Modeling and Optimization of Product Profiles in Biomass Pyrolysis

Udaya Bhaskar Reddy Ragula, Sriram Devanathan and Sindhu Subramanian

## Abstract

Biomass feed comes in many varieties, but have common chief constituents of hemicellulose, cellulose, and lignin. As the relative proportions of these constituents may vary, customization of the pyrolysis process conditions is required to produce a desired product profile. By recognizing the sources of variation, the reactor settings may be intelligently controlled, to achieve optimal operation. These considerations include biomass classification, feed rate, moisture content, particle size, and interparticle thermal gradients (which arise during pyrolysis based on heating rate and temperature distribution). This chapter addresses the optimization of product profiles during biomass pyrolysis from a modeling perspective. Fundamental models for packed bed and fluidized bed pyrolyzers are developed, using kinetics from existing literature. The proposed optimization approach (inclusive of the kinetic and process models) can guide practical achievement of desired product profiles of the biomass pyrolysis process.

Keywords: biomass pyrolysis, product profiles, kinetics, modeling and optimization

## 1. Introduction

The chief constituents of any biomass are hemicellulose, cellulose, lignin and inerts [1]. The amounts of the chief constituents of various biomasses along with the elemental compositions are presented in Table 1. Generally, pyrolysis is a process where biomass or other carbonaceous material is heated in the absence of air supply or in the presence of inert gas supply such as nitrogen. During the biomass pyrolysis, the bonds between high molecular compounds are broken and low molecular weight compounds are formed. The range of pyrolysis products are gases, liquids, char, and ash. The ash is the inert material present in biomass. The fractions of these products depend on composition of biomass and process conditions such as heating rate, biomass feed rate, particle size, moisture content, and the rate of heat transfer. The rate of heat transfer depends greatly on mixing conditions in the pyrolyzers.

Table 1 provides the chief constituents of biomass in wt% on dry and ash free basis along with the elemental composition for various biomasses. These chief constituents will vary depending on the type, location, and age of the plant. The error in values reported in Table 1 is 2–5%. The wt% of hemicellulose varies from


kinetics are provided in Section 3. These kinetic parameters are obtained from a combination of prior published work and the authors' experimental data, based on thermogravimetric analysis (TGA) and lumped parameter models. Section 4 deals

accounting for the variation in product profiles due to variation in feed and operating parameters. Section 5 introduces the optimization of product profiles for these

The primary products of the pyrolysis are gas, oil, char and water. The ratios of these products depend on parameters such as particle size, heating rate, degradation temperature, and feed rate of the material [14, 17–22]. The different pyrolysis processes are flash/fast pyrolysis, intermediate pyrolysis, and slow pyrolysis [1, 17, 20, 23–25]. The pyrolysis process is classified based on the heating rate as well as the degradation temperature of biomass. The operating parameters of different pyroly-

Flash/fast pyrolysis is performed with very high heating rates (1000 K/s) for less than a second. Since, the entire biomass particles are required to be heated for a very short time, only particles of very small size (<1 mm) can be pyrolyzed using this method. Because of small particle size and very short residence times, the thermal gradients within the particles are small, and hence, there is low product variation. This method is preferable if the number of components in the product stream is required to be low. With most of the biomasses during fast pyrolysis, the

Intermediate pyrolysis is performed at medium heating rates. This kind of pyrolysis is preferred for biomass in the particle size range of 1–5 mm. During the intermediate pyrolysis, there are huge thermal gradients within the particle itself. Hence, intermediate pyrolysis is preferred when different products are required from the same biomass. Further, this type of reactor does not require a separate unit

with the modeling of continuous packed bed and fluidized bed pyrolyzers,

two types of pyrolyzers under continuous operation.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

2. Different pyrolysis processes

DOI: http://dx.doi.org/10.5772/intechopen.85581

sis processes are given in Figure 1.

products are usually non-condensable gases.

for product separation, especially, oil and water.

Different types of pyrolysis and their operating conditions.

2.2 Medium/intermediate pyrolysis

2.1 Flash/fast pyrolysis

Figure 1.

45

#### Table 1.

Different biomasses, their constituents and elemental analysis.

17.4 to 52.9, the wt% of cellulose varies from 12 to 54.5% and the wt% of lignin varies from 16.1 to 60.2.

As shown in Table 1, there is significant variation across different biomasses. Thus, products resulting from biomass pyrolysis will also vary widely, and hence, it will be difficult to control the product profiles if enough attention is not paid to these chief constituents along with the biomass processing conditions.

This book chapter considers packed bed and fluidized bed pyrolyzers, which vary significantly in their heat transfer capabilities because of the mixing conditions. Section 2 of the chapter describes different pyrolysis processes and their associated product profiles. In view of their importance in pyrolysis, the reaction Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

kinetics are provided in Section 3. These kinetic parameters are obtained from a combination of prior published work and the authors' experimental data, based on thermogravimetric analysis (TGA) and lumped parameter models. Section 4 deals with the modeling of continuous packed bed and fluidized bed pyrolyzers, accounting for the variation in product profiles due to variation in feed and operating parameters. Section 5 introduces the optimization of product profiles for these two types of pyrolyzers under continuous operation.

## 2. Different pyrolysis processes

The primary products of the pyrolysis are gas, oil, char and water. The ratios of these products depend on parameters such as particle size, heating rate, degradation temperature, and feed rate of the material [14, 17–22]. The different pyrolysis processes are flash/fast pyrolysis, intermediate pyrolysis, and slow pyrolysis [1, 17, 20, 23–25]. The pyrolysis process is classified based on the heating rate as well as the degradation temperature of biomass. The operating parameters of different pyrolysis processes are given in Figure 1.

#### 2.1 Flash/fast pyrolysis

Flash/fast pyrolysis is performed with very high heating rates (1000 K/s) for less than a second. Since, the entire biomass particles are required to be heated for a very short time, only particles of very small size (<1 mm) can be pyrolyzed using this method. Because of small particle size and very short residence times, the thermal gradients within the particles are small, and hence, there is low product variation. This method is preferable if the number of components in the product stream is required to be low. With most of the biomasses during fast pyrolysis, the products are usually non-condensable gases.

#### 2.2 Medium/intermediate pyrolysis

Intermediate pyrolysis is performed at medium heating rates. This kind of pyrolysis is preferred for biomass in the particle size range of 1–5 mm. During the intermediate pyrolysis, there are huge thermal gradients within the particle itself. Hence, intermediate pyrolysis is preferred when different products are required from the same biomass. Further, this type of reactor does not require a separate unit for product separation, especially, oil and water.

Figure 1. Different types of pyrolysis and their operating conditions.

17.4 to 52.9, the wt% of cellulose varies from 12 to 54.5% and the wt% of lignin

these chief constituents along with the biomass processing conditions.

19.1 44.6 36.3 40–

Nerium oleander 23.5 48.2 28.4 44 6.7 0.5–

As shown in Table 1, there is significant variation across different biomasses. Thus, products resulting from biomass pyrolysis will also vary widely, and hence, it will be difficult to control the product profiles if enough attention is not paid to

43

4.7– 6.0

0.8– 4.9

1.2

0.04– 0.7

0.04– 0.2

34– 38

30– 35

[14, 15]

[14, 16]

Type of biomass Biomass constituents Elemental composition Ref.#

Rice husk 19.1 44.6 36.3 39.8 5.7 0.5 0.2 39.8 [2] Cotton stalk 23.5 48.2 28.4 46.8 6.4 0.3 0.2 46.8 [2] Rice straw 28.5 54.5 17.0 38.8 6.7 0.2 0.2 38.8 [2] Wheat straw 25.5 49.3 25.2 41.7 5 0.4 0.3 41.7 [2] Corn stalk 52.9 28.0 19.1 43.8 5.7 0.9 0.1 48.9 [3] Corn cob 35.4 47.0 17.7 43.6 5.8 0.7 1.3 48.6 [3] Elephant grass 34.3 31.5 34.2 44.5 5.4 1.4 — 31.8 [3] Hazelnut shell 17.4 25.4 57.2 52.3 6.5 5.2 9.2 26.8 [3]

Switch grass 51.6 32.3 16.1 44.7 5.7 0.3 — 49.1 [4] Hazelnut husk 22.8 38.2 38.9 42.6 5.5 1.1 0.1 50.6 [5] Walnut shell 25.4 23.5 51.1 47.5 6.3 0.4 — 45.6 [6] Pinewood waste 28.4 43.0 28.5 49.3 6 0.04 — 44.5 [7] Apple pomace 27.8 47.5 24.7 47.9 6.6 0.7 — 37.4 [8] Chestnut shells 23.3 32.6 44.0 48.1 5.4 0.6 — 45.7 [9] Cherry stones 28.0 28.1 43.9 51.1 7.2 3 — 38.6 [9] Grape seeds 22.9 16.9 60.2 51.5 6.3 1.7 — 40.3 [9] P. juliflora 18.8 51.6 29.6 43.3 6.32 1.3 0.07 48.9 [10] Cashew nut shells 18.6 41.3 40.1 58.3 7 0.7 0.06 32 [3, 11] Coconut shell 27.9 40.3 31.9 53.9 5.7 0.1 0.02 39.4 [3, 11] Cagon grass 28.8 50.9 20.4 44.3 5.6 0.8 0.09 49.0 [12] Karanja fruit Hull 48.4 12.0 39.6 45.1 6.1 — 0.36 48.4 [13] Cotton stalk 23.5 48.2 28.4 46.8 6.4 0.3 0.2 46.8 [2]

C (%)

33.1 42.7 24.2 45.1 6.05 0.3 — 42.7 [3]

H (%) N (%) S (%) O

(%)

LIG\* (%)

HEMI\* (%)

Recent Advances in Pyrolysis

Sugarcane bagasse

CEL\* (%)

This book chapter considers packed bed and fluidized bed pyrolyzers, which vary significantly in their heat transfer capabilities because of the mixing conditions. Section 2 of the chapter describes different pyrolysis processes and their associated product profiles. In view of their importance in pyrolysis, the reaction

varies from 16.1 to 60.2.

HEMI, hemicellulose; CEL, cellulose; LIG, lignin.

Different biomasses, their constituents and elemental analysis.

Hibiscus rosa sinensis

\*

44

Table 1.

#### 2.3 Slow pyrolysis

Slow pyrolysis is performed at very low heating rates 0.1 K/s. The particle size during slow pyrolysis is between 2 and 10 mm. Slow pyrolysis is preferred if the charring characteristic of biomass is very high [4]. This is also the preferred type of pyrolysis when the residence time required during the pyrolysis is high.

Type of biomass Model Activation energy–Ea

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

Integral method

OFW model KAS model

reaction model

mechanism

OFW model CR model

KAS model DE Algorithm CR model

CR model

OFW model KAS model CR model

KAS model OFW model Starink model DAEM

equation

Logistic DAEM

OFW model Friedman model

Integral method

Cotton stalk Direct method

DOI: http://dx.doi.org/10.5772/intechopen.85581

Rice straw Kissinger model

Wheat straw nth-order

Elephant grass Three component

Hazelnut husk KAS model

Hazelnut shell Friedman model

Walnut shell Arrhenius model

Apple pomace Friedman model

Chestnut shells Friedman model

Cherry stones Arrhenius

Grape seeds Gaussian DAEM

P. juliflora KAS model

Sugarcane bagasse Direct method

47

Rice husk Gaussian DAEM 85–110 105

Corn stalk KAS model 62.7 107

Switch grass KAS model 77.4 108

(kJ/mol)

Hemicellulose: 102.0 Cellulose: 98.5 Hemicellulose: 127.8 Cellulose: 72.5

> 172.6 192.7 193.6

Cellulose and hemicellulose: 78 (n-0.65) Lignin: 80 (n-2.7)

Hemicellulose: 46.5– 65.5 Cellulose: 108–127.2 Lignin: 45.6–53.5

> 127.8 131.1 —

222.3 216.3 First zone: 35–153 Second zone: 20–135 35.9

> 69.3 101.6

> 197.7 213.0 201.7 —

127.2–194.8 152.7–196.7 154.1–196.1 153.2–196.9 175.2

Hemicellulose: 197.7 Cellulose: 213.0 Lignin: 201.7

> 188 3 190 2

> > 204.0 203.2 219.3

Hemicellulose: 53.5, cellulose: 43 Hemicellulose: 87.7, cellulose: 77

Pine wood Arrhenius model 150 10<sup>11</sup> [39]

Corn cob CR model 64–80 10<sup>3</sup> [35]

Pre-exponential factor-A (min<sup>1</sup>

> 1.22 0.45 2.6 0.35

10<sup>11</sup> 10<sup>22</sup> 10<sup>15</sup>

107 10<sup>6</sup>

10<sup>4</sup> –10<sup>6</sup> 10<sup>8</sup> –10<sup>11</sup> 10<sup>2</sup> –10<sup>3</sup>

— — 10<sup>5</sup> –10<sup>6</sup>

— — 10<sup>3</sup> –10<sup>8</sup> 10<sup>4</sup> –10<sup>6</sup> 102

> 105 105

— — — 10<sup>1</sup> –10<sup>3</sup>

— — — — 1011–1015

> 10<sup>10</sup> 107 102

10<sup>2</sup> 10<sup>2</sup>

10<sup>11</sup> 105 10<sup>21</sup>

)

–107 [31]

–10<sup>8</sup> [34]

–109 [34]

Reference

[32]

[32]

[33]

[36]

[37]

[38]

[6]

[8]

[40]

[41]

[42]

[10]

[32]

Because of variations in the feed and the pyrolysis process, the thermal gradients are very high. Due to these thermal gradients within the particle and in the reactor, the reactions that the particles are undergoing are different, resulting in different product profiles. To design pyrolysis reactors with controlled product selectivity, it is important to understand which reactions are taking place in the reactor/biomass particle. This means that it is required to understand the kinetic parameters such as reaction order, activation energy, and pre-exponential factor for the reaction [14, 17, 23, 26–29]. Since, biomass particles are undergoing complicated reaction networks, the estimation of kinetic parameters were limited to the overall process in the early stages. Later, the kinetic parameters were estimated using the staged decomposition of biomass due to its chief constituents of biomass, namely, hemicellulose, cellulose, and lignin, and their relative proportions in a biomass [11, 14, 20, 30]. In the next section of this chapter, the kinetics of different biomasses are presented along with their corresponding parameters.

## 3. Pyrolysis kinetics and product profiles

Kinetics play an important role in the design of pyrolysis reactors. These kinetic parameters depend on the type of the biomass feed, mixing conditions in the reactor, and inter-particle & intra-particle thermal and mass transfer limitations. As mentioned earlier, the biomass pyrolysis consists of complicated reaction networks, making it difficult to analyze the pyrolysis kinetic parameters for individual reactions.

