to exhaust gas, the heat loss due to unburned gas, the heat loss due to radiation, and the heat loss due to sensible heat in slag, respectively. We can determine the four parameters using the data from the test. The fuel-consumption rate of a boiler B can be determined as

B=

D (ish.s − i f , w )

ηQnet , ar

,

(20)

where D is the boiler output and ish, s and if, w are superheated steam and feed water enthalpies. The total mass of the ash in the fuel mash can be determined as

mash = B × Aar .

(21)

According to the fly ash concentration ρ and the gas volume V measured in the flue gas, the fly ash mass rate mfa can be determined as

m fa = V × ρ .

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

(22)

78

Zhengqi Li This gives another mass ratio of fly ash

α /fa =

m fa mash

×100% .

If we determine η using fly ash

α /fa : (23)

α /fa instead of α fa , we have a new mass ratio of

α . We repeat this process until // fa

α nfa − α nfa−1 α

n fa

≤ 1% , at which point α nfa

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

is the mass ratio of fly ash. Using this method, the mass ratios of fly ash and boiler slag are 42% and 58% for the 28-tph load.

5.3.5. Boiler Efficiency and Main Run Parameters The efficiency and main run parameters of the boiler are shown in Table 10. The flue gas temperatures of the biomass-fired boiler with the 28- and 18-tph loads were 48 and 29°C lower than that of the coal-fired boiler with the 35-tph load, respectively, and the carbon contents in the biomass fly ash and boiler slag were higher. Because there was less biomass ash (see Table 9), there was little effect on the boiler efficiency. For the 28-tph load, the boiler efficiency was 81.56% and higher than that for the coal-fired boiler. For the 28-tph load, the fly ash concentration was 482 mg/m3 in the flue gas, and the contents of NOx and SO2 at 6% O2 were 257 and 84 mg/m3. 5.3.6. Fly–Ash Residue Biomass fly ash is one kind of plant ash. It contains many kinds of ash elements, such as potassium, calcium, magnesium, sulfur, iron, and silicon. The content of potassium is the largest. The measured content of potassium in the biomass fly ash was 3.603%, and it became 4.342% as potassium converted to potassium oxide. This is close to the content of potassium oxide in the plant ash, which was 5.28%. Therefore, we can use the biomass fly ash as fertilizer.

5.4. Conclusions A 35-tph coal-fired travelling grate boiler has been successfully retrofitted to co-fire biomass with the 28-tph load. The boiler efficiency was 81.56% and the

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Combustion Characteristics and NO Formation for Biomass Blends…

79

emissions of NOx, SO2 and fly ash at 6% O2 were 257, 84 and 482 mg/m3.The mass ratios of the biomass fly ash and boiler slag were 42% and 58%. For the 18tph load, the fuel burning zone was nearer to the inlet than was the case for the 28tph load, and the contents of CO and NO in the fuel burning zone above the grate were lower.

5.5 Nomenclature C cfa Carbon in fly ash

Cbsc Carbon in boiler slag x The distance from the measurement points to the fuel inlet along the grate length y The distance from the measurement points to the inner face of the right wall along the grate width z The distance from the measurement points to the grate along the furnace height α fa The supposed mass ratio of fly ash Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

α /fa The calculated mass ratio of fly ash in the first time

α //fa The calculated mass ratio of fly ash in the second time α nfa The calculated mass ratio of fly ash in the n time α nfa−1 The calculated mass ratio of fly ash in the n-1 time α bs The mass ratio of boiler slag q2 The heat loss due to exhaust gas

q3 The heat loss due to unburned gas q 4 The heat loss due to unburned carbon in fly ash and boiler slag

q4fa The heat losses due to unburned carbon in fly ash q4bs The heat losses due to unburned carbon in boiler slag q5 The heat loss due to radiation q6 The heat loss due to sensible heat in slag Aar The contents of fuel ash (as received basis)

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

80

Zhengqi Li

C cfa Unburned combustible in flue dust

Cbsc Unburned combustible in slag Qnet ,ar The fuel gross calorific value

η The boiler efficiency

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

B The fuel-consumption rate of a boiler D The boiler output ish, s Superheated steam enthalpies if, w Feed water enthalpies mash The total mass of the ash in fuel ρ The fly ash concentration V The gas volume mfa The fly ash mass rate

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Chapter 6

NUMERICAL SIMULATIONS OF BIOMASS COMBUSTION AND NO FORMATION IN A FIXED BED

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

6.1 Mathematical Model 6.1.1 Physical Model The mathematical model describes the combustion of biomass in the grate boiler. Radiation as a heat source is supplied by the free space above the grate. Air is introduced from under the grate with a uniform distribution. Biomass moving forward with the bed is heated by radiation from the flame and the furnace wall and transfers the heat to the nether fuel. On the bed, heat is mainly transferred vertically. The biomass is supplied by a feeder at invariable constant speed. The burning state of biomass on the grate relates to the moving time of the bed. At a time t, the current horizontal position of fuel is calculated using [29]

x = xo + ut ,

(24)

where x is the current horizontal position at time t (m), x0 is the initial horizontal position as x = 0 (m), u is the bed transport velocity (m/s), and t is the time (s). As shown in Fig. 38, the current horizontal position of fuel is determined by the time and bed transport velocity. Under the same initial conditions and external boundary conditions, the extent of burning at time t is equivalent to that at position x, so it is reasonable to describe the combustion of biomass on the bed approximately using a one-dimensional unsteady fixed-bed mathematical model.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

82

Zhengqi Li The main assumptions of the model are as follows. 1) The model is one-dimensional, and the air consists of oxygen and nitrogen; 2) No fragmentation or attrition of solid particles takes place; 3) The channel that may exist in the actual process is ignored, and the effects of turbulence and its impact on the gas in the bed are neglected; 4) The releasing time of the volatile and the corresponding heat variation are neglected. h o t g a s a n d v o la tile

r a d ia tio n fr o m th e fre e s p a c e fu e l s u p p lie d to th e g r a te

r a d ia n t h e a t

fla m e

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

x=ut

th e p r im a r y a ir

th e p rim a r y a ir

Figure 38. Schematic view of a moving bed transforming to a fixed bed.

6.1.2 Combustion Model The combustion of waste biomass in the bed can be separated into four consecutive or overlapping subprocesses: moisture evaporation, volatile release/char formation, combustion of the volatiles and char gasification. The subprocesses can be described by different reaction rate equations. 1) Moisture evaporation When the temperature of solid phase Ts is below 100°C, the moisture in the fuel diffuses to a gas phase with a lower vapor concentration. Equation (25) describes the drying process. When Ts < 373K,

r

H 2O

(

= k m A Cs , H 2 O − C g , H 2O

),

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

(25)

Numerical Simulations of Biomass Combustion… where

rH 2O

is the volumetric vaporization rate (kg/(m3s)),

83

C s ,H 2O

is the

C

concentrations of moisture in the solid phase state (kg/m3), g ,H 2O is the concentration of moisture in the gas phase (kg/m3), and km is the mass transfer coefficient (m/s).

km =

where

ShDH2 O

DH2O

d

,

(26)

is the molecular diffusivity (m2/s) and Sh is the Sherwood number.