#### 3.1 Pyrolysis kinetics

Different kinetic models were proposed by several authors working on the kinetics of the pyrolysis of biomass. Currently, there is no single model for the evaluation of kinetic parameters primarily based on the thermogravimetric (TGA) studies. The different types of kinetic models tested for the biomass pyrolysis are: (1) isoconversion (ISO) models such as Kissinger-Akahira-Sunose (KAS) model and Ozawa-Flynn-Wall (OFW) model, (2) generalized nth order reaction model, (3) Coats-Redfern (CR) model, (4) three component mechanism with CR model, (5) Gaussian Distributed Activation Energy model (Gaussian DAEM), (6) Friedmann model, and (7) Starink model. The differences lie in their approach to relate them to rate of biomass degradation with kinetic parameters-mainly the activation energy and the pre-exponential factor.

The different models to analyze the pyrolysis thermal degradation data, were classified mainly into three categories namely, accelerating, decelerating, and sigmoidal model based on the shape of rate of degradation vs. time [17, 29]. The accelerating models are the ones whose rate of decomposition increases with the increase in time. For those biomasses which fall under this category a simple power law model with respect to rate of decomposition is more suitable [29]. The decelerating models are used whose rate of reaction decreases with increase in time, such as CR model and integral models (such as KAS and OFW models). This means, there are mass transfer limitations (diffusion limited) during the reaction due to the formation of products or inert layers that slows down the reaction. The sigmoidal models


#### Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

2.3 Slow pyrolysis

Recent Advances in Pyrolysis

Slow pyrolysis is performed at very low heating rates 0.1 K/s. The particle size during slow pyrolysis is between 2 and 10 mm. Slow pyrolysis is preferred if the charring characteristic of biomass is very high [4]. This is also the preferred type of

Because of variations in the feed and the pyrolysis process, the thermal gradients are very high. Due to these thermal gradients within the particle and in the reactor, the reactions that the particles are undergoing are different, resulting in different product profiles. To design pyrolysis reactors with controlled product selectivity, it is important to understand which reactions are taking place in the reactor/biomass particle. This means that it is required to understand the kinetic parameters such as reaction order, activation energy, and pre-exponential factor for the reaction [14, 17, 23, 26–29]. Since, biomass particles are undergoing complicated reaction networks, the estimation of kinetic parameters were limited to the overall process in the early stages. Later, the kinetic parameters were estimated using the staged decomposition of biomass due to its chief constituents of biomass, namely, hemicellulose, cellulose, and lignin, and their relative proportions in a biomass [11, 14, 20, 30]. In the next section of this chapter, the kinetics of different biomasses are

Kinetics play an important role in the design of pyrolysis reactors. These kinetic parameters depend on the type of the biomass feed, mixing conditions in the reactor, and inter-particle & intra-particle thermal and mass transfer limitations. As mentioned earlier, the biomass pyrolysis consists of complicated reaction networks, making it difficult to analyze the pyrolysis kinetic parameters for individual reactions.

Different kinetic models were proposed by several authors working on the kinetics of the pyrolysis of biomass. Currently, there is no single model for the evaluation of kinetic parameters primarily based on the thermogravimetric (TGA) studies. The different types of kinetic models tested for the biomass pyrolysis are: (1) isoconversion (ISO) models such as Kissinger-Akahira-Sunose (KAS) model and Ozawa-Flynn-Wall (OFW) model, (2) generalized nth order reaction model, (3) Coats-Redfern (CR) model, (4) three component mechanism with CR model, (5) Gaussian Distributed Activation Energy model (Gaussian DAEM), (6) Friedmann model, and (7) Starink model. The differences lie in their approach to relate them to rate of biomass degradation with kinetic parameters-mainly the activation energy

The different models to analyze the pyrolysis thermal degradation data, were classified mainly into three categories namely, accelerating, decelerating, and sigmoidal model based on the shape of rate of degradation vs. time [17, 29]. The accelerating models are the ones whose rate of decomposition increases with the increase in time. For those biomasses which fall under this category a simple power law model with respect to rate of decomposition is more suitable [29]. The decelerating models are used whose rate of reaction decreases with increase in time, such as CR model and integral models (such as KAS and OFW models). This means, there are mass transfer limitations (diffusion limited) during the reaction due to the formation of products or inert layers that slows down the reaction. The sigmoidal models

pyrolysis when the residence time required during the pyrolysis is high.

presented along with their corresponding parameters.

3. Pyrolysis kinetics and product profiles

3.1 Pyrolysis kinetics

and the pre-exponential factor.

46


Biomass type Pyrolysis type Pyrolysis conditions Reactor type Products (wt%) Ref #

Fluidized bed Gas: 15%

Fixed bed Gas: 32.7%

Fluidized bed Gas: 23%

Fixed bed Gas: 44.8%

Fixed bed Gas: 54%

Fixed bed Gas: 30%

Fluidized bed Gas: 26.9%

Fixed bed Gas: 24%

Fluidized bed Gas: 35%

Fixed bed Gas: 12%

Fluidized bed Gas: 46%

Fixed bed Gas: 49%

Fluidized bed Gas: 36.3%

Fixed bed Gas: 26%

Fixed bed Gas: 21.5

Fixed bed Gas: 27

Fixed bed Gas: 55%

Fixed bed Gas: 24.4%

Liquids: 37–40% Char: 47%

Liquids.: 30.1% Char: 31.8%

Liquids: 36% Char: 16%

Liquids: 17.1% Char: 38%

Liquids: 6% Char: 40%

Liquids: 45% Char: 28%

Liquids: 34.9% Char: 28.0%

Liquids: 34% Char: 43%

Liquids: 38% Char: 28%

Liquids: 50% Char: 38%

Liquids: 36% Char: 18%

Liquids: 28% Char: 23

Liquids: 37.6% Char: 27.8%

Liquids: 44.7% Char: 29.3%

Liquids: 22.5% Water: 25% Char: 31%

Liquids: 34% Water: 11% Char: 28%

Liquids: 10% Char: 35%

Liquids: 43% Char: 32.6%

[45]

[46]

[47]

[24, 48]

[20]

[49, 50]

[28]

[51]

[52]

[53]

[54]

[55]

[56, 57]

[58]

[59]

[59]

[20]

[60]

HR: >3.33 K/s Size: <0.5 mm

HR: >0.33 K/s Size: <0.21 mm

> HR: 9 K/s Size: 1 mm

HR: 0.3 K/s Size: 3 mm

HR: 3.3 K/s Size: <0.5 mm

HR: 0.16 K/s Size: 0.3 mm

HR: 250–1000 K/s Size: 0.25–1 mm

> HR: 0.03 K/s Size: 1.33 mm

HR: 8.3 K/s Size: <0.5 mm

HR: 0.83 K/s Size: <1 mm

HR: 1000 K/s Size: 1–2 mm

HR: 10 K/s Size: 0.5–2 mm

HR: 4000 K/s Size: 0.8–1.2 mm

HR: 0.16 K/s Size: 0.21 mm

HR: 0.1 K/s Size: 0.85–1.8 mm

HR: 5 K/s Size: 0.85–1.8 mm

HR: 3.3 K/s Size: <0.5 mm

HR: 0.35 K/s Size: <0.5 mm

Slow Temp.: >500°C

Slow Temp.: 600°C

Slow Temp.:600°C

Slow Temp.: 450°C

Slow Temp.: 400°C

Intermediate Temp.: 577°C

Slow Temp.: 500°C

Intermediate Temp.: 550°C

Slow Temp.: 420°C

Rice husk Intermediate Temp.: >500°C

DOI: http://dx.doi.org/10.5772/intechopen.85581

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

Cotton stalk Intermediate Temp.: 490°C

Rice straw Intermediate Temp.:500°C

Wheat straw Fast/flash Temp.: 525°C

Corn stalks Intermediate Temp.: 500°C

Corn cob Fast/flash Temp.: 550°C

Elephant grass Fast/flash Temp.: 480–520°C

Hazelnut shell Slow Temp.:550°C

Sugarcane bagasse Intermediate Temp.: 500°C

49

#### Table 2.

Kinetic parameters of pyrolysis of different biomasses.

represent the autocatalytic type reactions during biomass pyrolysis, such as OFW and KAS models. The choice of the kinetic model purely depends on the rate of decomposition vs. time curve.

The isoconversion model and Gaussian DAEM will give only the activation of energy for the overall thermal degradation process, which is further used in Arrhenius type model for finding the pre-exponential factor.

Table 2 provides the kinetic parameters reported by researchers for different biomasses that are widely used in pyrolysis.

Comparing Tables 1 and 2, for the biomasses with high hemicellulose content, the activation energy is between 50 and 65 kJ/kmol, and the pre-exponential factor is between 10<sup>3</sup> and 108 min�<sup>1</sup> . The activation energy for high cellulose content is between 170 and 220 kJ/kmol and the pre-exponential factor is between 10<sup>7</sup> and 10<sup>21</sup> min�<sup>1</sup> . For the biomasses with high lignin content, the activation energy ranges from 65 to 140 kJ/kmol and the pre-exponential factor ranges from 10�<sup>1</sup> to 10<sup>7</sup> min�<sup>1</sup> . These wide variations in the kinetic parameters indicate that different reactions are occurring within the biomass particle.

#### 3.2 Pyrolysis product profiles: lumped product distribution

It has been mentioned earlier that the products of biomass pyrolysis vary widely. There are not many studies aimed at intrinsic kinetic parameter estimation for each individual products of pyrolysis. The pyrolysis products are lumped based on their standard phase. The lumped reaction model for the biomass pyrolysis is given in Eqs. (1) and (2) [43, 44].

$$\begin{aligned} \rightarrow \text{Gas} \\ \text{Reaction 1}: \text{ Biomass} \rightarrow \text{Tar/Liquids} \\ \rightarrow \text{Char} \end{aligned} \tag{1}$$

The tar formed during the primary biomass decomposition shall further decompose into gas and char.

$$\begin{aligned} \text{Reaction 2}: \operatorname{Tor} &\xrightarrow[\text{Char}]{\longrightarrow \text{Gas}} \end{aligned} \tag{2}$$


#### Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

represent the autocatalytic type reactions during biomass pyrolysis, such as OFW and KAS models. The choice of the kinetic model purely depends on the rate of decom-

(kJ/mol)

Cellulose: 174.3

Cellulose: 216.0

213.9 60–64

> 61.0 68.5

Cellulose: 9–62 Lignin: 65–142

Cellulose: 11–43 Lignin: 71–109

Pre-exponential factor-A (min�<sup>1</sup>

> 10<sup>13</sup> 101 –102

> > 10<sup>6</sup> 10<sup>6</sup>

10<sup>6</sup> –10<sup>8</sup> 10�<sup>1</sup> –10<sup>3</sup> 104 –107

102 –10<sup>4</sup> 10�<sup>1</sup> –10<sup>3</sup> 102 –105

10�<sup>1</sup> –10�<sup>4</sup> 10�<sup>2</sup> –10�<sup>3</sup>

10�<sup>1</sup> –10�<sup>5</sup> 10�<sup>1</sup> –10�<sup>3</sup> )

Reference

[11]

[11]

[12]

[13]

[14]

[14]

The isoconversion model and Gaussian DAEM will give only the activation of energy for the overall thermal degradation process, which is further used in Arrhe-

Table 2 provides the kinetic parameters reported by researchers for different

Comparing Tables 1 and 2, for the biomasses with high hemicellulose content, the activation energy is between 50 and 65 kJ/kmol, and the pre-exponential factor

. For the biomasses with high lignin content, the activation energy ranges

. These wide variations in the kinetic parameters indicate that different

between 170 and 220 kJ/kmol and the pre-exponential factor is between 10<sup>7</sup> and

It has been mentioned earlier that the products of biomass pyrolysis vary widely. There are not many studies aimed at intrinsic kinetic parameter estimation for each individual products of pyrolysis. The pyrolysis products are lumped based on their standard phase. The lumped reaction model for the biomass pyrolysis is

The tar formed during the primary biomass decomposition shall further decom-

Reaction 2 : Tar ! Gas

from 65 to 140 kJ/kmol and the pre-exponential factor ranges from 10�<sup>1</sup> to

. The activation energy for high cellulose content is

! Gas ! Tar=Liquids ! Char

! Char (2)

(1)

nius type model for finding the pre-exponential factor.

Type of biomass Model Activation energy–Ea

Cashew nut shells CR model Hemicellulose: 130.2

Coconut shell CR model Hemicellulose: 179.6

DAEM Global kinetic model

OFW model

Hibiscus rosa sinensis CR model Hemicellulose: 55–91

Nerium oleander CR model Hemicellulose: 29–51

reactions are occurring within the biomass particle.

3.2 Pyrolysis product profiles: lumped product distribution

Reaction 1 : Biomass

biomasses that are widely used in pyrolysis.

Kinetic parameters of pyrolysis of different biomasses.

position vs. time curve.

Imperata cylindrical (Cagon grass)

Recent Advances in Pyrolysis

Karanja fruit Hull KAS model

is between 10<sup>3</sup> and 108 min�<sup>1</sup>

given in Eqs. (1) and (2) [43, 44].

pose into gas and char.

48

10<sup>21</sup> min�<sup>1</sup>

Table 2.

10<sup>7</sup> min�<sup>1</sup>


From the data presented in Table 3, it can be generalized that the liquid products will be more in fluidized bed reactors, whereas more char is seen in fixed bed reactors. Further, slow pyrolysis results in more gas and char when compared to fast pyrolysis. This might be due to the liquid products formed in reaction 1, shown in Eq. (1), being further consumed by reaction 2, shown in Eq. (2) resulting in more gas and char. The relative proportions of these products depend on various param-

The biomass pyrolysis has been mainly carried out in two types of reactors: fixed

bed (or packed bed) reactors and fluidized bed reactors, by many authors, as presented in Table 3. The process models for these two reactors are developed from fundamental laws of conservation of mass and energy along with empirical relations for properties such as specific heat, density, diffusivity, mass and heat transfer coefficients, etc. The packed bed pyrolyzers are further classified as down-draft and updraft pyrolysis reactors. The detailed comparison of the general packed bed and

The biomass usually contains moisture. The moisture needs to be removed before the pyrolysis stage. If this is carried to the pyrolysis stage, the gasification reactions such as steam reformation may kick off resulting in undesirable products.

any) fed from bottom.

leaves from the top of the pyrolyzer.

above the bottom of the pyrolyzer.

Fluidized bed reactor • A bed of fine inert solid material is present at the bottom of the

sticking to the wall.

Comparison of fixed bed and fluidized bed pyrolyzers.

enough (1–3 m/s) to agitate the material.

products of the pyrolysis leave from the top.

• Operating temperature ranges from 300 to 750°C.

pyrolysis.

the gasifier.

• The biomass is fed from the top of the pyrolyzer, and the inert gas (if

• The pyrolysis gases and along with the liquid tar (in the form of gas)

• The ash (inert component of the biomass) is collected at the bottom of

• The feed is broken down, falling down the gasifier under gravity. A bed of hot char through which the gases are allowed to pass through (a secondary reaction zone) ensures the pyrolysis products travelling from top are further broken down. This increases the residence time through the pyrolysis stage. An exit for the pyrolysis products is provided just

• The biomass is fed from top of the pyrolyzer along with inert gas allowing the feed and gases move in the same direction.

• The ash collected under the grate at the bottom the pyrolyzer.

pyrolyzer. The inert gas is fed from the bottom of the pyrolyzer fast

• The biomass feed is fed in from the side, mixes with the inert gas and the

• The operating temperature is below 900°C to avoid ash melting and

• Operating temperature ranges from 300 to 750°C.

• Char resulting from pyrolysis falls down and may accelerate the

eters mentioned earlier.

Updraft fixed pyrolyzer

Downdraft fixed bed

pyrolyzer

Table 4.

51

4. Process modeling of pyrolysis reactors

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

DOI: http://dx.doi.org/10.5772/intechopen.85581

fluidized bed reactors is given in Table 4.

Gasifier type Specifications/conditions

#### Table 3.

Product profiles from pyrolysis of different biomass in fixed and fluidized bed pyrolyzers.

For practical purposes, the products are classified into gases, liquids, and char. The lumped product distribution from pyrolysis of different biomasses, different pyrolysis methods, pyrolysis conditions and two different pyrolyzers (fixed bed and fluidized) are provided (Table 3).

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

From the data presented in Table 3, it can be generalized that the liquid products will be more in fluidized bed reactors, whereas more char is seen in fixed bed reactors. Further, slow pyrolysis results in more gas and char when compared to fast pyrolysis. This might be due to the liquid products formed in reaction 1, shown in Eq. (1), being further consumed by reaction 2, shown in Eq. (2) resulting in more gas and char. The relative proportions of these products depend on various parameters mentioned earlier.

## 4. Process modeling of pyrolysis reactors

The biomass pyrolysis has been mainly carried out in two types of reactors: fixed bed (or packed bed) reactors and fluidized bed reactors, by many authors, as presented in Table 3. The process models for these two reactors are developed from fundamental laws of conservation of mass and energy along with empirical relations for properties such as specific heat, density, diffusivity, mass and heat transfer coefficients, etc. The packed bed pyrolyzers are further classified as down-draft and updraft pyrolysis reactors. The detailed comparison of the general packed bed and fluidized bed reactors is given in Table 4.

The biomass usually contains moisture. The moisture needs to be removed before the pyrolysis stage. If this is carried to the pyrolysis stage, the gasification reactions such as steam reformation may kick off resulting in undesirable products.


Table 4. Comparison of fixed bed and fluidized bed pyrolyzers.

For practical purposes, the products are classified into gases, liquids, and char. The lumped product distribution from pyrolysis of different biomasses, different pyrolysis methods, pyrolysis conditions and two different pyrolyzers (fixed bed and

Biomass type Pyrolysis type Pyrolysis conditions Reactor type Products (wt%) Ref #

Fluidized bed Gas: 16.5%

Fixed bed Gas: 43.2%

Fixed bed Gas: 15.6%

TGA Gas: 10%

Fluidized bed Gas: 22%

Fixed bed Gas: 26%

Fixed bed Gas: 25%

Fixed bed Gas: 71.5%

Fixed bed Gas: 51%

Fixed bed Gas: 30–33%

Fixed bed Gas: 57.4%

Fluidized bed Gas: 12.5%

Fixed bed Gas: 44

Fixed bed Gas: 49.1–74.1%

TGA Gas: 8.8–47.6%

Liquids: 57.9% Char: 20.0%

Liquids: 27.5% Char: 29.3%

Liquids: 31% Char: 27.5%

Liquids: 25% Char: 40%

Liquids: 67% Char: 11%

Liquids: 43% Char: 23% Water: 14%

Liquids: 58% Char: 18%

Liquids: 25.4% Char: 3%

Liquids: 10% Char: 39%

Liquids: 38–44% Char: 22–31%

Liquids: 23.5% Char: 19.1%

Liquids: 32–58% Char: 20–56.8%

Liquid: 62.5% Char: 25%

Liquid: 38.3% Char: 36.8%

Oil: 3.2–20.8% Char: 22.6–30.5% [28]

[61]

[62]

[63]

[64]

[65]

[66]

[23]

[20]

[67]

[68]

[41]

[57, 69]

[70]

[12]

HR: 1000 K/s Size: 0.25–1 mm

HR: 0.17 K/s Size: 0 < 1 mm

HR: 5 K/s Size: 0.6–1.8 mm

HR: 0.16–1 K/s Size: 0.075 mm

HR: >1000 K/s Size: 0.25–0.425 mm

HR: 5 K/s Size: 06–0.85 mm

> HR: 0.16 K/s Size: <1 mm

HR: 0.08–0.3 K/s Size: 420–840 μm

HR: 3.3 K/s Size: <0.5 mm

HR: 1 K/s Size: 1.18–1.8 mm

> HR: 0.166 K/s Size: 0.25 mm

HR: 0.083–0.33 K/s Size: 0.32–2 mm

HR: >4000 K/s Size: 0.25–0.5 mm

HR: 0.33 K/s Size: 0.2–0.5 mm

HR: 0.36 K/s Size: 0.25–1 mm

Slow Temp.: 500°C

Slow Temp.: 550°C

Intermediate Temp.: 500°C

Slow Temp.: 700°C

Slow Temp.: 550°C

Slow Temp.: 600°C

Product profiles from pyrolysis of different biomass in fixed and fluidized bed pyrolyzers.

Switch grass Fast/flash Temp.: 510°C

Recent Advances in Pyrolysis

Walnut shell Intermediate Temp.: 500°C

Pine wood Fast/flash Temp.: 400–500°C

Apple pomace Slow Temp.: 400°C

Coconut shell Intermediate Temp.: 500°C

Cashew nut shells Slow Temp.: 400–450°C

Cherry stones Slow Temp.: 400–500°C

P. juliflora Fast/flash Temp.: 450°C

Cogon grass Slow Temp.: 500°C

Temp., temperature; HR, heating rate; size, biomass particle size.

fluidized) are provided (Table 3).

Table 3.

50

The modeling of the packed bed and the fluidized bed pyrolyzers is discussed in detail below.

The major difference between the updraft pyrolyzer and the downdraft pyrolyzer is only the inlet and outlet locations, which changes the residence time and mixing conditions. For the purpose of the modeling of biomass pyrolysis and optimization of product profiles, the updraft and downdraft pyrolyzers are treated as fixed bed pyrolyzers.

#### 4.1 Modeling of packed bed pyrolysis reactor

The modeling of the packed bed pyrolysis reactors is divided into drying stage and the pyrolysis stage. The modeling of these two stages has been is presented in detail below.

#### 4.1.1 Modeling of drying stage in packed bed pyrolyzers

The drying of biomass particles will happen in two stages: (a) constant rate period and (b) falling rate period. The rate of drying in these two stages is given separately by the following equations.

The rate of drying during the constant rate period is given by [71]

$$\frac{dX}{dt} = -k\_c = 1.3 \times 10^{-9} T\_{\rm g}^{4.112} v\_{\rm g}^{0.219} \tag{3}$$

only the solid biomass as reactant, most of the reactions are treated as first order reactions using Arrhenius type model [73]. The Arrhenius model parameters for the

The volume and the moles of the products of the pyrolysis zone was found with

dXBiomass �rBiomass

(6)

; Cp is

) Ea (J/mol)

�rBiomass <sup>¼</sup> rGas <sup>þ</sup> rLiquids <sup>þ</sup> rChar � �

where V is the volume of the pyrolyzer in l, FBiomass is the molar feed rate of biomass in mol/s; Xf is the desired biomass conversion; �rBiomass is the rate of biomass consumption in mol/(lit. s); rGas is the rate of formation of gaseous products in mol/(lit. s); rLiquids is the rate of formation of liquid products in mol/(lit. s);

Since, the pyrolysis stage is non-isothermal in nature, it is essential to model the temperature along the length of the pyrolyzer which involves external heating. The energy balance equation was obtained based on the shell energy balance to find the temperature profile along the length of the pyrolysis zone. The steady state energy

<sup>4</sup> � <sup>T</sup><sup>4</sup> � � � <sup>Δ</sup>HRFAOdXA <sup>¼</sup> 0 (7)

) Ea (J/mol) A (s�<sup>1</sup>

lumped reaction model that produces gaseous, liquid, and char for different

constituent of biomass are given in Table 5.

DOI: http://dx.doi.org/10.5772/intechopen.85581

the help of the plug flow reactor (PFR) design equation.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

V FBiomass

rChar is the rate of formation of char in mol/(lit. s).

k d2 T dr<sup>2</sup> � <sup>ρ</sup>Cpvz

radial direction and one in axial direction.

@<sup>r</sup> <sup>¼</sup> R, k dT

@<sup>r</sup> <sup>¼</sup> <sup>0</sup>; k dT

Biomass component↓ A (s�<sup>1</sup>

\*HEMI, hemicellulose; CEL, cellulose; LIG, lignin.

Kinetic parameters for lumped models for biomass pyrolysis.

@z ¼ 0; T ¼ To

(m2 K4

Table 5.

53

balance equation results in the following differential equation:

<sup>R</sup> <sup>σ</sup> TW

The boundary conditions for solving the energy balance equation are:

Pyrolysis product! Gas Tar/liquids Char

) Ea (J/mol) A (s�<sup>1</sup>

HEMI\* 2.1 � 1016 1.8 � <sup>10</sup><sup>5</sup> 8.7 � <sup>10</sup><sup>14</sup> 2.0 � <sup>10</sup><sup>5</sup> 2.6 � 1011 1.5 � 105 CEL\* 2.8 � <sup>10</sup><sup>19</sup> 2.4 � <sup>10</sup><sup>5</sup> 3.3 � <sup>10</sup><sup>14</sup> 1.9 � 105 1.3 � <sup>10</sup><sup>10</sup> 1.5 � 105 LIG\* 9.6 � <sup>10</sup><sup>8</sup> 1.1 � 105 1.5 � 109 1.4 � 105 7.7 � 106 1.1 � 105

dr <sup>¼</sup> Radiation flux

is the reactor temperature in K; r, z are the radial and axial directions.

where k is the thermal conductivity in W/(m. K); ρ is the density in kg/m3

the specific heat capacity in J/(kg. K); vz is the gas velocity in m/s; ε<sup>R</sup> is the emissivity of the body during radiation; σ is the Stephan-Boltzmann constant = 5.67E�08 W/

); R is the radius of oxidation zone in m; TW is the source temperature in K; T

dr <sup>¼</sup> <sup>0</sup>ð Þ Tempearature minimum condition

Since, Eq. (7) is combined convection diffusion equation with heat source (heat of reaction) and external heat supply, two boundary conditions are required: one in

dT dz <sup>þ</sup> <sup>ε</sup><sup>R</sup>

¼ ð Xf

0

The rate of drying during the falling rate period is given by [71]

$$\begin{split} \frac{dX}{dt} &= -K(X - X\_{eq}) \\ K &= 0.011 \exp\left(-201.8/T\_{\rm g}\right) \end{split} \tag{4}$$

Eqs. (3) and (4), provide an estimation of rate of drying in both constant and falling rate periods, with rate of drying depending on the particle temperature. During the pyrolysis process, the biomass particle temperature depends on the rate of heat transfer between the gas and the particle. This rate of heat transfer depends on the flow characteristics involving Reynolds and Prandtl number. For 20 < Re < 1000, the heat transfer coefficient is given by [71, 72]

$$\begin{aligned} h &= \text{3.2\%C}\_{\text{pg}} \text{G}\_{\text{g}} \text{Re}^{-0.65} \text{Pr}^{2\%}\\ \mathbf{Re}\_{p} &= \frac{\rho\_{\text{gas}} U\_{o} d\_{p}}{\mu} \end{aligned} \tag{5}$$

where kc is the mass transfer coefficient at constant drying period in s�<sup>1</sup> ; K is the mass transfer coefficient at falling drying period in s�<sup>1</sup> ; Re is the Reynolds number of gas; Pr is the Prandtl number of gas; Tg is the gas temperature in °C; vg is the gas velocity in m/s; Cpg is the heat capacity of gas in J/(kg K); εmf is the porosity at minimum fluidization; X is the moisture content in the biomass; Xeq is the moisture content at the end of constant rate period; Gg is the gas mass flux in kg/(m2 s).

#### 4.1.2 Modeling of pyrolysis in packed bed reactors

Eqs. (1) and (2) represents the reactions reported during pyrolysis of biomass with lumped product approach. Since the reaction presented in Eq. (1) involves

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

The modeling of the packed bed and the fluidized bed pyrolyzers is discussed in

The major difference between the updraft pyrolyzer and the downdraft pyrolyzer is only the inlet and outlet locations, which changes the residence time and mixing conditions. For the purpose of the modeling of biomass pyrolysis and optimization of product profiles, the updraft and downdraft pyrolyzers are treated as

The modeling of the packed bed pyrolysis reactors is divided into drying stage and the pyrolysis stage. The modeling of these two stages has been is presented in

The drying of biomass particles will happen in two stages: (a) constant rate period and (b) falling rate period. The rate of drying in these two stages is given

The rate of drying during the constant rate period is given by [71]

The rate of drying during the falling rate period is given by [71]

dt ¼ �K X � Xeq

dt ¼ �kc <sup>¼</sup> <sup>1</sup>:<sup>3</sup> � <sup>10</sup>�<sup>9</sup>T<sup>4</sup>:<sup>112</sup>

K ¼ 0:011 exp �201:8=Tg

Eqs. (3) and (4), provide an estimation of rate of drying in both constant and falling rate periods, with rate of drying depending on the particle temperature. During the pyrolysis process, the biomass particle temperature depends on the rate of heat transfer between the gas and the particle. This rate of heat transfer depends on the flow characteristics involving Reynolds and Prandtl number. For 20 < Re < 1000, the heat transfer coefficient is given by [71, 72]

<sup>h</sup> <sup>¼</sup> <sup>3</sup>:26CpgGgRe�0:65Pr2<sup>=</sup><sup>3</sup>

of gas; Pr is the Prandtl number of gas; Tg is the gas temperature in °C; vg is the gas velocity in m/s; Cpg is the heat capacity of gas in J/(kg K); εmf is the porosity at minimum fluidization; X is the moisture content in the biomass; Xeq is the moisture content at the end of constant rate period; Gg is the gas mass flux in kg/(m2 s).

Eqs. (1) and (2) represents the reactions reported during pyrolysis of biomass with lumped product approach. Since the reaction presented in Eq. (1) involves

Re<sup>p</sup> <sup>¼</sup> <sup>ρ</sup>gasUodp μ

mass transfer coefficient at falling drying period in s�<sup>1</sup>

4.1.2 Modeling of pyrolysis in packed bed reactors

52

where kc is the mass transfer coefficient at constant drying period in s�<sup>1</sup>

<sup>g</sup> v<sup>0</sup>:<sup>219</sup>

<sup>g</sup> (3)

(4)

(5)

; K is the

; Re is the Reynolds number

detail below.

detail below.

fixed bed pyrolyzers.