Sh = 2.0 + 1.1 × Sc1H32 O Re0.6

,

(27)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

where ScH2O is the Sherwood number of steam [44]. When the solid temperature is 100°C, the process of moisture evaporation can be described by Eq. (28). The evaporation rate is determined by the heat absorption of the solid phase and the latent heat of moisture vaporization [29]. When Ts = 373 K,

rH 2O = QH 2O γ

,

(28)

where γ is the latent heat of moisture vaporization (J/kg) and QH2O is the heat absorption for the solid phase, which includes heat transfer and radiative heat transfer (J/(m3s)).

QH 2 O = A[ h(Tg − Ts ) + ε sσ (Tg4 − Ts4 )]

,

(29)

where σ is the Stefan–Boltzmann constant. 2) Volatile release Volatile produced by the pyrolysis of waste biomass is a mixture of hydrocarbons (CHx). According to the assumptions made by Hawksley and Badzioch, volatile release is described by the first-order reaction equation. The

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

84

Zhengqi Li

release rate is proportional to the surplus of volatile [16]. Thus, the volatile release rate equation is

rvol = k vol mmol ,

(30)

where kvol is the rate constant of devolatilization (1/s) and mvol is the mass of volatiles remaining in the fuel (kg/m3). 3) Combustion of the volatiles The finite rate reaction equation is used to describe the combustion of volatile and CO in the gas phase:

x 2CH x + ( + 2)O 2 → x H 2 O + 2 CO 2 . 2 The combustion reaction of long-chain and ring hydrocarbons has a finite rate [35]:

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

rCH =

dCH = −k1Tg Pg 0.3 (CH )0.5 (CO ) exp[− E1 /( RTg )] , dt

(31)

where CH is the molar concentration of hydrocarbons (mol/m3) and CO is the molar concentration of oxygen (mol/m3). 1 CO + O2 → CO 2 2 The reaction rate is calculated using [45] rCO =

dCCO = − k2Tg pg0.3CCO CH1/22O CO1/2 exp(− E2 / RTg ) , dt

(32)

where CCO and CH2O are the molar concentrations of CO and stream respectively (mol/m3).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Numerical Simulations of Biomass Combustion…

85

4) Combustion of char Due to the combustion of volatiles and downward heat transfer, the char temperature increases gradually, and then char burns at a certain temperature. A diffusion/kinetics model is used to describe the combustion of char [46]:

Θ C + O2 → (2Θ − 1)CO + (2 − Θ )CO 2 , where Θ is the stoichiometric ratio for char combustion, expressed as

Θ = 2(rc + 1) /(rc + 2) ,

(33)

and rc is the ratio of the CO/CO2 formation rates, expressed as rc =

CCO = 12 exp(−3300 / Ts ) . CCO2

(34)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Oxygen diffusion, chemical reaction dynamics and ash influence the reaction rate of char, so the total char combustion rate rchar is expressed as [47]

rchar = k mchar po2 ,

(35)

where PO is the oxygen partial pressure at the particle surface (Pa), k is the char 2 chemical reaction rate constant, and fuel (kg/m3).

mchar is the mass of char remaining in the

6.1.3 Governing Equations for the Gas and Solid Phases The mass conservation equation of the solid phase [41] is

∂ ⎡⎣ ρ s (1 − ε ) ⎤⎦ / ∂t = − rs ,

(36)

where ρs is the density of the solid phase (kg/m3), ε is the porosity of the bed, and rs is the rate of conversion from the solid to the gas phase as a source term due to moisture evaporation, pyrolysis and char combustion (kg/(m3s)).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

86

Zhengqi Li

The energy equation for the solid phase [35, 48] includes the effects of the convective heat transfer between the gas phase and solid phase, heat gain of char combustion, latent heat of moisture vaporization and coupling thermal conductivity along the bed wall: ∂[ ρs (1 − ε )hs ] ∂ ∂T − (λeff s ) = hA(Tg − Ts ) + Qs + QH2 O , ∂t ∂x ∂x

(37)

where hs is the solid phase enthalpy (J/kg), h is the gas–solid heat transfer coefficient (w/(m2K)), A is the volumetric solid surface area (m2/m3), Tg and Ts are the gas and solid phase temperatures respectively (K), QH2O and Qs are the heat loss due to moisture evaporation and the heat gain due to char combustion in which the heat produced by the heterogeneous reaction is assumed to be first deposited in the solid phase (W/m3), and λeff is the effective thermal conductivity (w/(m2K)) [48]. λeff can be expressed as

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

λeff = λeff ,0 + 0.5λg PrRe ,

(38)

where λeff,0 is the thermal conductivity for no fluid flow (w/(m2K)) [48]. λeff,0 is given by

λeff,0 = ε (λg + hrv Δl ) +

(1 − ε )Δl , 1/(λg / lv + hrs ) + ls / λs

(39)

where lv is the equivalent thickness of a representative layer of fluid that has the same thermal resistance as the fluid film does (m).

lv = 0.15912Δl ( λg / λair )

0.3716

ε 1.7304 ,

(40)

where hrv and hrs are the effective radiation heat transfer coefficient of the voids and the radiation heat transfer coefficient for radiation at the contact surface respectively (W/(m2K)), Δl is a characteristic distance between two particles (m), ls = 2d/3 is the equivalent thickness of a layer of representative solid that has the same thermal resistance as a sphere does, λs is the thermal conductivity of the pure solid, λs = 0.05, and λair is the air thermal conductivity (W/(m2K)).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Numerical Simulations of Biomass Combustion… hrv = 0.1952[1 +

hrs = 0.1952(

ε (1 − ε s ) −1 Ts 3 ] ( ) , 2ε s (1 − ε ) 100

T εs )( s )3 , 2 − ε s 100

87 (41)

(42)

Δl = 0.96795d (1 − ε ) −1 3 ,

(43)

λair = 5.66 ×10 −5 Tg + 1.1×10 −2 ,

(44)

where εs is the solid emissivity [48]. The continuous equation for the gas phase [41] is ∂( ρgε )

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

∂t

+

∂ ( ρgν g ) ∂x

= rs .

(45)

Turbulence of gas flow with low velocity in the bed is neglected in simulation. The gas phase flow model uses the Darcy formula and the hydraulic diameter of the porous medium, and then obtains the momentum equation for the gas phase with a pressure gradient in the porous medium: ∂ ( ρgν g ) ∂t

+

∂ ( ρ gν gν g ) ∂x

=−

∂Pg ∂x

+ F (ν g ) .

(46)

The function F(vg) relates to the flow state. When Re < 10,

μ κ

F (ν g ) = − ν g .

(47)

When Re ≥ 10, F (ν g ) = −

μ K

ν g − ρg Cν g ν g

,

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

(48)

Zhengqi Li

88

K=

C=

d 2ε 3 150(1 − ε ) 3 ,

(49)

1.75(1 − ε ) dε 3 ,

(50)

where vg is the velocity of vapor (m/s), Pg is the hydrostatic pressure (Pa), ρg is the vapor density (kg/m3), K and C are constants, and κ is the permeability rate. The energy equation for the gas phase is affected by the convective heat transfer between the gas phase and solid phase, vertical heat conductivity of the gas phase in the bed and heat gain due to gas combustion [35]: ∂ ( ρgε hg )

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

∂t

+

∂Tg ∂ ( ρg vg hg − λg ) = hA(Ts − Tg ) + Qg , ∂x ∂x

(51)

where hg is the gas phase enthalpy (J/kg) and Qg is the heat gain of the gas due to combustion (W/m3). The gas phase is assumed to be a mixture of volatiles (CHx), CO, inert gas (N2), vapor, CO2 and O2. There is combustion and convection diffusion of components. The conservation equation for the chemical reaction expressed by yi is [35] ∂ ( ρ g ε yi ) ∂t

+

∂yi ⎞ ∂ ⎛ ⎜ ρ g vg yi − Γ i ⎟ = ri , ∂x ⎝ ∂x ⎠

(52)

where yi is the mass fraction of the species, i represents an individual species (i = vol (CHx), CO, N2, H2O, CO2 and O2), and Γi is the molecular diffusion coefficient. The source term ri represents the rate of mass production or subtraction of individual species (kg/(m3s)) and is determined from the combustion rate for gas components and other main reaction rates for the bed.

6.1.4 Model of NO Formation and Subtraction 1) Mechanism of NO formation On the basis of the NOx formation mechanism, there are three types of NOx: thermal NO, prompt NO and fuel NO. Fuel NO is formed by the oxidation of fuel-

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Numerical Simulations of Biomass Combustion…

89

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

N. Little thermal NO or prompt NO formed during the combustion of biomass with a low flame temperature. NO is mainly transformed by fuel-N. The simulations for NO formation by biomass only consider the formation of fuel NO. The formation mechanism of fuel NO is very complex. The formation and reduction of fuel NO relate with not only the characteristics and structure of the fuel and the distribution and composition of fuel-N in the volatile and char after the thermal decomposition, but also combustion conditions such as the temperature and concentrations of oxygen and other components. Figure 39 presents the proposed pathways of fuel-N conversion. Most fuel-N is released as NH3 and HCN together with a small amount of NO during the release of volatiles. The liberated NH3 and HCN together with char-N are then oxidized to NO in the presence of oxygen. The formed NO may reduce to HCN and N2 in the presence of H2, hydrocarbons (CH4 and C2H6), char, NH3 and HCN [34].

Figure 39. Pathways of fuel-N conversion.

2) Transformation of fuel-N The proportions of volatile-N and char-N transformed by fuel-N depend on the fuel species, pyrolysis temperature, residence time and heating rate. When the amount of volatile, pyrolysis temperature and heating rate increase, the amount of volatile-N increases while the amount of char-N decreases. Generally speaking, the mass fraction of char-N is about 25% in the temperature range of 900 to 1200 K [34]. Compared with coal, biomass fuels are characterized by higher O/N ratios, which may lead to the release of NO with volatiles. Experiments by Lin and Dam-

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

90

Zhengqi Li

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Johansen [18] showed that up to 1–4% of fuel-N is directly converted to NO during the release of volatiles. According to the above analysis and the mass fractions of tar and light gas during the pyrolysis of biomass [50], the ratio of volatile-N, char-N and NO transformed by fuel-N assumed in the model is Ngas/Nchar/NNO = 71/25/4. The mass fractions of HCN, NH3 and N2 in the volatile depend not only on the biomass species but also on the pyrolysis temperature and heating rate. Leppalahti, Koljonen [47] and Weissinger [51] reported that NH3 is the main nitrogen compound in the volatile during biomass pyrolysis. According to experiment data for wood pyrolysis operated using an entrained flow reactor [52], the mass ratio of HCN, NH3 and N2 assumed in the model is 10/90/0. 3) Chemical reactions for the transformation of fuel-N Table 11 presents 11 global homogeneous and heterogeneous chemical reactions for the transformation of fuel-N. R1–R5 are conversion reactions of NH3 to NO and N2. Basically, the reaction constants except that for R4 were obtained directly from the literature. Considering the alkali metal in the biomass as a catalyst in NH3 conversion to NO and N2, the reaction constant for R4 is six times that reported by Jensen et al. [53]. Biomass pyrolysis mostly yields CO, CO2, tar, small quantities of light hydrocarbons, and H2. The hydrocarbon and H2 mixture formed by biomass pyrolysis may be a highly efficient NO reduction agent [54–56]. R6 and R7 are included in the reaction network to investigate the contribution of the NO reaction with gaseous hydrocarbons and hydrogen. The global reaction rates are obtained from Chen et al. [55] and Johnsson [57]. Atakan and Hartlieb [58] found that the most NO consumed in R6 is transformed to N2 and HCN, and a minor fraction is transformed to NH3. It is assumed that 70% of NO is directly reduced to N2 and 30% to HCN in R6. The reaction rate of R7 is six times that reported by Johnsson [57] since the mineral content and particularly high alkali metal content in the biomass may have a great catalytic effect on the NO reduction rate [59]. The global reaction scheme for HCN, R8, was obtained from Desroches-Ducarne et al. [60].