Recent Advances in Pyrolysis

4.1 Modeling of packed bed pyrolysis reactor

4.1.1 Modeling of drying stage in packed bed pyrolyzers

dX

dX

separately by the following equations.

only the solid biomass as reactant, most of the reactions are treated as first order reactions using Arrhenius type model [73]. The Arrhenius model parameters for the lumped reaction model that produces gaseous, liquid, and char for different constituent of biomass are given in Table 5.

The volume and the moles of the products of the pyrolysis zone was found with the help of the plug flow reactor (PFR) design equation.

$$\frac{V}{F\_{Biomass}} = \int \frac{dX\_{Biomass}}{-r\_{Biomass}} \tag{6}$$

$$\left(-r\_{Biomass} = r\_{Gas} + r\_{Liquid} + r\_{Char}\right)$$

where V is the volume of the pyrolyzer in l, FBiomass is the molar feed rate of biomass in mol/s; Xf is the desired biomass conversion; �rBiomass is the rate of biomass consumption in mol/(lit. s); rGas is the rate of formation of gaseous products in mol/(lit. s); rLiquids is the rate of formation of liquid products in mol/(lit. s); rChar is the rate of formation of char in mol/(lit. s).

Since, the pyrolysis stage is non-isothermal in nature, it is essential to model the temperature along the length of the pyrolyzer which involves external heating. The energy balance equation was obtained based on the shell energy balance to find the temperature profile along the length of the pyrolysis zone. The steady state energy balance equation results in the following differential equation:

$$k\frac{d^2T}{dr^2} - \rho \mathbf{C}\_p v\_x \frac{dT}{dz} + \frac{e\_R}{R} \sigma \left(T\_W{}^4 - T^4\right) - \Delta H\_R F\_{AO} dX\_A = 0 \tag{7}$$

where k is the thermal conductivity in W/(m. K); ρ is the density in kg/m3 ; Cp is the specific heat capacity in J/(kg. K); vz is the gas velocity in m/s; ε<sup>R</sup> is the emissivity of the body during radiation; σ is the Stephan-Boltzmann constant = 5.67E�08 W/ (m2 K4 ); R is the radius of oxidation zone in m; TW is the source temperature in K; T is the reactor temperature in K; r, z are the radial and axial directions.

Since, Eq. (7) is combined convection diffusion equation with heat source (heat of reaction) and external heat supply, two boundary conditions are required: one in radial direction and one in axial direction.

The boundary conditions for solving the energy balance equation are:

$$\begin{aligned} \@{} \@{} r &= R, \, k \frac{dT}{dr} = \text{Radiation flux} \\ \@{} \@{} r &= 0; \, k \frac{dT}{dr} = 0 \, (Tempearature \, minimum \, condition) \\ \@{} \@{} x &= 0; \, T = T\_o \end{aligned}$$


#### Table 5.

Kinetic parameters for lumped models for biomass pyrolysis.

## 4.2 Modeling of fluidized bed pyrolysis reactor

As mentioned in the modeling of the packed bed pyrolysis, the pyrolysis process occurs after the drying of the biomass.

Based on the two-phase model, the overall mass transfer coefficient is estimated

þ 5:85

DAB 0:5 g0:<sup>25</sup> d1:<sup>25</sup> b

DABεmfUb d3 b

)

¼ 1 kcb þ 1 kdc

Re<sup>p</sup> <sup>¼</sup> <sup>ρ</sup>gasUodp μ

<sup>p</sup>ρgas <sup>ρ</sup>solid � <sup>ρ</sup>gas � �<sup>g</sup>

where kdc is the mass transfer coefficient between dense emulsion and cloud in m/s; kcb is the mass transfer coefficient between bubble and cloud in m/s; koverall is the overall mass transfer coefficient between bubble and cloud in m/s; hc is the overall heat transfer coefficient in W/(m2 K); Re<sup>p</sup> is the Particle Reynolds number; Ar is the Archimedes number; Umbf is the bubble velocity at minimum fluidization

the bed voidage at minimum fluidization; ρsolid is the density of biomass particle in

Due to vigorous mixing in the fluidized beds, the following assumptions are

3. Reactions occurring only in the solid phase (only in the emulsion phase).

and bubble phase) can be obtained from the fundamental component balance:

Based on these assumptions, the following models (separately for emulsion phase

made for modeling of the pyrolysis stage in the fluidized bed reactors:

The heat transfer coefficient is estimated using the correlation given in Eq. (11) [76].

<sup>0</sup>:<sup>35</sup>Ar<sup>0</sup>:<sup>25</sup> Vgas ≤ Uo

� � (11)

<sup>μ</sup><sup>2</sup> (13)

/s; Ub is the bubble velocity in m/s; εmf is

; db is the bubble diameter in m.

!

!

(8)

(9)

(10)

(12)

using Eqs. (8)–(10) [75].

Cloud-bubble mass transfer coefficient

DOI: http://dx.doi.org/10.5772/intechopen.85581

Dense-cloud mass transfer coefficient

Overall mass transfer coefficient (in s�<sup>1</sup>

hc ¼ 0:15

Ar ¼

; ρgas is the density of fluidizing gas in kg/m<sup>3</sup>

4.2.2 Modeling of pyrolysis stage in fluidized bed pyrolyzers

in m/s; DAB is the binary diffusivity in m<sup>2</sup>

1. Isothermal operation of the reactor,

2. No radial variations, and

kg/m<sup>3</sup>

55

kcb ¼ 1:5

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

Umbf db

kdc ¼ 6:77

1 koverall

kg dp Re<sup>p</sup>

d3

### 4.2.1 Modeling of drying stage in fluidized bed pyrolyzers

The equations for the drying stage in a fluidized bed pyrolyzers are similar to those in a packed bed pyrolyzer, as provided in Section 4.1.1. Because there is vigorous mixing in the fluidized bed pyrolysis, the extent of heat and mass transfer is very high. Kunii and Levenspiel confirmed that there is more than one phase during bubbling fluidization [74]. These two phases are named as bubble phase (primarily the fluid used as fluidizing the medium) and emulsion phase (mixture of biomass particles and fluid). The interface between these two phases is named as "cloud." The concept of two-phase model along with the mass transfer between the two phases is depicted in Figure 2. It is to be noted that the primary reactions during biomass pyrolysis shall occur only in the solid phase (i.e., in the emulsion phase) and the secondary reactions occur in both phases.

The most widely used Kunii and Levenspiel model (K-L model) expresses the overall heat and mass transfer coefficient in a bubbling fluidized bed considering the resistance for heat and mass transfer between the bubble-cloud interface and resistance for heat and mass transfer between emulsion and cloud [74]. The cloudbubble interface heat and mass transfer are functions of gas velocity and conduction/diffusion from a thin cloud layer into the bubble. The emulsion and cloud transfer are only due to conduction/diffusion between the emulsion phase and the cloud boundary. They had also suggested additional mass transfer resulting from particles dispersed in the bubbles. However, recent advanced imaging technique, have shown bubble free particles in most cases.

Figure 2. The two-phase model for fluidized bed pyrolyzer.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

Based on the two-phase model, the overall mass transfer coefficient is estimated using Eqs. (8)–(10) [75].

Cloud-bubble mass transfer coefficient

4.2 Modeling of fluidized bed pyrolysis reactor

4.2.1 Modeling of drying stage in fluidized bed pyrolyzers

phase) and the secondary reactions occur in both phases.

have shown bubble free particles in most cases.

Figure 2.

54

The two-phase model for fluidized bed pyrolyzer.

occurs after the drying of the biomass.

Recent Advances in Pyrolysis

As mentioned in the modeling of the packed bed pyrolysis, the pyrolysis process

The equations for the drying stage in a fluidized bed pyrolyzers are similar to those in a packed bed pyrolyzer, as provided in Section 4.1.1. Because there is vigorous mixing in the fluidized bed pyrolysis, the extent of heat and mass transfer is very high. Kunii and Levenspiel confirmed that there is more than one phase during bubbling fluidization [74]. These two phases are named as bubble phase (primarily the fluid used as fluidizing the medium) and emulsion phase (mixture of biomass particles and fluid). The interface between these two phases is named as "cloud." The concept of two-phase model along with the mass transfer between the two phases is depicted in Figure 2. It is to be noted that the primary reactions during biomass pyrolysis shall occur only in the solid phase (i.e., in the emulsion

The most widely used Kunii and Levenspiel model (K-L model) expresses the overall heat and mass transfer coefficient in a bubbling fluidized bed considering the resistance for heat and mass transfer between the bubble-cloud interface and resistance for heat and mass transfer between emulsion and cloud [74]. The cloudbubble interface heat and mass transfer are functions of gas velocity and conduction/diffusion from a thin cloud layer into the bubble. The emulsion and cloud transfer are only due to conduction/diffusion between the emulsion phase and the cloud boundary. They had also suggested additional mass transfer resulting from particles dispersed in the bubbles. However, recent advanced imaging technique,

$$k\_{cb} = 1.5 \frac{U\_{mbf}}{d\_b} + 5.85 \left(\frac{D\_{AB} \, ^{0.5} \text{g}^{0.25}}{d\_b^{1.25}}\right) \tag{8}$$

### Dense-cloud mass transfer coefficient

$$k\_{dc} = 6.77 \left( \frac{D\_{AB} \varepsilon\_{mf} U\_b}{d\_b^3} \right) \tag{9}$$

Overall mass transfer coefficient (in s�<sup>1</sup> )

$$\frac{1}{k\_{overall}} = \frac{1}{k\_{cb}} + \frac{1}{k\_{dc}}\tag{10}$$

The heat transfer coefficient is estimated using the correlation given in Eq. (11) [76].

$$h\_c = 0.15 \frac{k\_g}{d\_p} \text{Re}\_p^{0.35} Ar^{0.25} \left(V\_{gas} \le U\_o\right) \tag{11}$$

$$\text{Re}\_p = \frac{\rho\_{gas} U\_o d\_p}{\mu} \tag{12}$$

$$Ar = \frac{d\_p^3 \rho\_{gas} \left(\rho\_{solid} - \rho\_{gas}\right) \mathbf{g}}{\mu^2} \tag{13}$$

where kdc is the mass transfer coefficient between dense emulsion and cloud in m/s; kcb is the mass transfer coefficient between bubble and cloud in m/s; koverall is the overall mass transfer coefficient between bubble and cloud in m/s; hc is the overall heat transfer coefficient in W/(m2 K); Re<sup>p</sup> is the Particle Reynolds number; Ar is the Archimedes number; Umbf is the bubble velocity at minimum fluidization in m/s; DAB is the binary diffusivity in m<sup>2</sup> /s; Ub is the bubble velocity in m/s; εmf is the bed voidage at minimum fluidization; ρsolid is the density of biomass particle in kg/m<sup>3</sup> ; ρgas is the density of fluidizing gas in kg/m<sup>3</sup> ; db is the bubble diameter in m.

#### 4.2.2 Modeling of pyrolysis stage in fluidized bed pyrolyzers

Due to vigorous mixing in the fluidized beds, the following assumptions are made for modeling of the pyrolysis stage in the fluidized bed reactors:


Based on these assumptions, the following models (separately for emulsion phase and bubble phase) can be obtained from the fundamental component balance:

$$
abla \text{bble phase}: \; \epsilon u\_b \frac{d \mathbf{C}\_{A,b}}{dz} - k\_L (\mathbf{C}\_{A,b} - \mathbf{C}\_{A,\epsilon}) = \mathbf{0} \tag{14}$$

component (i.e., x1 = 100%) while its opposite edge holds a value of x1 = 0 (with

However, it may be noted that in certain experiments the sum of proportions of the three components may be deliberately constrained to equal a specific value. This feature has been used in our experimental design. For example, referring to Figure 3, the three vertices represent hemicellulose, cellulose and lignin. The maximum fraction of these three constituents is 0.764, 0.742 and 0.7444 for hemicellulose, cellulose and lignin, respectively. The 0.119, 0.117, and 0.139 marked on the sides of the triangle represent the minimum values for the biomass constituents.

The analysis of minimization of gas yield is separated into two types: (a) fast pyrolysis in fluidized bed pyrolyzer and (b) slow pyrolysis in fixed bed pyrolyzer. These two were chosen to account for the extremities of the process conditions. The contour plot for fast pyrolysis in fluidized bed is shown in Figure 3. The three vertices of the triangle are the maximum points of hemicellulose, cellulose, and lignin. For the minimization of gaseous products during fast pyrolysis in fluidized bed, the hemicellulose content in biomass is required to be high, when compared to the cellulose and lignin. More than 40% hemicellulose and 20% of each of lignin and cellulose in the biomass feed will minimize the gas yield during fast pyrolysis in fluidized bed. This is easily achievable, because there are biomasses with hemicellulose content higher than 40% (please refer to Table 1). Considering fast pyrolysis of wheat straw in fluidized bed reactor, which contains 25.5% hemicellulose, 49.3% cellulose, and 25.2% of lignin, the gas yield is 30%, which is consistent with the

The gas yield during slow pyrolysis in fixed bed pyrolysis is higher than that of fast pyrolysis in fluidized bed reactors. This is due to the fact that the residence time in the fixed bed reactors is higher when compared to that of fluidized bed reactors. Due to longer residence times, there are secondary reactions, that is, the conversion of liquid into gas and char. The optimal minimum gas yield during slow biomass pyrolysis in fixed bed reactors is 16%, as shown in contour plot in Figure 4. The minimum gas yield can be obtained at various conditions. Specifically, it is obtained either at high hemicellulose content or at medium cellulose and lignin content in the

x2 + x3 = 100%). The same holds for the other two vertices and edges.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

DOI: http://dx.doi.org/10.5772/intechopen.85581

5.2 Optimization of gas yield

Figure 3.

57

experimentally obtained data presented in Table 3.

Contour plot of gas yield in biomass pyrolysis for fast pyrolysis in fluidized bed reactor.

$$\text{Eumulation phase}: \ (\mathbf{1} - \varepsilon)u\_{\varepsilon} \frac{d\mathbf{C}\_{A,\varepsilon}}{dx} + k\_{L}(\mathbf{C}\_{A,b} - \mathbf{C}\_{A,\varepsilon}) - (\mathbf{1} - \varepsilon)\mathbf{D}\_{\text{eff}}\frac{d^{2}\mathbf{C}\_{A,\varepsilon}}{dx^{2}} + (-r\_{A})\rho\_{\varepsilon}(\mathbf{1} - \varepsilon) = \mathbf{0};\tag{15}$$

where b is the bubble phase and e is the emulsion phase.

#### 5. Optimization of product profiles in biomass pyrolysis

As discussed in Section 1, since the product profiles of biomass pyrolysis are known to greatly depend on various parameters, it is necessary to determine the parameters that have the largest effect on the ratios of important lumped products discussed in Section 2.

#### 5.1 Response surface optimization methodology: the mixture design

The following quantifiable factors were chosen for the optimization. The biomass-based factors are:

a. Hemicellulose fraction

b.Cellulose fraction

c. Lignin fraction

The process-based parameters have been divided into two categorical type parameters.

a. The type of pyrolysis-slow and fast/flash pyrolysis

b.The type of reactor-fixed bed and fluidized bed reactors

The variation of constituent composition of biomass is obtained from Table 1. The amounts of gaseous, liquid, and products for each biomass for different process parameters and different reactor types are obtained from Table 3.