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated, 2009. ProQuest

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Table 11. Reactions of fuel-N conversion included in the models. Chemical reaction

Reaction rate

Kinetic constant

Remark

Reference

R1

NH3 + NO + 0.25O2 → N 2 + 1.5H 2

0.5 0.5 0.5 r1 = kCNH CNO CO2 3

k = 1.1× 1012 exp(−27680 / Tg )

Homogeneous

[53]

R2

NH3 + 1.25O2 → NO + 1.5H 2O

r2 = kCNH3 CO2

k = 5.07 ×1014 exp(−38160 / Tg )

Homogeneous

[53]

Char-catalyzed

[53]

Char-catalyzed

[53，59]

k = 6.75 ×10 exp(−10000 / Tg ) 6

R3

NH3 + 1.25O2 → NO + 1.5H 2O

r3 = kCNH3 CO2 /(CO2 + k′)

R4

NH3 + 0.75O2 → 0.5N 2 + 1.5H 2O

r4 = kCNH3 CO2 /(CO2 + k′)

R5

2 / 3NH3 + NO → 5 / 6N 2 + H 2O

0.64 0.29 r5 = kCNO CNH3

k = 8.2 ×104 exp(−12000 / Tg )

Char-catalyzed

[54]

r6 = kCNOCCi H j

k = 2.7 ×106 exp(−9466 / Tg )

HCN:N2 = 30:70

[55，58]

r7 = kCNOCH2

k = 1.2 ×107 exp(−8623 / Tg )

Char-catalyzed

[57，59]

Homogeneous

[60]

Heterogeneous

[64] [61，65]

R6 R7

∑ ∑

i, j

Ci H j + NO → HCN + H 2O + N 2

i, jCi H j + NO → CO + H 2 O + N 2

H 2 + NO → 0.5N 2 + H 2 O HCN + 0.5O2 → CNO;

R8

r8 = kC O2 C HCN r11 = k ′C O2 C HCN

CNO + 0.5O2 → NO + CO;

(1 /(1 + k 2 / k1C NO ))

CNO + NO → N2 + 0.5O2 + CO

r21 = kC O2 C HCN ( k 2 / k1 /(1 + k 2 / k1C NO ))

R9 R10 R11

1 NO + C → N2 + CO 2

[C − N ] +

ψ +1 2

O 2 → β NO + [ C − O ]

Fuel-N release during devolatilization and char combustion

0.7 r9 = kCNO

k ′ = 0.054 k = 1.7 ×107 exp(−10000 / Tg )

k ′ = 0.054

k = 2.14 ×105 exp(−10000 / Tg ) k ′ = 2.14 × 105 exp(−10000 / Tg ) k2 / k1 = 1.02 × 109 exp(−25460 / Tg )

k = f (t ) × g (T ) × 1.24 ×106 exp( −16180 / Tg )

r10 = −α rchar

α = N /C

Heterogeneous

r11 = −γ rvol(CH )

γ = ( N / vol )initial

Homogeneous

Zhengqi Li

92

Several mechanisms have been proposed for the reduction of NO on the surface of char [61–63]. These mechanisms result in the release of N2 and chemisorbed oxygen as CO. NO reduction depends not only on the temperature but also on the physical structure and chemical composition. The strong NO–char reaction is mainly attributed to the catalytic effect of the alkali metal in the biomass. It was found that the high concentration of CO has a positive effect on the NO–char reaction [64]. The rate of the NO–char reaction as a function of time and temperature may be expressed as R9. The coefficient f(t) is based on the experimental data of Garijo et al. [66] and is expressed as

f (t ) = 0.23 + 0.77 exp(−t / 200) , while g(t) is an empirical coefficient and is −4 given as g (t ) = 8exp((0.7 − 7.4 × 10 T ) / 0.07) . To account for the contribution of char–N to NO during char combustion, the De Soete model [61] was adopted in the reaction simulation. The model proposes that the char-bound nitrogen is oxidized to NO via R10. The rate of this reaction is assumed to be proportional to the rate of char combustion with the constant of

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

proportionality being defined as ψ . [C–N] and [C–O] represent the char-bound nitrogen and oxygen respectively. Similar to the case for NO formation during char combustion, the rate of fuel–N release during volatile release and combustion may be expressed as R11, where g is the ratio of N and volatiles and is assumed to be

γ = ( N / vol )initial .

6.1.5 Variations in the Bed Height and Fuel Density During combustion, the volume of fuel particles varies gradually with the volatile release and char combustion and the height and volume of the bed decrease. Hence, the volume contraction of the bed due to mass loss should be considered in model selection. In the moisture evaporation stage, the fuel density decreases and it is appropriate to adopt the variable density model. During volatile release and char combustion, the diameters of fuel particles decrease with the precipitation of reaction products, and thus a shrinking core model is chosen. The variations in particle geometry, density and governing micro-unit volume are calculated using

ρ s = ρ s (1 − U b ) , ys,wat > 0

(53)

0

d p = d p0 ((1 − U b ) (1 − y s ,wat ))

1/ 3

, ys,wat ≤ 0

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

(54)

Numerical Simulations of Biomass Combustion…

Vm ⎛ d p =⎜ V0 ⎜⎝ d p 0

⎞ ⎟⎟ ⎠

3

⎛ ε p0 ⎜⎜ ⎝ εp

⎞ ⎟⎟ , ys,wat ≤ ⎠

0

93

(55)

where ρs and ρs0 are the current density and initial density of fuel particles (kg/m3), dp and dp0 are the current diameter and initial diameter of fuel particles (m) respectively, ys,wat is the current water mass fraction and ys,wat0 is the initial value in the waste fuel, εp is the current volume fraction of solid phase and εp0 is the initial value in fuel, and Ub is the ratio of mass loss/initial mass in the governing micro-unit.

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

6.2 Computation Method and Grid having a Varying Dimension The differential equations were discretized by the finite-volume method. Governing equations were solved by the SIMPLEST algorithm. The grid system in the computational domain was reestablished in every time step to obtain the next convergence solution. Figure 40 is the flowchart of the calculation. The boundary of the computational domain varies during the combustion, so a grid having a varying dimension was used in the model. The grid system was determined from the size of the particle after the previous time step [67]. The computational domain was divided into 70 grids. During the simulation, the width and length of the grid were invariant, with only the height Hgrid varying.

(

H grid = H 0 × ( d ' d 0 ) × (1 − ε ' ) + H 0 × (1 − ε ' ) − (1 − ε )

),

(56)

where H0 is the initial height of the grid (m), d0 is the initial diameter of particles and d is the particle diameter at time t (m), and ε is the porosity at time t. The width and length of the computational domain were invariant, but the height of the domain varied. The coordinate of each grid Xc was '

'

(

)

3 X C = X 0 + H 0 × ( d ' d ) × (1 − ε ) (1 − ε ' ) ,

(57)

where X0 is the initial coordinate of the lower vertex of the grid pattern. Thus, the height variation of the bed can be calculated to describe the dynamic combustion.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Zhengqi Li

94

Fuel temperature>100

No

Yes

Moisture evaporation model

Mass transfer model caused by the concentration difference

Fuel temperature increases to 300

Volatile release model

Fuel temperature increases to 300

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Combustion model of volatile Fuel temperature keeps increasing, the bed height decreases

Fuel temperature keeps increasing, the bed height decreases

Combustion model of char

Burn out

Figure 40. Flowchart of the calculation.

6.3 Computational Boundary and Initial Conditions The object to be simulated is a fixed-bed test facility. Boundary conditions include the primary air being introduced from under the grate with a uniform distribution, initial temperatures of the fuel bed and primary air both being 20°C, the environment pressure being atmospheric pressure, the temperature above the bed being 900°C and the combustion lasting 2500–3500 s. Initial conditions are as porosity of the bed of 0.58, biomass diameter of 17 mm, bed height of 540 mm and fuel weight of 1.1 kg. The proximate and ultimate analysis results for the fuel are listed in Table 6.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Numerical Simulations of Biomass Combustion…

95

The simulations were carried out with a time step of 1.0 s and a total computational time of 4800 s. Relative residuals of the governing equations in each time step are all less than 0.1% [68].

6.4 Numerical Simulations of Combustion and NO Formation for Different Primary Air Flow Rates

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 41 shows that the profiles of bed temperature against the primary air flow rate obtained from simulation and experiment are in good agreement. Both profiles indicate that the bed temperature first increases and then decreases with an increase in the primary air flow rate. The trend was discussed in section 4.3.

Figure 41. Bed temperatures from experiment and simulation.

Figure 42 compares the profiles of O2 concentration obtained from simulation and experiment. For a primary air flow rate of 0.029 kg/m2·s, the simulation results are in good agreement with the experimental data, but for a primary air flow rate of 0.077 kg/m2·s, there are obvious differences. Because of the greater influence of air flow turbulence on the combustion process due to the increased primary air flow rate, fuel particles are easily broken up and so the actual process of combustion is different from that simulated.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Zhengqi Li

96

O2 concentration (%)

25

experiment data simulation results 2 0.029 kg/m s

20

15

10

5

0 0

500

1000

1500

2000

time (s)

2500

3000

3500

20

O2 concentration (%)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

25

experiment data simulation results 2 0.077 kg/m s

15

10

5

0 0

500

1000

1500

time (s)

2000

2500

Figure 42. O2 concentration profiles from experiment and simulation.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

3000

Numerical Simulations of Biomass Combustion…

97

CO concentration (%)

10 10

experiment data simulation results 2 0.029 kg/m s

88

66

44

22

00 0

500

1000

1500

2000

time (s)

2500

3000

3500

experiment data simulation results 2 0.077 kg/m s

8

CO concentration (%)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

10

6

4

2

0 0

500

1000

1500

time (s)

2000

2500

Figure 43. CO concentration profiles from experiment and simulation.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

3000

Zhengqi Li

98 14

experiment data simulation results 2 0.029 kg/m s

CO2 concentration (%)

12

10

8

6

4

2

0 0

500

1000

1500

2000

time (s)

2500

3000

3500

14

CO2 concentration (%)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

12

experiment data simulation results 2 0.077 kg/m s

10

8

6

4

2

0 0

500

1000

1500

2000

2500

3000

time (s)

Figure 44. CO2 concentration profiles from experiment and simulation.

Figure 43 compares the profiles of the CO concentration obtained from simulation and experiment. For a primary air flow rate of 0.029 kg/m2·s, the

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Numerical Simulations of Biomass Combustion…

99

simulation results agree well with experiment results. When the primary air flow rate increased to 0.077 kg/m2·s, the simulation results are uniform with the experimental data. Figure 44 compares the profiles of the CO2 concentration obtained from simulation and experiment. The simulation results agree well with experiment data for the different primary air flow rates. There is little difference between simulation results and experiment data at the beginning of CO2 release. Figure 45 shows that the profiles of the NO concentration obtained from simulation and experiment agree well. Both profiles have two peaks during the combustion. The NO concentration in the experiment decreases slowly in the later stage, while the simulated concentration decreases quickly. In addition, the NO releases later in the simulation than it does in the experiment. Figure 46 compares the profiles of the mean combustion velocity obtained from simulation and experiment. The simulation results and experiment data are in agreement for the primary air flow rates of 0.029 and 0.044 kg/m2·s, but the simulation results are lower than experiment data when the rate increases to 0.077 and 0.092 kg/m2·s. The main reason for this is that the simulation results for the mean combustion velocity are obtained by calculating the moisture evaporation rate and combustion velocities of the volatile and char whereas experiment data are obtained by measuring the mass loss rate of fuel in the bed during combustion. Parts of ash and char in the bed could be blown away at the high primary air rate, but these factors are not taken into considered in the simulation. As shown in the figure, the mean combustion velocity first increases and then decreases with an increase in the primary air flow rate. Figure 47 shows profiles of the mean NO concentrations for different primary air flow rates. The NO concentration is a minimum when the primary air flow rate is 0.044 kg/m2·s. For a lower primary air flow rate, simulation results are lower than experiment data, but for a primary air flow rate above 0.044 kg/m2·s, the simulation results are higher. This is possibly related to the channeling phenomenon at a high primary air flow rate.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Zhengqi Li

100 1000

experiment data simulation results 2 0.029 kg/m s

NO concentration (ppm)

800

600

400

200

0 0

500

1000

1500

2000

time (s)

2500

3000

3500

experiment data simulation results 2 0.077 kg/m s

400

NO concentration (ppm)

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

500

300

200

100

0 0

500

1000

1500

2000

2500

time (s) Figure 45. NO concentration profiles from experiment and simulation.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

3000

Numerical Simulations of Biomass Combustion…

0.044

2

combustion velocity ( kg/m .s)

0.040 0.036 0.032 experiment data simulation results

0.028 o

0.024 0.020 0.016 0.012 0.008 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 2

primary air flow rate ( kg/m .s)

1000 3 NO concentration ( mg/Nm ( %6O2) )

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

Figure 46. Combustion velocity vs. primary air flow rate.

900 800

o

experiment data simulation results

700 600 500 400 300 200 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 2 primary air flow rate ( kg/m .s)

Figure 47. Mean NO emission concentration vs. primary air flow rate.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

101

102

Zhengqi Li

6.5 Conclusions

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

A numerical model of biomass combustion and NO formation in a fixed bed were presented. The SIMPLEST transient algorithm and a grid with a varying dimension were used. Simulation results of the bed temperature, gas species concentration, combustion velocity and NO emission concentration coincided with experiment data for different primary air rates. The numerical simulation can be used to study the combustion of biomass and the associated NO formation.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

ACKNOWLEDGMENTS

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

This work was sponsored by the Heilongjiang Provincial Natural Science Foundation of China (Contract No.: E200623).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved. Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

REFERENCES [1] [2] [3] [4]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]