Since hemicellulose, cellulose, and lignin are biomass constituents, the sum of their fractions must be equal to one. Accordingly, a mixture design was chosen for the statistical design of the experiments.

A total of 52 data sets were selected combining Tables 1 and 3 (based on the variations in mixture design), and the experiments conducted and data analyzed.

The depletion of fossil fuels has created interest in obtaining the fuels from alternate sources such as biomass, especially for transportation fuels. Therefore, the objective of the product profiles from biomass pyrolysis is aimed to maximize the liquid (tar) products and simultaneously minimize the production of gas and char. The reason for the minimization of gaseous products is that they may contain carbon monoxide and carbon dioxide which are of less calorific value when compared to the hydrocarbons.

Ternary diagrams are common in representing factor levels in a mixture design. In such a diagram, with three factors (x1, x2, and x3), the vertex represents a pure

### Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

component (i.e., x1 = 100%) while its opposite edge holds a value of x1 = 0 (with x2 + x3 = 100%). The same holds for the other two vertices and edges.

However, it may be noted that in certain experiments the sum of proportions of the three components may be deliberately constrained to equal a specific value. This feature has been used in our experimental design. For example, referring to Figure 3, the three vertices represent hemicellulose, cellulose and lignin. The maximum fraction of these three constituents is 0.764, 0.742 and 0.7444 for hemicellulose, cellulose and lignin, respectively. The 0.119, 0.117, and 0.139 marked on the sides of the triangle represent the minimum values for the biomass constituents.

## 5.2 Optimization of gas yield

Bubble phase : εub

dCA,e

where b is the bubble phase and e is the emulsion phase.

5. Optimization of product profiles in biomass pyrolysis

5.1 Response surface optimization methodology: the mixture design

The following quantifiable factors were chosen for the optimization.

The process-based parameters have been divided into two categorical type

The variation of constituent composition of biomass is obtained from Table 1. The amounts of gaseous, liquid, and products for each biomass for different process

Since hemicellulose, cellulose, and lignin are biomass constituents, the sum of their fractions must be equal to one. Accordingly, a mixture design was chosen for

A total of 52 data sets were selected combining Tables 1 and 3 (based on the variations in mixture design), and the experiments conducted and data analyzed. The depletion of fossil fuels has created interest in obtaining the fuels from alternate sources such as biomass, especially for transportation fuels. Therefore, the objective of the product profiles from biomass pyrolysis is aimed to maximize the liquid (tar) products and simultaneously minimize the production of gas and char. The reason for the minimization of gaseous products is that they may contain carbon monoxide and carbon dioxide which are of less calorific value when

Ternary diagrams are common in representing factor levels in a mixture design. In such a diagram, with three factors (x1, x2, and x3), the vertex represents a pure

a. The type of pyrolysis-slow and fast/flash pyrolysis

b.The type of reactor-fixed bed and fluidized bed reactors

parameters and different reactor types are obtained from Table 3.

Emulsion phase : ð Þ 1 � ε ue

Recent Advances in Pyrolysis

discussed in Section 2.

The biomass-based factors are:

the statistical design of the experiments.

compared to the hydrocarbons.

56

a. Hemicellulose fraction

b.Cellulose fraction

c. Lignin fraction

parameters.

dCA,b

dz <sup>þ</sup> kLð Þ� CA, <sup>b</sup> � CA,e ð Þ <sup>1</sup> � <sup>ε</sup> Deff

As discussed in Section 1, since the product profiles of biomass pyrolysis are known to greatly depend on various parameters, it is necessary to determine the parameters that have the largest effect on the ratios of important lumped products

dz � kLð Þ¼ CA,b � CA,e 0 (14)

dz<sup>2</sup> þ �ð Þ rA <sup>ρ</sup>eð Þ¼ <sup>1</sup> � <sup>ε</sup> <sup>0</sup>;

(15)

d2 CA,e

> The analysis of minimization of gas yield is separated into two types: (a) fast pyrolysis in fluidized bed pyrolyzer and (b) slow pyrolysis in fixed bed pyrolyzer. These two were chosen to account for the extremities of the process conditions. The contour plot for fast pyrolysis in fluidized bed is shown in Figure 3. The three vertices of the triangle are the maximum points of hemicellulose, cellulose, and lignin. For the minimization of gaseous products during fast pyrolysis in fluidized bed, the hemicellulose content in biomass is required to be high, when compared to the cellulose and lignin. More than 40% hemicellulose and 20% of each of lignin and cellulose in the biomass feed will minimize the gas yield during fast pyrolysis in fluidized bed. This is easily achievable, because there are biomasses with hemicellulose content higher than 40% (please refer to Table 1). Considering fast pyrolysis of wheat straw in fluidized bed reactor, which contains 25.5% hemicellulose, 49.3% cellulose, and 25.2% of lignin, the gas yield is 30%, which is consistent with the experimentally obtained data presented in Table 3.

> The gas yield during slow pyrolysis in fixed bed pyrolysis is higher than that of fast pyrolysis in fluidized bed reactors. This is due to the fact that the residence time in the fixed bed reactors is higher when compared to that of fluidized bed reactors. Due to longer residence times, there are secondary reactions, that is, the conversion of liquid into gas and char. The optimal minimum gas yield during slow biomass pyrolysis in fixed bed reactors is 16%, as shown in contour plot in Figure 4. The minimum gas yield can be obtained at various conditions. Specifically, it is obtained either at high hemicellulose content or at medium cellulose and lignin content in the

Figure 3. Contour plot of gas yield in biomass pyrolysis for fast pyrolysis in fluidized bed reactor.

#### Figure 4.

Contour plot of gas yield in biomass pyrolysis for slow pyrolysis in fixed bed reactor.

biomass. Considering the wheat straw decomposition during slow pyrolysis in fixed bed, the gas yield is 16–26% as given by Figure 4.

content is high, the liquid yield ranges from 40 to 60%. For example, one of the biomasses containing high hemi cellulose is switch grass, for which, if decomposed during fast pyrolysis in fluidized bed, the liquid yield is more than 51% (from

Figure 6 shows the liquid yield during slow pyrolysis in fixed bed pyrolyzes. Comparing Figure 5 with Figure 6, it can be concluded that, the liquid yield is higher in fast pyrolysis in fluidized bed than slow pyrolysis in fixed bed. Further, the maximum liquid yield that can be obtained is 40%. This can be obtained only at high hemicellulose content, medium cellulose content and low lignin content. The liquid yield for the high hemicellulose content biomasses, such as switch grass,

Figure 5), which is consistent with the experimentally obtained data.

Contour plot of liquid yield in biomass pyrolysis for slow in fixed bed reactor.

Contour plot of liquid yield in biomass pyrolysis for fast pyrolysis in fluidized bed reactor.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

DOI: http://dx.doi.org/10.5772/intechopen.85581

Figure 5.

Figure 6.

59

The gas yield during biomass pyrolysis may be obtained from statistical modeling, presented in Eq. (16). This equation was obtained after removing the model terms that were not statistically significant, that is, terms with p-value < 0.05.

$$\text{GasYield}(\text{\%}) = -20.4 \, ^\circ H + 108.8 \, ^\circ C + 111.3 \, ^\circ L - 377.8 \, ^\circ C \, ^\circ L - 18.3 \, ^\circ H \, ^\circ P + 50.1 \, ^\circ H \, ^\circ L \, ^\circ R \tag{16}$$

where H is the hemicellulose fraction in biomass; C is the cellulose fraction in biomass; L is the lignin fraction in biomass; P is the pyrolysis type (fast or slow); R is the reactor type (fluidized bed or fixed bed).

The coefficients present in Eq. (16) represents whether the effect is positive or negative. The coefficient for hemicellulose is negative. This means, the higher the hemicellulose content, lesser will be the gas yield. From Eq. (16), there are interactions between cellulose-lignin, hemicellulose-pyrolysis type and hemicelluloselignin-reactor type. These interactions could be either synergistic (positive coefficient) or antagonistic (negative coefficient). For the gas yield%, the cellulose-lignin and hemicellulose-pyrolysis are antagonistic, whereas hemicellulose-lignin-reactor type is synergistic.

Equations similar to gas yield%, presented in Eq. (16) were also obtained for liquid yield% and char yield%.

#### 5.3 Optimization of liquid yield

As mentioned in Section 5.1, it is important find the level of biomass constituents and operating conditions for pyrolysis to maximize the liquid yield, while minimizing both gas and char yield.

Figure 5 presents the contour plot of liquid yield as function of biomass constituents during fast pyrolysis. The maximum liquid yield may be obtained at high hemicellulose, low lignin and low cellulose content. For biomasses whose cellulose content is high, the liquid yields are very low and for the biomasses whose lignin

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

#### Figure 5.

biomass. Considering the wheat straw decomposition during slow pyrolysis in fixed

GasYieldð Þ¼� % <sup>20</sup>:4<sup>∗</sup> <sup>H</sup> <sup>þ</sup> <sup>108</sup>:8<sup>∗</sup><sup>C</sup> <sup>þ</sup> <sup>111</sup>:<sup>3</sup> <sup>∗</sup> <sup>L</sup> � <sup>377</sup>:8<sup>∗</sup>C<sup>∗</sup> <sup>L</sup> � <sup>18</sup>:<sup>3</sup> <sup>∗</sup> <sup>H</sup><sup>∗</sup> <sup>P</sup> <sup>þ</sup> <sup>50</sup>:<sup>1</sup> <sup>∗</sup> <sup>H</sup><sup>∗</sup> <sup>L</sup><sup>∗</sup> <sup>R</sup>

where H is the hemicellulose fraction in biomass; C is the cellulose fraction in biomass; L is the lignin fraction in biomass; P is the pyrolysis type (fast or slow); R

The coefficients present in Eq. (16) represents whether the effect is positive or negative. The coefficient for hemicellulose is negative. This means, the higher the hemicellulose content, lesser will be the gas yield. From Eq. (16), there are interactions between cellulose-lignin, hemicellulose-pyrolysis type and hemicelluloselignin-reactor type. These interactions could be either synergistic (positive coefficient) or antagonistic (negative coefficient). For the gas yield%, the cellulose-lignin and hemicellulose-pyrolysis are antagonistic, whereas hemicellulose-lignin-reactor

Equations similar to gas yield%, presented in Eq. (16) were also obtained for

As mentioned in Section 5.1, it is important find the level of biomass constituents and operating conditions for pyrolysis to maximize the liquid yield, while minimiz-

Figure 5 presents the contour plot of liquid yield as function of biomass constit-

uents during fast pyrolysis. The maximum liquid yield may be obtained at high hemicellulose, low lignin and low cellulose content. For biomasses whose cellulose content is high, the liquid yields are very low and for the biomasses whose lignin

(16)

The gas yield during biomass pyrolysis may be obtained from statistical modeling, presented in Eq. (16). This equation was obtained after removing the model terms that were not statistically significant, that is, terms with p-value < 0.05.

bed, the gas yield is 16–26% as given by Figure 4.

Contour plot of gas yield in biomass pyrolysis for slow pyrolysis in fixed bed reactor.

is the reactor type (fluidized bed or fixed bed).

type is synergistic.

Figure 4.

Recent Advances in Pyrolysis

liquid yield% and char yield%.

5.3 Optimization of liquid yield

ing both gas and char yield.

58

Contour plot of liquid yield in biomass pyrolysis for fast pyrolysis in fluidized bed reactor.

content is high, the liquid yield ranges from 40 to 60%. For example, one of the biomasses containing high hemi cellulose is switch grass, for which, if decomposed during fast pyrolysis in fluidized bed, the liquid yield is more than 51% (from Figure 5), which is consistent with the experimentally obtained data.

Figure 6 shows the liquid yield during slow pyrolysis in fixed bed pyrolyzes. Comparing Figure 5 with Figure 6, it can be concluded that, the liquid yield is higher in fast pyrolysis in fluidized bed than slow pyrolysis in fixed bed. Further, the maximum liquid yield that can be obtained is 40%. This can be obtained only at high hemicellulose content, medium cellulose content and low lignin content. The liquid yield for the high hemicellulose content biomasses, such as switch grass,

during slow pyrolysis in fixed bed reactor, the liquid yield will vary from 25 to 40%, which can be obtained from Figure 6.

obtained during fast pyrolysis in fluidized bed from biomasses containing higher amount of lignin, medium amounts of hemicellulose and lower amounts of cellulose (lignocellulosic biomass in other words). For example, the high lignin biomasses such as hazelnut shell will yield less char during fast pyrolysis in fluidized bed

lignin during slow pyrolysis in fixed bed operation.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

DOI: http://dx.doi.org/10.5772/intechopen.85581

Low char yields may be obtained for biomasses with high hemicellulose and high

Biomasses vary widely in their chief constituents namely, hemicellulose, cellu-

lose, and lignin. The order of the degradation temperature for the three chief constituents is hemicellulose < cellulose < lignin. For obtaining controlled and useful product profiles such as liquid fuels via pyrolysis, it is important to understand the pyrolysis process conditions, such as heating rate and temperature. For optimization, the pyrolysis operating parameters are classified as slow and fast pyrolysis and the reactors are classified as fluidized bed and fixed bed. The data required for optimization was obtained from many sources, including experimental data for various biomasses and process conditions. The response surface optimization was performed using mixture design with three levels for biomass constituents and two levels each for pyrolysis type and reactor type. The objective of this book chapter is to model the reaction using lumped (gas, liquids, and char) approach and find the feed and operating parameters to maximize the liquid yield. In general, the liquid yield was found to be higher during fast pyrolysis/fluidized bed reactors when compared to slow pyrolysis/fixed bed reactors. This is due to the large residence time in the slow pyrolysis/fixed bed reactors. The maximum yield may be obtained for the biomasses containing high hemicellulose such as switch grass and

The authors recommend further optimization with respect to particle size,

The authors acknowledge the financial support from the Center of Excellence in Advanced Materials & Green Technologies, granted under the Frontier Areas of Science and Technology (FAST) scheme to establish the Centers of Excellence by Ministry of Human Resource Development (MHRD), Government of India.

The authors would like to thank the students who attended the course "Waste to

Energy part of Masters in Materials Science and Engineering program, Amrita Vishwa Vidyapeetham," who had helped in obtaining the data and supported the

reactor.

6. Conclusions

corn stock.

heating rate, and biomass feed rate.

There is no conflict of interest.

Notes/Thanks/Other declarations

Acknowledgements

Conflict of interest

work through discussions.

61

## 5.4 Optimization of char yield

Minimization of char is also important during biomass pyrolysis. Figure 7 shows the contour plot of char yield during fast pyrolysis in fluidized bed and Figure 8 shows the char yield during slow pyrolysis in fixed bed. Lower char yields may be

#### Figure 7.

Contour plot of char yield in biomass pyrolysis for fast pyrolysis in fluidized bed reactor.