Peter, MK. Bioresource Technology, 200283, 37–46. Matti, P. Biomass and Bioenergy, 200427, 613–620. Yao, BY, Changkook, R, Adela, K. Fuel, 200584, 2116-2130. Liao, CP, Yang, YJ, Wu, CZ, Huang, HT. Biomass and Bioenergy, 200427, 111-117. Sun, ZJ. Transactions of the CSAE, 200420 (5), 1-5. Fisher, T, Hajaligol, M, Waymack, B, Kellogg, D. Journal of Analytical and Applied Pyrolysis, 200262, 331-349. Demirbas, MF, Bala, M. Energy Conversion and Management, 200647, 2371-2381. Atul, S, Rajeswara, RT. Bioresource Technology, 199967, 53-59. Blasi, CD, Branca, C, DErrico, G. Thermochimica Acta, 2000364, 133-142. Cao, Q, Xie, KC, Bao, WR, Shen, SG. Bioresource Technology, 200494, 83-89. Cai, JM, Liu, RH. Bioresource Technology, 200899, 2795–2799. Zabaniotou, AA, Kantarelis, EK. Bioresource Technology, 200899, 3174– 3181. Orfao, JJM, Antunes, FJA, Figueiredo, JL. Fuel, 199978, 349-358. Meszaros, E, Varhegyi, G, Jakab, E. Energy & Fuels, 200418, 497-507. Li ZQ, Zhao W, Meng BH, Liu CL, Zhu QY, Zhao GB. Bioresource Technology, 2008 99(16), 7616 – 7622. Varhegyi, G, Antal, MJ, Jakab, E, Szabao, P. Journal of Analytical and Applied Pyrolysis, 199742, 73-87. Hu, S, Jess, A, Xu, MH. Fuel, 200786, 2778-2788. Haykiri-Acma, H, Yaman, S, Kucukbayrak, S. Renewable Energy, 200631, 803–810.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

106 [19] [20] [21] [22] [23] [24]

[25] [26] [27] [28] [29]

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41]

[42] [43]

References Teng, H, Wei, YC. Ind. Eng. Chem. Res., 199837, 3806-3811. Miura, K, Maki, T. Energy & Fuels, 199812, 864-869. Cai, J, Liu, R. Bioresource Technology, 200899(8), 2795-2799. Sonobe, T, Worasuwannarak, N. Fuel, 200887, 414-421. Miura, K. Energy & Fuels, 19959, 302-327. Criado, JM, Pérez-Maqueda, LA, Sánchez-Jiménez, PE. J. Therm. Anal. Calorim., 200582, 671-675. Hashimoto, K, Miura, K, Watanabe, T. AIChE J., 198228,: 737-746. Li ZQ, Liu CL, Chen, ZC, Qian, J, Zhao W, Zhu, QY. Bioresource Technology, 2009100, 948-952 Maki, T. Energy & Fuels, 199711, 972–977. Saastanainen, JJ, Taipale, R. Combustion and Flame, 2000123, 214-226. Van der Lans, RP, Pedersen, LT, Jensen, A, Glarborg, P, Dam-Johansen, K. Biomass and Bioenergy, 200019, 199-208. Friberg, R, Blasiak, W. Biomass and Bioenergy, 200223, 189-208. Jan, F, Ulrik, H. Energy & Fuels, 200317,: 1251-1258. Yang, YB, Sharifi, VN, Swithenbank, J. Fuel, 200483, 1553-1562. Thunman, H, Leckner, B. Proceedings of the Combustion Institute, 200530, 2939-2946. Zhou, H, Jensen, AD, Glarborg, P, Kavaliauskas, A. Fuel, 200685, 705-716. Zhou, H, Jensen, AD, Glarborg, P, Jensen, PA. Kavaliauskas, A. Fuel, 200584, 389-403. Shin, D, Choi, S. Combustion and Flame, 2000121, 167-180. Changkook, R, Yao, BY, Adela, K, Nicola, EY, Vida, NS, Jim, S. Fuel, 200685, 1039-1046. Zhao, W, Li, ZQ, Wang, DW, Zhu, QY, Sun, R, Meng BH and Zhao GB. Bioresource Technology, 2008 99(8), 2956-2963. Li, ZQ, Zhao, W, Zhao GB, Zhang, FS, Zhu, QY. Energy & Fuels, 200822, 2009-2014. Yang, YB; Changkook, R; Adela, K; Sharifi, VN. Fuel, 200584, 2026-2038. Gort, R. On the propagation of reaction front in a packed bed: thermal conversion of municipal waste and biomass. Academic Dissertation, University of Twente, Erschede, The Netherlands, 1995. Zhao, W, Li, ZQ, Zhao GB, Zhang, FS, Zhu, QY. Energy Conversion and Management, 2008 49, 3560-3565. Li, ZQ, Zhao, W, Li, RY, Wang, ZW, Li, Y, Zhao GB. Bioresource Technology, 2009100(7), 2278-2283.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

References

107

[44] Yang YB, Yamauchi H, Nasserzadeh V, Swithenbank J. Fuel, 200382(18), 2205-2221. [45] Goh YR, Lim CN, Chan KH, et al. The Second International Symposium on Incineration and Flue Gas Treatment Technology, 1999, 4-6. [46] Hu S, Jess A, Xu MH. Fuel, 200786, 2778-2788. [47] Leppälahti J, Koljonen T. Fuel Processing Technology, 199543, 1-45. [48] Fjellerup J, Henriksen U, Jensen AD, Jensen PA, Glarborg P. Energy & Fuels, 200317, 1251-1258. [49] Lin W, Dam-Johansen K. CHEC report: NO emissions in combustion of annual biomass. Dept of Chem Eng, Technical University of Denmark; 2005. [50] Bassilakis R, Carangelo RM, Wójtowicz MA. Fuel, 200180, 1765-1786. [51] Weissinger A, Fleckl T, Obernberger I. Combustion and Flame, 2004137, 403-417. [52] Aho MJ, Hämäläinena JP, Tummavuori JL. Combustion and Flame, 199395, 22-30. [53] Jensen A, Johnsson JE, Andries J, Laughlin K, Read G, Mayer M. Fuel, 199574, 1555-1569. [54] Johnsson JE, Jensen A, Nielsen JS. Kinetics of heterogeneous NO and N2O reduction at FBC conditions. In: R.B. Reuther, Editor, Proceedings of the 15th International Conference on Fluidized Bed Combustion, Savannah, GAPaper No. FBC 99-0099, ASME (1999). [55] Chen W, Smoot LD, Hill SC, Fletcher TH. Energy & Fuels, 199610, 10461052. [56] Rudiger H, Greul U, Spliethoff H, Hein KRG.. Combustion Science and Technology, 1996121, 299-315. [57] Johnson JE. A new NOx module for the IEA-model. 21st IEA–AFBC Meeting, Beograd, 1990. [58] Atakan B, Hartlieb AT. Applied Physics B: Lasers and Optics, 200071, 697702. [59] Bueno-López A, Garcia-garcia A. Carbon, 200442, 1565-1574. [60] Desroches-Ducarne EJ, Dolignier C, Marty E, Martin G., Delfosse L. Fuel, 199877, 1399-1410. [61] Soete GGD, Croiset E, Richard JR. Combustion and Flame, 1999117,: 140154. [62] Chan LK, Sarofim AF, Beer JM. Combustion and Flame, 198352, 37-45. [63] Johnsson JE. Fuel, 199473, 1398-1415. [64] Sørensen CO, Johnsson JE, Jensen A. Energy & Fuels, 200115, 1359-1368.

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

108

References

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

[65] Zhou H, Flamant G, Gauthier D. Chemical Engineering Science, 200459, 4205-4215. [66] Garijo EG, Jensen AD, Glarborg P. Energy & Fuels, 200317, 1429-1436. [67] Zhou LX, Li Y. Powder Technology, 2000113, 70-79. [68] Zhao, W. Experimental study and numerical simulation on combustion characteristics of biomass waste in a fixed bed. Ph. Dissertation, Harbin Institute of Technology, Harbin, P.R. China, 2009 (in Chinese).

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

INDEX

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

A absorption, 46, 83 activated carbon, 3 activation energy, ix, 5, 11, 18, 19, 20, 22, 25, 28, 41, 44 active centers, 11 agent, 3, 90 air, ix, 19, 31, 33, 34, 38, 39, 40, 41, 42, 44, 45, 46, 51, 52, 53, 54, 57, 58, 59, 61, 62, 63, 64, 65, 68, 69, 70, 82, 86, 94, 95, 98, 99, 101, 102 algorithm, 93, 102 alkali, 90, 92 ash, ix, 4, 33, 34, 46, 58, 67, 68, 69, 76, 77, 78, 79, 80, 85, 99 assumptions, 28, 82, 83 atmospheric pressure, 94 availability, 54

B behavior, 18 biomass, ix, 1, 3, 4, 5, 6, 8, 18, 19, 21, 25, 31, 33, 38, 58, 61, 62, 63, 64, 65, 67, 68, 69, 70, 75, 76, 78, 81, 82, 83, 89, 90, 92, 94, 102, 106, 107, 108 blends, 58, 62, 63, 64, 65, 67, 68, 69, 70, 75, 76 blocks, 67

Boltzmann constant, 83 boundary conditions, 81 burning, ix, 31, 32, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 49, 51, 52, 53, 54, 55, 57, 58, 59, 61, 63, 64, 65, 69, 70, 75, 79, 81 burnout, 36, 70

C calcium, 78 carbon, 3, 4, 19, 33, 34, 46, 58, 76, 77, 78, 79 catalyst, 90 catalytic effect, 19, 90, 92 cellulose, ix, 5, 6, 8, 11 CH4, 89 char combustion, 61, 62, 85, 86, 91, 92 chemical reactions, 90 China, ix, 1, 103, 108 chromium, 69 cleaning, 67 CO2, ix, 32, 34, 36, 45, 51, 57, 62, 85, 88, 90, 98, 99 coal, 1, 19, 20, 21, 22, 25, 27, 28, 67, 68, 78, 89 combustion, ix, 1, 31, 32, 33, 35, 36, 37, 38, 39, 41, 42, 45, 46, 49, 51, 52, 53, 54, 55, 57, 58, 59, 61, 62, 63, 65, 69, 70, 81, 82, 84, 85, 86, 88, 89, 91, 92, 93, 94, 95, 99, 102, 107, 108 combustion chamber, 32, 33

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Index

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

110

combustion characteristics, ix, 31, 70, 108 complexity, 8 components, ix, 3, 5, 11, 69, 88, 89 composition, 3, 31, 37, 63, 89, 92 concentration, x, 31, 34, 36, 37, 38, 40, 41, 42, 44, 45, 46, 49, 50, 51, 55, 57, 62, 63, 64, 65, 66, 68, 69, 77, 78, 80, 82, 83, 84, 92, 95, 96, 97, 98, 99, 100, 101, 102 conductivity, 86, 88 conservation, 85, 88 consumption, 1, 54, 77, 80 convection, 31, 39, 88 convective, 86, 88 convergence, 93 conversion, 1, 3, 8, 20, 21, 22, 25, 28, 85, 89, 90, 91, 106 conversion rate, 20, 21, 22, 25 corn, ix, 1, 3, 4, 6, 7, 8, 10, 11, 12, 17, 18, 20, 21, 22, 25, 28, 33, 34, 37, 41, 42, 44, 45, 46, 51, 57, 62 correlation, 25 correlation coefficient, 25 cortex, 67 coupling, 86 cracking, 3 crops, ix, 1 cyclone, 67, 69

D Darcy, 87 decay, 53, 61, 62 decomposition, ix, 3, 4, 5, 6, 8, 11, 18, 89 definition, 5 degradation, 42 degradation rate, 42 Denmark, 107 density, 33, 34, 44, 46, 58, 63, 64, 67, 85, 88, 92, 93 derivative thermogravimetry, 3 developing countries, 1 deviation, 18 devolatilization, 37, 49, 54, 62, 84, 91 differential equations, 93 diffusion, 9, 85, 88

diffusivity, 83 distribution, ix, 19, 20, 25, 28, 42, 45, 60, 69, 71, 72, 73, 74, 76, 81, 89, 94 distribution function, 20 drying, 37, 42, 45, 61, 82 dust, 77, 80

E electricity, 67 emission, x, 41, 45, 57, 65, 66, 101, 102 energy, 1, 11, 22, 42, 86, 88 energy consumption, 1 environment, 94 equilibrium, 49, 52 evaporation, 53, 54, 82, 83, 85, 86, 92, 99 extinction, 31

F fertilizer, 78 film, 86 fire, 67, 68, 70, 76, 78 flame, 49, 51, 52, 53, 54, 57, 61, 68, 70, 81, 89 flow rate, 4, 31, 33, 34, 39, 40, 41, 42, 44, 45, 46, 51, 54, 57, 58, 59, 61, 62, 63, 64, 65, 70, 95, 98, 99, 101 flue gas, 32, 37, 49, 51, 57, 63, 67, 68, 69, 77, 78 fluid, 86 forestry, 67 fragmentation, 82 fuel, ix, 3, 31, 32, 34, 36, 37, 38, 39, 41, 42, 45, 46, 47, 49, 51, 52, 53, 54, 55, 57, 58, 59, 61, 62, 63, 65, 67, 68, 69, 70, 75, 77, 79, 80, 81, 82, 84, 85, 88, 89, 90, 91, 92, 93, 94, 95, 99 furnaces, 1

G gas, ix, 1, 3, 4, 5, 31, 32, 34, 36, 37, 39, 45, 46, 51, 57, 58, 59, 61, 62, 67, 68, 69, 70,