Contour plot of char yield in biomass pyrolysis for slow pyrolysis in fixed bed reactor.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

obtained during fast pyrolysis in fluidized bed from biomasses containing higher amount of lignin, medium amounts of hemicellulose and lower amounts of cellulose (lignocellulosic biomass in other words). For example, the high lignin biomasses such as hazelnut shell will yield less char during fast pyrolysis in fluidized bed reactor.

Low char yields may be obtained for biomasses with high hemicellulose and high lignin during slow pyrolysis in fixed bed operation.

## 6. Conclusions

during slow pyrolysis in fixed bed reactor, the liquid yield will vary from 25 to 40%,

Minimization of char is also important during biomass pyrolysis. Figure 7 shows the contour plot of char yield during fast pyrolysis in fluidized bed and Figure 8 shows the char yield during slow pyrolysis in fixed bed. Lower char yields may be

Contour plot of char yield in biomass pyrolysis for fast pyrolysis in fluidized bed reactor.

Contour plot of char yield in biomass pyrolysis for slow pyrolysis in fixed bed reactor.

which can be obtained from Figure 6.

5.4 Optimization of char yield

Recent Advances in Pyrolysis

Figure 7.

Figure 8.

60

Biomasses vary widely in their chief constituents namely, hemicellulose, cellulose, and lignin. The order of the degradation temperature for the three chief constituents is hemicellulose < cellulose < lignin. For obtaining controlled and useful product profiles such as liquid fuels via pyrolysis, it is important to understand the pyrolysis process conditions, such as heating rate and temperature. For optimization, the pyrolysis operating parameters are classified as slow and fast pyrolysis and the reactors are classified as fluidized bed and fixed bed. The data required for optimization was obtained from many sources, including experimental data for various biomasses and process conditions. The response surface optimization was performed using mixture design with three levels for biomass constituents and two levels each for pyrolysis type and reactor type. The objective of this book chapter is to model the reaction using lumped (gas, liquids, and char) approach and find the feed and operating parameters to maximize the liquid yield. In general, the liquid yield was found to be higher during fast pyrolysis/fluidized bed reactors when compared to slow pyrolysis/fixed bed reactors. This is due to the large residence time in the slow pyrolysis/fixed bed reactors. The maximum yield may be obtained for the biomasses containing high hemicellulose such as switch grass and corn stock.

The authors recommend further optimization with respect to particle size, heating rate, and biomass feed rate.

#### Acknowledgements

The authors acknowledge the financial support from the Center of Excellence in Advanced Materials & Green Technologies, granted under the Frontier Areas of Science and Technology (FAST) scheme to establish the Centers of Excellence by Ministry of Human Resource Development (MHRD), Government of India.

### Conflict of interest

There is no conflict of interest.

#### Notes/Thanks/Other declarations

The authors would like to thank the students who attended the course "Waste to Energy part of Masters in Materials Science and Engineering program, Amrita Vishwa Vidyapeetham," who had helped in obtaining the data and supported the work through discussions.

Recent Advances in Pyrolysis

## Author details

Udaya Bhaskar Reddy Ragula\*, Sriram Devanathan and Sindhu Subramanian Department of Chemical Engineering and Materials Science, Amrita Vishwa Vidyapeetham, Coimbatore, India

References

ef5027373

[1] Basu P. Biomass Gasification and Pyrolysis Handbook. Kidlington, Oxford, UK: Elsevier; 2010. DOI: 10.1016/B978-0-12-374988-8.00001-5

DOI: http://dx.doi.org/10.5772/intechopen.85581

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

pyrolysis kinetics of apple pomace as waste biomass. International Journal of Hydrogen Energy. 2014;39:16619-16627. DOI: 10.1016/j.ijhydene.2014.06.012

[9] Özsin G, Pütün AE. Kinetics and evolved gas analysis for pyrolysis of food processing wastes using TGA/ MS/FT-IR. Waste Management. 2017;

[10] Chandrasekaran A, Ramachandran S, Subbiah S. Determination of kinetic parameters in the pyrolysis operation and thermal behavior of Prosopis juliflora using thermogravimetric analysis. Bioresource Technology. 2017;

64:315-326. DOI: 10.1016/j. wasman.2017.03.020

233:413-422. DOI: 10.1016/j.

[11] Tsamba AJ, Yang W, Blasiak W. Pyrolysis characteristics and global kinetics of coconut and cashew nut shells. Fuel Processing Technology. 2006;87:523-530. DOI: 10.1016/j.

[12] Li LL, Fu XB, Wang XN, Tian YY, Qin S. Pyrolytic characteristics and kinetic studies of agricultural wastesfour kinds of grasses. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2016;38: 1156-1162. DOI: 10.1080/ 15567036.2013.856967

[13] Islam MA, Asif M, Hameed BH. Pyrolysis kinetics of raw and hydrothermally carbonized Karanj (Pongamia pinnata) fruit hulls via thermogravimetric analysis. Bioresource Technology. 2015;179:227-233. DOI: 10.1016/j.biortech.2014.11.115

[14] Subramanian S, Ragula UBR. Pyrolysis kinetics of Hibiscus rosa sinensis and Nerium oleander. Biofuels. 2018; 7269:1-15. DOI: 10.1080/17597269.

2018.1432274

biortech.2017.02.119

fuproc.2005.12.002

Christopher J, Lamba B, Tuli DK, et al. Physical and chemical characterization of various indian agriculture residues for biofuels production. Energy & Fuels. 2015;29:3111-3118. DOI: 10.1021/

comprehensive review on the pyrolysis of lignocellulosic biomass. Renewable Energy. 2018;129:695-716. DOI: 10.1016/j.renene.2017.04.035

[4] Jahirul MI, Rasul MG, Chowdhury AA, Ashwath N. Biofuels production through biomass pyrolysis—A

technological review. Energies. 2012;5: 4952-5001. DOI: 10.3390/en5124952

[5] Braga RM, Melo DMA, Aquino FM, Freitas JCO, Melo MAF, Barros JMF, et al. Characterization and comparative study of pyrolysis kinetics of the rice husk and the elephant grass. Journal of Thermal Analysis and Calorimetry. 2014;115:1915-1920. DOI: 10.1007/

[6] Uzun BB, Yaman E. Pyrolysis kinetics of walnut shell and waste polyolefins using thermogravimetric analysis. Journal of the Energy Institute.

2016;90:1-13. DOI: 10.1016/j.

Forest Refine. 2013;3:1-4

[7] Räsänen T, Athanassiadis D. Basic chemical composition of the biomass components of pine, spruce and birch.

[8] Guerrero MRB, Marques Da Silva Paula M, Zaragoza MM, Gutiérrez JS, Velderrain VG, Ortiz AL, et al. Thermogravimetric study on the

s10973-013-3503-7

joei.2016.09.001

63

[2] Raj T, Kapoor M, Gaur R,

[3] Dhyani V, Bhaskar T. A

\*Address all correspondence to: u\_bhaskarreddy@cb.amrita.edu

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

## References

[1] Basu P. Biomass Gasification and Pyrolysis Handbook. Kidlington, Oxford, UK: Elsevier; 2010. DOI: 10.1016/B978-0-12-374988-8.00001-5

[2] Raj T, Kapoor M, Gaur R, Christopher J, Lamba B, Tuli DK, et al. Physical and chemical characterization of various indian agriculture residues for biofuels production. Energy & Fuels. 2015;29:3111-3118. DOI: 10.1021/ ef5027373

[3] Dhyani V, Bhaskar T. A comprehensive review on the pyrolysis of lignocellulosic biomass. Renewable Energy. 2018;129:695-716. DOI: 10.1016/j.renene.2017.04.035

[4] Jahirul MI, Rasul MG, Chowdhury AA, Ashwath N. Biofuels production through biomass pyrolysis—A technological review. Energies. 2012;5: 4952-5001. DOI: 10.3390/en5124952

[5] Braga RM, Melo DMA, Aquino FM, Freitas JCO, Melo MAF, Barros JMF, et al. Characterization and comparative study of pyrolysis kinetics of the rice husk and the elephant grass. Journal of Thermal Analysis and Calorimetry. 2014;115:1915-1920. DOI: 10.1007/ s10973-013-3503-7

[6] Uzun BB, Yaman E. Pyrolysis kinetics of walnut shell and waste polyolefins using thermogravimetric analysis. Journal of the Energy Institute. 2016;90:1-13. DOI: 10.1016/j. joei.2016.09.001

[7] Räsänen T, Athanassiadis D. Basic chemical composition of the biomass components of pine, spruce and birch. Forest Refine. 2013;3:1-4

[8] Guerrero MRB, Marques Da Silva Paula M, Zaragoza MM, Gutiérrez JS, Velderrain VG, Ortiz AL, et al. Thermogravimetric study on the

pyrolysis kinetics of apple pomace as waste biomass. International Journal of Hydrogen Energy. 2014;39:16619-16627. DOI: 10.1016/j.ijhydene.2014.06.012

[9] Özsin G, Pütün AE. Kinetics and evolved gas analysis for pyrolysis of food processing wastes using TGA/ MS/FT-IR. Waste Management. 2017; 64:315-326. DOI: 10.1016/j. wasman.2017.03.020

[10] Chandrasekaran A, Ramachandran S, Subbiah S. Determination of kinetic parameters in the pyrolysis operation and thermal behavior of Prosopis juliflora using thermogravimetric analysis. Bioresource Technology. 2017; 233:413-422. DOI: 10.1016/j. biortech.2017.02.119

[11] Tsamba AJ, Yang W, Blasiak W. Pyrolysis characteristics and global kinetics of coconut and cashew nut shells. Fuel Processing Technology. 2006;87:523-530. DOI: 10.1016/j. fuproc.2005.12.002

[12] Li LL, Fu XB, Wang XN, Tian YY, Qin S. Pyrolytic characteristics and kinetic studies of agricultural wastesfour kinds of grasses. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2016;38: 1156-1162. DOI: 10.1080/ 15567036.2013.856967

[13] Islam MA, Asif M, Hameed BH. Pyrolysis kinetics of raw and hydrothermally carbonized Karanj (Pongamia pinnata) fruit hulls via thermogravimetric analysis. Bioresource Technology. 2015;179:227-233. DOI: 10.1016/j.biortech.2014.11.115

[14] Subramanian S, Ragula UBR. Pyrolysis kinetics of Hibiscus rosa sinensis and Nerium oleander. Biofuels. 2018; 7269:1-15. DOI: 10.1080/17597269. 2018.1432274

Author details

Recent Advances in Pyrolysis

62

Vidyapeetham, Coimbatore, India

provided the original work is properly cited.

Udaya Bhaskar Reddy Ragula\*, Sriram Devanathan and Sindhu Subramanian Department of Chemical Engineering and Materials Science, Amrita Vishwa

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

\*Address all correspondence to: u\_bhaskarreddy@cb.amrita.edu

[15] Vinodhini S, Malathy NS. Bioprospecting of plants fibre of Coimbatore district of Tamil Nadu. International Journal of Plant Sciences. 2009;4:444-445

[16] Jabli M, Tka N, Ramzi K, Saleh TA. Physicochemical characteristics and dyeing properties of lignin-cellulosic fibers derived from Nerium oleander. Journal of Molecular Liquids. 2018;249: 1138-1144. DOI: 10.1016/j.molliq. 2017.11.126

[17] Suriapparao DV, Vinu R. Effects of biomass particle size on slow pyrolysis kinetics and fast pyrolysis product distribution. Waste and Biomass Valorization. 2017;9:1-13. DOI: 10.1007/ s12649-016-9815-7

[18] Gaston KR, Jarvis MW, Pepiot P, Smith KM, Frederick WJ, Nimlos MR. Biomass pyrolysis and gasification of varying particle sizes in a fluidized-bed reactor. Energy & Fuels. 2011;25: 3747-3757. DOI: 10.1021/ef200257k

[19] Niu Y, Tan H, Liu Y, Wang X, Xu T. The effect of particle size and heating rate on pyrolysis of waste capsicum stalks biomass. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2013;35: 1663-1669. DOI: 10.1080/ 15567036.2010.509084

[20] Tsai WT, Lee MK, Chang YM. Fast pyrolysis of rice straw, sugarcane bagasse and coconut shell in an induction-heating reactor. Journal of Analytical and Applied Pyrolysis. 2006; 76:230-237. DOI: 10.1016/j.jaap. 2005.11.007

[21] Lignite M. Effects of heating rate and particle size on pyrolysis kinetics of mengen lignite. Energy Sources. 2001;23:337-344. DOI: 10.1080/ 009083101300110887

[22] Basu P. Biomass Gasification and Pyrolysis: Practical Design and Theory. Kidlington, Oxford, UK: Academic Press; 2010. https://doi.org/10.1016/ C2009-0-20099-7

[30] Kirubakaran V, Sivaramakrishnan V, Nalini R, Sekar T, Premalatha M, Subramanian P. A review on gasification of biomass. Renewable and Sustainable Energy Reviews. 2009;13:179-186. DOI:

DOI: http://dx.doi.org/10.5772/intechopen.85581

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

[37] Ceylan S, Topçu Y. Pyrolysis kinetics of hazelnut husk using

10.1016/j.fuel.2006.07.005

10.1007/s11814-017-0308-6

1993;36:291-298

67:165-190

[39] Wagenaar BM, Prins W, Swaaij WPM v. Flash pyrolysis kinetics of pine wood. Fuel Processing Technology.

[40] Özsin G, Pütün AE. Co-pyrolytic behaviors of biomass and polystyrene: Kinetics, thermodynamics and evolved gas analysis. Korean Journal of Chemical Engineering. 2018;35:428-437. DOI:

[41] Gonzalez JF, Encinar JM, Canito JL, Sabio E, Chacon M. Pyrolysis of cherry stones: Energy uses of the different fractions and kinetic study. Journal of Analytical and Applied Pyrolysis. 2003;

[42] Fiori L, Valbusa M, Lorenzi D,

[43] Radmanesh R, Courbariaux Y, Chaouki J, Guy C. A unified lumped approach in kinetic modeling of biomass pyrolysis. Fuel. 2006;85:1211-1220. DOI:

pyrolysis of grape residues. Bioresource Technology. 2012;103:389-397. DOI: 10.1016/j.biortech.2011.09.113

[44] Cao L, Yuan X, Jiang L, Li C, Xiao Z, Huang Z, et al. Thermogravimetric characteristics and kinetics analysis of oil cake and torrefied biomass blends. Fuel. 2016;175:129-136. DOI: 10.1016/j.

[45] Tsai WT, Lee MK, Chang YM. Fast pyrolysis of rice husk: Product yields

Fambri L. Modeling of the devolatilization kinetics during

10.1016/j.fuel.2005.11.021

fuel.2016.01.089

thermogravimetric analysis. Bioresource Technology. 2014;156:182-188. DOI: 10.1016/j.biortech.2014.01.040

[38] Haykiri-Acma H, Yaman S. Synergy in devolatilization characteristics of lignite and hazelnut shell during copyrolysis. Fuel. 2007;86:373-380. DOI:

[31] Williams PT, Besler S. The pyrolysis of rice husks in a thermogravimetric analyser and static batch reactor. Fuel.

[32] El-Sayed SA, Mostafa ME. Pyrolysis characteristics and kinetic parameters determination of biomass fuel powders by differential thermal gravimetric analysis (TGA/DTG). Energy

Conversion and Management. 2014;85: 165-172. DOI: 10.1016/j.enconman.

[33] Mani T, Murugan P, Abedi J, Mahinpey N. Pyrolysis of wheat straw in a thermogravimetric analyzer: Effect of

particle size and heating rate on

kinetics. Chemical Engineering

devolatilization and estimation of global

Research and Design. 2010;88:952-958. DOI: 10.1016/j.cherd.2010.02.008

[34] Shen J, Igathinathane C, Yu M, Pothula AK. Biomass pyrolysis and combustion integral and differential reaction heats with temperatures using thermogravimetric analysis/differential scanning calorimetry. Bioresource Technology. 2015;185:89-98. DOI: 10.1016/j.biortech.2015.02.079

[35] Xiwen Y, Kaili X, Yu L. Assessing the effects of different process parameters on the pyrolysis Beha.: EBSCOhost. Bio Resources. 2017;12: 2748-2767. DOI: 10.1002/asi.22787

[36] Collazzo GC, Broetto CC, Perondi D, Junges J, Dettmer A, Dornelles Filho AA, et al. A detailed non-isothermal kinetic study of elephant grass pyrolysis from different models. Applied Thermal Engineering. 2017;110:1200-1211. DOI: 10.1016/j.applthermaleng.2016.09.012

65

10.1016/j.rser.2007.07.001

1993;72:151-159. DOI: 10.1016/

0016-2361(93)90391-E

2014.05.068

[23] Guerrero MRB, Salinas Gutiérrez JM, Meléndez Zaragoza MJ, López Ortiz A, Collins-Martínez V. Optimal slow pyrolysis of apple pomace reaction conditions for the generation of a feedstock gas for hydrogen production. International Journal of Hydrogen Energy. 2016;41:23232-23237. DOI: 10.1016/j.ijhydene.2016.10.066

[24] Pratap A, Chouhan S. A slow pyrolysis of cotton stalk (Gossypium arboretum) waste for bio-oil production. Journal of Pharmaceutical, Chemical and Biological Sciences. 2015;3:143-149

[25] Demirbas MF, Balat M. Biomass pyrolysis for liquid fuels and chemicals: A review. Journal of Scientific and Industrial Research. 2007;66:797-804

[26] Polat S, Apaydin-Varol E, Pütün AE. Thermal decomposition behavior of tobacco stem. Part II: Kinetic analysis. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2016;38:3073-3080. DOI: 10.1080/ 15567036.2015.1129374

[27] Raveendran K, Ganesh A, Khilar KC. Pyrolysis characteristics of biomass and biomass components. Fuel. 1996;75: 987-998

[28] Greenhalf CE, Nowakowski DJ, Harms AB, Titiloye JO, Bridgwater AV. A comparative study of straw, perennial grasses and hardwoods in terms of fast pyrolysis products. Fuel. 2013;108: 216-230. DOI: 10.1016/j.fuel.2013. 01.075

[29] Vyazovkin S, Burnham AK, Criado JM, Pérez-Maqueda LA, Popescu C, Sbirrazzuoli N. ICTAC kinetics committee recommendations for performing kinetic computations on thermal analysis data. Thermochimica Acta. 2011;520:1-19. DOI: 10.1016/j. tca.2011.03.034

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

[30] Kirubakaran V, Sivaramakrishnan V, Nalini R, Sekar T, Premalatha M, Subramanian P. A review on gasification of biomass. Renewable and Sustainable Energy Reviews. 2009;13:179-186. DOI: 10.1016/j.rser.2007.07.001

[15] Vinodhini S, Malathy NS. Bioprospecting of plants fibre of Coimbatore district of Tamil Nadu. International Journal of Plant Sciences.

Recent Advances in Pyrolysis

[16] Jabli M, Tka N, Ramzi K, Saleh TA. Physicochemical characteristics and dyeing properties of lignin-cellulosic fibers derived from Nerium oleander. Journal of Molecular Liquids. 2018;249: 1138-1144. DOI: 10.1016/j.molliq.

Kidlington, Oxford, UK: Academic Press; 2010. https://doi.org/10.1016/

[23] Guerrero MRB, Salinas Gutiérrez JM, Meléndez Zaragoza MJ, López Ortiz A, Collins-Martínez V. Optimal slow pyrolysis of apple pomace reaction conditions for the generation of a feedstock gas for hydrogen production. International Journal of Hydrogen Energy. 2016;41:23232-23237. DOI: 10.1016/j.ijhydene.2016.10.066

[24] Pratap A, Chouhan S. A slow pyrolysis of cotton stalk (Gossypium arboretum) waste for bio-oil production. Journal of Pharmaceutical, Chemical and Biological Sciences. 2015;3:143-149

[25] Demirbas MF, Balat M. Biomass pyrolysis for liquid fuels and chemicals: A review. Journal of Scientific and Industrial Research. 2007;66:797-804

[26] Polat S, Apaydin-Varol E, Pütün AE. Thermal decomposition behavior of tobacco stem. Part II: Kinetic analysis. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2016;38:3073-3080. DOI: 10.1080/

[27] Raveendran K, Ganesh A, Khilar KC. Pyrolysis characteristics of biomass and biomass components. Fuel. 1996;75:

[28] Greenhalf CE, Nowakowski DJ, Harms AB, Titiloye JO, Bridgwater AV. A comparative study of straw, perennial grasses and hardwoods in terms of fast pyrolysis products. Fuel. 2013;108: 216-230. DOI: 10.1016/j.fuel.2013.

[29] Vyazovkin S, Burnham AK, Criado JM, Pérez-Maqueda LA, Popescu C, Sbirrazzuoli N. ICTAC kinetics committee recommendations for performing kinetic computations on thermal analysis data. Thermochimica Acta. 2011;520:1-19. DOI: 10.1016/j.

15567036.2015.1129374

987-998

01.075

tca.2011.03.034

C2009-0-20099-7

[17] Suriapparao DV, Vinu R. Effects of biomass particle size on slow pyrolysis kinetics and fast pyrolysis product distribution. Waste and Biomass

Valorization. 2017;9:1-13. DOI: 10.1007/

[18] Gaston KR, Jarvis MW, Pepiot P, Smith KM, Frederick WJ, Nimlos MR. Biomass pyrolysis and gasification of varying particle sizes in a fluidized-bed reactor. Energy & Fuels. 2011;25: 3747-3757. DOI: 10.1021/ef200257k

[19] Niu Y, Tan H, Liu Y, Wang X, Xu T. The effect of particle size and heating rate on pyrolysis of waste capsicum stalks biomass. Energy Sources, Part A:

[20] Tsai WT, Lee MK, Chang YM. Fast pyrolysis of rice straw, sugarcane bagasse and coconut shell in an induction-heating reactor. Journal of Analytical and Applied Pyrolysis. 2006;

76:230-237. DOI: 10.1016/j.jaap.

[21] Lignite M. Effects of heating rate and particle size on pyrolysis kinetics of mengen lignite. Energy Sources. 2001;23:337-344. DOI: 10.1080/

[22] Basu P. Biomass Gasification and Pyrolysis: Practical Design and Theory.

2005.11.007

64

009083101300110887

Recovery, Utilization, and Environmental Effects. 2013;35: 1663-1669. DOI: 10.1080/ 15567036.2010.509084

2009;4:444-445

2017.11.126

s12649-016-9815-7

[31] Williams PT, Besler S. The pyrolysis of rice husks in a thermogravimetric analyser and static batch reactor. Fuel. 1993;72:151-159. DOI: 10.1016/ 0016-2361(93)90391-E

[32] El-Sayed SA, Mostafa ME. Pyrolysis characteristics and kinetic parameters determination of biomass fuel powders by differential thermal gravimetric analysis (TGA/DTG). Energy Conversion and Management. 2014;85: 165-172. DOI: 10.1016/j.enconman. 2014.05.068

[33] Mani T, Murugan P, Abedi J, Mahinpey N. Pyrolysis of wheat straw in a thermogravimetric analyzer: Effect of particle size and heating rate on devolatilization and estimation of global kinetics. Chemical Engineering Research and Design. 2010;88:952-958. DOI: 10.1016/j.cherd.2010.02.008

[34] Shen J, Igathinathane C, Yu M, Pothula AK. Biomass pyrolysis and combustion integral and differential reaction heats with temperatures using thermogravimetric analysis/differential scanning calorimetry. Bioresource Technology. 2015;185:89-98. DOI: 10.1016/j.biortech.2015.02.079

[35] Xiwen Y, Kaili X, Yu L. Assessing the effects of different process parameters on the pyrolysis Beha.: EBSCOhost. Bio Resources. 2017;12: 2748-2767. DOI: 10.1002/asi.22787

[36] Collazzo GC, Broetto CC, Perondi D, Junges J, Dettmer A, Dornelles Filho AA, et al. A detailed non-isothermal kinetic study of elephant grass pyrolysis from different models. Applied Thermal Engineering. 2017;110:1200-1211. DOI: 10.1016/j.applthermaleng.2016.09.012

[37] Ceylan S, Topçu Y. Pyrolysis kinetics of hazelnut husk using thermogravimetric analysis. Bioresource Technology. 2014;156:182-188. DOI: 10.1016/j.biortech.2014.01.040

[38] Haykiri-Acma H, Yaman S. Synergy in devolatilization characteristics of lignite and hazelnut shell during copyrolysis. Fuel. 2007;86:373-380. DOI: 10.1016/j.fuel.2006.07.005

[39] Wagenaar BM, Prins W, Swaaij WPM v. Flash pyrolysis kinetics of pine wood. Fuel Processing Technology. 1993;36:291-298

[40] Özsin G, Pütün AE. Co-pyrolytic behaviors of biomass and polystyrene: Kinetics, thermodynamics and evolved gas analysis. Korean Journal of Chemical Engineering. 2018;35:428-437. DOI: 10.1007/s11814-017-0308-6

[41] Gonzalez JF, Encinar JM, Canito JL, Sabio E, Chacon M. Pyrolysis of cherry stones: Energy uses of the different fractions and kinetic study. Journal of Analytical and Applied Pyrolysis. 2003; 67:165-190

[42] Fiori L, Valbusa M, Lorenzi D, Fambri L. Modeling of the devolatilization kinetics during pyrolysis of grape residues. Bioresource Technology. 2012;103:389-397. DOI: 10.1016/j.biortech.2011.09.113

[43] Radmanesh R, Courbariaux Y, Chaouki J, Guy C. A unified lumped approach in kinetic modeling of biomass pyrolysis. Fuel. 2006;85:1211-1220. DOI: 10.1016/j.fuel.2005.11.021

[44] Cao L, Yuan X, Jiang L, Li C, Xiao Z, Huang Z, et al. Thermogravimetric characteristics and kinetics analysis of oil cake and torrefied biomass blends. Fuel. 2016;175:129-136. DOI: 10.1016/j. fuel.2016.01.089

[45] Tsai WT, Lee MK, Chang YM. Fast pyrolysis of rice husk: Product yields

and compositions. Bioresource Technology. 2007;98:22-28. DOI: 10.1016/j.biortech.2005.12.005

[46] Vieira FR, Luna CMR, Arce GLAF, Avila I. A study of biochar yield from slow pyrolysis of rice husk; In: 24th ABCM-Brazilian Association of Engineering and Mechanical Sciences; 2017. p. 1-6

[47] Ali N, Saleem M, Shahzad K, Chughtai A. Bio-oil production from fast pyrolysis of cotton stalk in fluidized bed reactor. Arabian Journal for Science and Engineering. 2015;40:3019-3027. DOI: 10.1007/s13369-015-1801-z

[48] Sidhu GK, Sandhya. Engineering properties of cotton stalks (Gossypium hirsitum L.), Indian. Journal of Agricultural Research. 2015;49:456-459. DOI: 10.18805/ijare.v49i5.5811

[49] Park J, Lee Y, Ryu C, Park Y. Slow pyrolysis of rice straw: Analysis of products properties, carbon and energy yields. Bioresource Technology. 2014; 155:63-70. DOI: 10.1016/j.biortech. 2013.12.084

[50] Song H, Andreas J, Minhou X. Kinetic study of Chinee biomass slow pyrolysis: Comparison of different kinetic models. Fuel. 2007;86:2778-2788. DOI: 10.1016/j.fuel.2007.02.031

[51] Mohanty P, Nanda S, Pant KK, Naik S, Kozinski JA, Dalai AK. Evaluation of the physiochemical development of biochars obtained from pyrolysis of wheat straw, timothy grass and pinewood: Effects of heating rate. Journal of Analytical and Applied Pyrolysis. 2013;104:485-493. DOI: 10.1016/j.jaap.2013.05.022

[52] Fu P, Li Z, Bai X, Altilia WY, Pietrangeli F, Torre AD, et al. Bio-oil production from fast pyrolysis of corn stalk in a fluidized bed. Applied Mechanics and Materials. 2014;

672–674:143-146. DOI: 10.4028/www. scientific.net/AMM.672-674.143

hazelnut shell: A study on mass transfer limitations on product yields and characterization of the pyrolysis oil. Energy Sources. 2000;22:913-924. DOI:

DOI: http://dx.doi.org/10.5772/intechopen.85581

Modeling and Optimization of Product Profiles in Biomass Pyrolysis

reactor. Journal of Analytical and Applied Pyrolysis. 2009;84:179-184. DOI: 10.1016/j.jaap.2009.02.001

[67] Sundaram EG, Natarajan E. Pyrolysis of coconut shell: An

10.24200/tjer.vol6iss2pp33-39

Subramanian P, Uma D.

2016.1261205

[68] Mythili R, Venkatachalam P,

Characterization of bioresidues for biooil production through pyrolysis. Bioresource Technology. 2013;138:71-78. DOI: 10.1016/j.biortech.2013.03.161

[69] Mythili R, Subramanian P, Uma D. Biofuel production from Prosopis juliflora in fluidized bed reactor. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2017;39: 741-746. DOI: 10.1080/15567036.

[70] Chandrasekaran A, Ramachandran S, Subbiah S. Modeling, experimental validation and optimization of Prosopis juliflora fuelwood pyrolysis in fixed-bed

tubular reactor. Bioresource Technology. 2018;264:66-77. DOI: 10.1016/j.biortech.2018.05.013

[71] Xu J, Qiao L. Mathematical

DOI: 10.1021/ef3008745

Pana-Suppamassadu K,

modeling of coal gasification processes in a well-stirred reactor: Effects of devolatilization and moisture content. Energy & Fuels. 2012;26:5759-5768.