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Index 75, 77, 78, 79, 80, 82, 83, 84, 85, 86, 87, 88, 90, 102 gas phase, 46, 82, 83, 84, 85, 86, 87, 88 gases, 45, 54 gasification, 82 Gaussian, 19, 20, 25, 28 generation, 31, 49, 52 grids, 93

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

H H2, 89, 90 heat, 31, 34, 37, 39, 42, 49, 51, 52, 53, 54, 55, 57, 61, 67, 69, 76, 77, 79, 81, 82, 83, 85, 86, 88 heat capacity, 42 heat conductivity, 88 heat loss, 34, 49, 52, 61, 76, 77, 79, 86 heat transfer, 31, 39, 49, 52, 83, 85, 86, 88 heating rate, 4, 6, 7, 8, 11, 17, 18, 20, 21, 41, 44, 89, 90 height, 33, 38, 42, 44, 46, 51, 58, 68, 70, 79, 92, 93, 94 hemicellulose, ix, 5, 8, 11, 18 heterogeneous, 8, 11, 18, 37, 86, 90, 107 high temperature, 6 hydro, 83, 84, 89, 90 hydrocarbons, 83, 84, 89, 90 hydrogen, 4, 33, 90 hydrostatic pressure, 88

I IEA, 107 inert, 4, 88 interaction, 3 iron, 78 iteration, 11 kinetic equations, 5 kinetics, ix, 3, 11, 18, 85

L lignin, ix, 5, 8, 11

111

linear, 6, 21, 22, 25, 28, 59 losses, 7, 76, 79

M magnesium, 78 mass loss, ix, 5, 7, 18, 31, 34, 36, 37, 38, 41, 54, 55, 57, 58, 59, 92, 93, 99 mass transfer, 83 measurement, 6, 32, 33, 36, 59, 60, 70, 75, 79 metal content, 90 metallurgy, 3 mixing, 46 models, ix, 4, 8, 10, 11, 17, 18, 91 moisture, ix, 31, 37, 51, 53, 54, 55, 57, 67, 68, 69, 82, 83, 85, 86, 92, 99 moisture content, 37, 54, 67, 68 molecular weight, 3 momentum, 87

N natural, 1 natural gas, 1 Netherlands, 106 network, 90 nickel, 69 Nielsen, 107 nitrogen, 4, 33, 82, 90, 92 NO, i, v, vii, ix, 31, 32, 33, 37, 40, 41, 42, 44, 45, 49, 50, 51, 55, 57, 58, 62, 63, 64, 65, 66, 67, 70, 74, 75, 79, 81, 88, 89, 90, 92, 95, 99, 100, 101, 102, 107 normal, 68

O oil, 1 oxidation, 31, 37, 38, 39, 40, 41, 42, 45, 46, 47, 49, 51, 52, 54, 57, 61, 63, 65, 88 oxidation rate, 49, 54 oxide, 78 oxygen, 4, 31, 33, 37, 38, 39, 54, 61, 82, 84, 85, 89, 92

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Index

112 Oxygen, 85

S P

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

parameter, 8, 18, 31 particles, 4, 37, 46, 49, 51, 55, 57, 82, 86, 92, 93, 95 pathways, 89 permeability, 88 porosity, 42, 46, 64, 85, 93, 94 porous, 87 ports, 69 potassium, 78 power plant, 67 precipitation, 92 pressure, 68, 85, 87, 88, 94 probe, 32, 62, 69 production, 3, 88 propagation, 31, 37, 38, 40, 41, 42, 44, 45, 46, 47, 48, 49, 51, 52, 53, 54, 55, 57, 61, 63, 65, 106 proportionality, 92 pyrolysis, ix, 3, 4, 5, 6, 7, 8, 10, 11, 12, 18, 19, 21, 37, 61, 70, 83, 85, 89, 90

R radiation, 46, 69, 77, 79, 81, 86 random, 33 range, 4, 11, 25, 51, 68, 89 raw materials, 41 reaction order, 5, 11 reaction rate, 8, 31, 53, 82, 84, 85, 88, 90 reaction time, 37, 59, 62 reactivity, 5 refractory, 32 relationships, 22, 23, 24, 26, 27, 28, 29, 30 residuals, 95 resistance, 86 resources, ix, 1 room temperature, 51

sample, 4, 21, 33, 44, 45 sampling, 32, 37, 62, 69 shape, 18 Sherwood number, 83 shoulders, 5 silicon, 69, 78 simulation, 8, 10, 11, 18, 87, 92, 93, 95, 96, 97, 98, 99, 100, 102, 108 simulations, 89, 95 slag, ix, 67, 68, 76, 77, 78, 79, 80 SO2, ix, 68, 78, 79 solid phase, 82, 83, 85, 86, 88, 93 spatial, 48, 49 species, ix, 19, 32, 39, 69, 88, 89, 90, 102 speed, 81 stages, 31, 45, 61, 65 stainless steel, 32, 69 standard deviation, 20 statistics, 1 steady state, 58 steel, 32, 69 subtraction, 88 sulphur, 4, 33, 67 Sun, 105, 106 superheated steam, 68, 77 supply, 54, 67 surface area, 86 surplus, 84 switching, 58 synthesis, 3

T tar, 39, 90 temperature, ix, 6, 7, 20, 21, 22, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 45, 46, 47, 48, 49, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 63, 64, 68, 69, 70, 71, 75, 76, 82, 83, 85, 89, 90, 92, 94, 95, 102 temperature gradient, 37, 48, 49, 55, 56, 61 TGA, 3, 4, 41, 44 thermal decomposition, ix, 89

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,

Index thermal resistance, 86 thermogravimetric, ix, 3, 4, 21 transfer, 31, 39, 49, 52, 83, 85, 86, 88 transformation, 5, 90 transport, 81 travel, 38 turbulence, 82, 95

113

vapor, 69, 82, 88 variation, 9, 19, 49, 51, 54, 57, 63, 82, 93 velocity, x, 38, 42, 47, 48, 49, 51, 52, 53, 54, 55, 57, 81, 87, 88, 99, 101, 102 voids, 86 volatility, 67

W U Ub, 93 ultimate analysis, 4, 20, 33, 34, 46, 51, 58, 94 uniform, 37, 45, 46, 81, 94, 99

water, 32, 51, 68, 69, 77, 80, 93 weight loss, 21 wheat, 31, 36, 37, 39 wood, 31, 51, 58, 67, 90

V yield, 63

Copyright © 2009. Nova Science Publishers, Incorporated. All rights reserved.

values, 4, 11, 18, 20, 21, 28, 33, 37, 38, 42

Y

Corn Straw and Biomass Blends : Combustion Characteristics and NO Formation, Nova Science Publishers, Incorporated,