[72] Narataruksa P, Tungkamani S,

Hunpinyo P, et al. Conversion enhancement of tubular fixed-bed reactor for Fischer-Tropsch synthesis using static mixer. Journal of Natural Gas Chemistry. 2012;21:435-444. DOI: 10.1016/S1003-9953(11)60388-5

Keeratiwintakorn P, Nivitchanyong S,

[73] Wu J, Zhang H, Ying W, Fang D. Simulation and analysis of a tubular fixed-bed Fischer-Tropsch synthesis

experimental investigation. Journal of Engineering Research. 2009;6:33. DOI:

10.1080/00908310051128291

[61] Chen T, Liu R, Scott NR. Characterization of energy carriers obtained from the pyrolysis of white ash, switchgrass and corn stoverbiochar, syngas and bio-oil. Fuel Processing Technology. 2016;142: 124-134. DOI: 10.1016/j.fuproc.

[62] Onay Ö, Beis SH, Koçkar ÖM. Pyrolysis of walnut shell in a well-swept fixed-bed reactor. Energy Sources. 2004;26:771-782. DOI: 10.1080/

[63] Yuan HR, Liu RH. Study on pyrolysis kinetics of walnut shell. Journal of Thermal Analysis and Calorimetry. 2007;89:983-986. DOI:

10.1504/IJGEI.2008.018006

[64] Suttibak S, Sriprateep K, Pattiya A. Production of bio-oil from pine sawdust by rapid pyrolysis in a fluidized-bed reactor. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2015;37: 1440-1446. DOI: 10.1080/ 15567036.2011.631091

[65] Uzun BB, Kanmaz G. Effect of operating parameters on bio-fuel production from waste furniture sawdust. Waste Management & Research. 2013;31:361-367. DOI: 10.1177/0734242X12470402

[66] Wang Z, Cao J, Wang J. Pyrolytic characteristics of pine wood in a slowly heating and gas sweeping fixed-bed

67

j.jaap.2010.10.001

2015.09.034

00908310490451402

[60] Carrier M, Hugo T, Gorgens J, Knoetze H. Comparison of slow and vacuum pyrolysis of sugar cane bagasse. Journal of Analytical and Applied Pyrolysis. 2011;90:18-26. DOI: 10.1016/

[53] Cordella M, Berrueco C, Santarelli F, Paterson N, Kandiyoti R, Millan M. Yields and ageing of the liquids obtained by slow pyrolysis of sorghum, switchgrass and corn stalks. Journal of Analytical and Applied Pyrolysis. 2013; 104:316-324. DOI: 10.1016/j. jaap.2013.07.001

[54] Zhang H, Xiao R, Wang D, He G, Shao S, Zhang J, et al. Biomass fast pyrolysis in a fluidized bed reactor under N2, CO2, CO, CH4 and H2 atmospheres. Bioresource Technology. 2011;102:4258-4264. DOI: 10.1016/j. biortech.2010.12.075

[55] Demirbas A. Effects of temperature and particle size on bio-char yield from pyrolysis of agricultural residues. Journal of Analytical and Applied Pyrolysis. 2004;72:243-248. DOI: 10.1016/j.jaap.2004.07.003

[56] Sousa J, Bezerra M, Almeida M, Moure G, Mesa-Perez J, Caramao E. Characteristics of bio-oil from the fast pyrolysis of elephant grass (Pennisetum purpureum schumach) in a fluidized bed reactor. American Chemical Science Journal. 2016;14:1-10. DOI: 10.9734/ ACSJ/2016/25843

[57] Boateng AA, Mullen CA, Goldberg N, Hicks KB, Jung HJG, Lamb JFS. Production of bio-oil from alfalfa stems by fluidized-bed fast pyrolysis. Industrial and Engineering Chemistry Research. 2008;47:4115-4122. DOI: 10.1021/ie800096g

[58] Strezov V, Evans TJ, Hayman C. Conversion of elephant grass (Pennisetum purpureum schum) to biogas, bio-oil and charcoal. Bioresource Technology. 2008;99:8394-8399. DOI: 10.1016/j.biortech.2008.02.039

[59] Koçkar ÖM, Onay Ö, Pütün AE, Pütün E. Fixed-bed pyrolysis of

Modeling and Optimization of Product Profiles in Biomass Pyrolysis DOI: http://dx.doi.org/10.5772/intechopen.85581

hazelnut shell: A study on mass transfer limitations on product yields and characterization of the pyrolysis oil. Energy Sources. 2000;22:913-924. DOI: 10.1080/00908310051128291

and compositions. Bioresource Technology. 2007;98:22-28. DOI: 10.1016/j.biortech.2005.12.005

Recent Advances in Pyrolysis

2017. p. 1-6

2013.12.084

[46] Vieira FR, Luna CMR, Arce GLAF, Avila I. A study of biochar yield from slow pyrolysis of rice husk; In: 24th ABCM-Brazilian Association of Engineering and Mechanical Sciences;

672–674:143-146. DOI: 10.4028/www. scientific.net/AMM.672-674.143

[53] Cordella M, Berrueco C, Santarelli F, Paterson N, Kandiyoti R, Millan M. Yields and ageing of the liquids obtained

switchgrass and corn stalks. Journal of Analytical and Applied Pyrolysis. 2013;

[54] Zhang H, Xiao R, Wang D, He G, Shao S, Zhang J, et al. Biomass fast pyrolysis in a fluidized bed reactor under N2, CO2, CO, CH4 and H2 atmospheres. Bioresource Technology. 2011;102:4258-4264. DOI: 10.1016/j.

[55] Demirbas A. Effects of temperature and particle size on bio-char yield from pyrolysis of agricultural residues. Journal of Analytical and Applied Pyrolysis. 2004;72:243-248. DOI: 10.1016/j.jaap.2004.07.003

[56] Sousa J, Bezerra M, Almeida M, Moure G, Mesa-Perez J, Caramao E. Characteristics of bio-oil from the fast pyrolysis of elephant grass (Pennisetum purpureum schumach) in a fluidized bed reactor. American Chemical Science Journal. 2016;14:1-10. DOI: 10.9734/

[57] Boateng AA, Mullen CA, Goldberg N, Hicks KB, Jung HJG, Lamb JFS. Production of bio-oil from alfalfa stems

[58] Strezov V, Evans TJ, Hayman C.

[59] Koçkar ÖM, Onay Ö, Pütün AE, Pütün E. Fixed-bed pyrolysis of

(Pennisetum purpureum schum) to biogas, bio-oil and charcoal. Bioresource Technology. 2008;99:8394-8399. DOI: 10.1016/j.biortech.2008.02.039

Conversion of elephant grass

by fluidized-bed fast pyrolysis. Industrial and Engineering Chemistry Research. 2008;47:4115-4122. DOI:

by slow pyrolysis of sorghum,

104:316-324. DOI: 10.1016/j.

jaap.2013.07.001

biortech.2010.12.075

ACSJ/2016/25843

10.1021/ie800096g

[47] Ali N, Saleem M, Shahzad K, Chughtai A. Bio-oil production from fast pyrolysis of cotton stalk in fluidized bed reactor. Arabian Journal for Science and Engineering. 2015;40:3019-3027. DOI: 10.1007/s13369-015-1801-z

[48] Sidhu GK, Sandhya. Engineering properties of cotton stalks (Gossypium hirsitum L.), Indian. Journal of

Agricultural Research. 2015;49:456-459.

[49] Park J, Lee Y, Ryu C, Park Y. Slow pyrolysis of rice straw: Analysis of products properties, carbon and energy yields. Bioresource Technology. 2014; 155:63-70. DOI: 10.1016/j.biortech.

[50] Song H, Andreas J, Minhou X. Kinetic study of Chinee biomass slow pyrolysis: Comparison of different kinetic models. Fuel. 2007;86:2778-2788.

DOI: 10.1016/j.fuel.2007.02.031

10.1016/j.jaap.2013.05.022

66

[52] Fu P, Li Z, Bai X, Altilia WY, Pietrangeli F, Torre AD, et al. Bio-oil production from fast pyrolysis of corn stalk in a fluidized bed. Applied Mechanics and Materials. 2014;

[51] Mohanty P, Nanda S, Pant KK, Naik S, Kozinski JA, Dalai AK. Evaluation of the physiochemical development of biochars obtained from pyrolysis of wheat straw, timothy grass and pinewood: Effects of heating rate. Journal of Analytical and Applied Pyrolysis. 2013;104:485-493. DOI:

DOI: 10.18805/ijare.v49i5.5811

[60] Carrier M, Hugo T, Gorgens J, Knoetze H. Comparison of slow and vacuum pyrolysis of sugar cane bagasse. Journal of Analytical and Applied Pyrolysis. 2011;90:18-26. DOI: 10.1016/ j.jaap.2010.10.001

[61] Chen T, Liu R, Scott NR. Characterization of energy carriers obtained from the pyrolysis of white ash, switchgrass and corn stoverbiochar, syngas and bio-oil. Fuel Processing Technology. 2016;142: 124-134. DOI: 10.1016/j.fuproc. 2015.09.034

[62] Onay Ö, Beis SH, Koçkar ÖM. Pyrolysis of walnut shell in a well-swept fixed-bed reactor. Energy Sources. 2004;26:771-782. DOI: 10.1080/ 00908310490451402

[63] Yuan HR, Liu RH. Study on pyrolysis kinetics of walnut shell. Journal of Thermal Analysis and Calorimetry. 2007;89:983-986. DOI: 10.1504/IJGEI.2008.018006

[64] Suttibak S, Sriprateep K, Pattiya A. Production of bio-oil from pine sawdust by rapid pyrolysis in a fluidized-bed reactor. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2015;37: 1440-1446. DOI: 10.1080/ 15567036.2011.631091

[65] Uzun BB, Kanmaz G. Effect of operating parameters on bio-fuel production from waste furniture sawdust. Waste Management & Research. 2013;31:361-367. DOI: 10.1177/0734242X12470402

[66] Wang Z, Cao J, Wang J. Pyrolytic characteristics of pine wood in a slowly heating and gas sweeping fixed-bed

reactor. Journal of Analytical and Applied Pyrolysis. 2009;84:179-184. DOI: 10.1016/j.jaap.2009.02.001

[67] Sundaram EG, Natarajan E. Pyrolysis of coconut shell: An experimental investigation. Journal of Engineering Research. 2009;6:33. DOI: 10.24200/tjer.vol6iss2pp33-39

[68] Mythili R, Venkatachalam P, Subramanian P, Uma D. Characterization of bioresidues for biooil production through pyrolysis. Bioresource Technology. 2013;138:71-78. DOI: 10.1016/j.biortech.2013.03.161

[69] Mythili R, Subramanian P, Uma D. Biofuel production from Prosopis juliflora in fluidized bed reactor. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 2017;39: 741-746. DOI: 10.1080/15567036. 2016.1261205

[70] Chandrasekaran A, Ramachandran S, Subbiah S. Modeling, experimental validation and optimization of Prosopis juliflora fuelwood pyrolysis in fixed-bed tubular reactor. Bioresource Technology. 2018;264:66-77. DOI: 10.1016/j.biortech.2018.05.013

[71] Xu J, Qiao L. Mathematical modeling of coal gasification processes in a well-stirred reactor: Effects of devolatilization and moisture content. Energy & Fuels. 2012;26:5759-5768. DOI: 10.1021/ef3008745

[72] Narataruksa P, Tungkamani S, Pana-Suppamassadu K, Keeratiwintakorn P, Nivitchanyong S, Hunpinyo P, et al. Conversion enhancement of tubular fixed-bed reactor for Fischer-Tropsch synthesis using static mixer. Journal of Natural Gas Chemistry. 2012;21:435-444. DOI: 10.1016/S1003-9953(11)60388-5

[73] Wu J, Zhang H, Ying W, Fang D. Simulation and analysis of a tubular fixed-bed Fischer-Tropsch synthesis

reactor with co-based catalyst. Chemical Engineering and Technology. 2010;33: 1083-1092. DOI: 10.1002/ceat. 200900610

[74] Kunii D, Levenspiel O. Fluidization Engineering. 2nd ed. Boston: Elsevier; 1991

[75] Ciesielczyk W, Iwanowski J. Analysis of fluidized bed drying kinetics on the basis of interphase mass transfer coefficient. Journal of Drying Technology. 2006;24:1153-1157

[76] Agarwal PK, Genetti WE, Lee YY. Drying and devolatilization of Mississippi lignite in a fluidized-bed. American Chemical Journal. 1984:29

**69**

**Chapter 5**

**Abstract**

**1. Introduction**

Production

Waste Plastics Valorization by Fast

Pyrolysis and in Line Catalytic

Steam Reforming for Hydrogen

*Itsaso Barbarias, Aitor Arregi, Maite Artetxe,* 

*Laura Santamaria, Gartzen Lopez, María Cortazar,* 

This chapter summarizes the most recent results obtained in the plastic waste pyrolysis-reforming strategy for hydrogen production. An original two-reactor configuration consisting of a conical spouted bed reactor for the pyrolysis step and a fluidized bed reactor for the pyrolysis volatile reforming is proposed. The fundamental aspects and challenges of this joint process are discussed in detail, and the prospects for the full-scale implementation of this valorization route are assessed. Thus, the influence the main reforming parameters (temperature, space time and steam/ plastic ratio) have in the pyrolysis-reforming of HDPE on product yields and catalyst stability are reported. Moreover, the role played by plastic composition on process performance is also described by studying the influence of following polymers: high density polyethylene (HDPE), polypropylene (PP), polyethylene terephthalate (PET) and polystyrene (PS). The operating conditions used for the valorization of different plastics have been as follows: pyrolysis temperature of 500°C, reforming temperature

of 700°C, space time of 16.7 gcatalyst min gplastic−1 and steam/plastic ratio of 4.

The current overall hydrogen production has reached a value of 7.7 EJ/year, and an increase to 10 EJ/year is expected by the year 2050 [1]. Hydrogen demand is mainly related to ammonia production (51%), oil refining (31%), methanol production (10%), and other uses (8%). Moreover, new fossil sources of lower quality than conventional oil are increasingly been used, but they require severe hydroprocessing for their upgrading. This is the case of oil sands, which surely will contribute to

Approximately 96% of the world hydrogen produced is of fossil origin. Thus, 48% is produced from natural gas reforming, 30% from oil fractions and the remaining 18% by means of coal gasification [2]. Accordingly, only 4% is produced

**Keywords:** hydrogen, pyrolysis, reforming, waste plastics, spouted bed

meeting the increasing requirements of hydrogen.

from renewable sources, such as water electrolysis.

*Maider Amutio, Javier Bilbao and Martin Olazar*

## **Chapter 5**

reactor with co-based catalyst. Chemical Engineering and Technology. 2010;33:

[74] Kunii D, Levenspiel O. Fluidization Engineering. 2nd ed. Boston: Elsevier;

Analysis of fluidized bed drying kinetics on the basis of interphase mass transfer

[76] Agarwal PK, Genetti WE, Lee YY.

1083-1092. DOI: 10.1002/ceat.

Recent Advances in Pyrolysis

[75] Ciesielczyk W, Iwanowski J.

coefficient. Journal of Drying Technology. 2006;24:1153-1157

Drying and devolatilization of Mississippi lignite in a fluidized-bed. American Chemical Journal. 1984:29

200900610

1991

68
