Renewable and Sustainable Energy Reviews 50 (2015) 10811096

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Biomass pyrolysisA review of modelling, process parameters

and catalytic studies
Abhishek Sharma a, Vishnu Pareek a,n, Dongke Zhang b
a
Department of Chemical Engineering, Curtin University, Kent Street, Bentley, Perth 6102, WA, Australia
b
Centre for Energy (M473), The University of Western Australia, 35 Stirling Highway, Crawley 6009, WA, Australia

art ic l e i nf o a b s t r a c t

Article history: Biomass as a form of energy source may be utilized in two different ways: directly by burning the
Received 5 August 2014 biomass and indirectly by converting it into solid, liquid or gaseous fuels. Pyrolysis is an indirect
Received in revised form conversion method, and can be described in simpler terms as a thermal decomposition of biomass under
13 April 2015
oxygen-depleted conditions to an array of solid, liquid and gaseous products, namely biochar, bio-oil and
Accepted 30 April 2015
fuel gas. However, pyrolysis of biomass is a complex chemical process with several operational and
Available online 6 June 2015
environmental challenges. Consequently, this process has been widely investigated in order to under-
Keywords: stand the mechanisms and kinetics of pyrolysis at different scales, viz. particle level, multi-phase
Biochar reacting ow, product distribution and reactor performance, process integration and control. However,
Biomass
there are a number of uncertainties in current biomass pyrolysis models, especially in their ability to
Hydrodynamic
optimize process conditions to achieve desired product yields and distribution.
Kinetics
Modelling The present contribution provides a critical review of the current status of mathematical modelling
Pyrolysis studies of biomass pyrolysis with the aim to identify knowledge gaps for further research and
opportunities for integration of biomass pyrolysis models of disparate scales. Models for the hydro-
dynamic behaviour of particles in pyrolysis, and their interaction with the reactive ow and the effect on
the performance of the reactors have also been critically analyzed. From this analysis it becomes
apparent that feedstock characteristics, evolving physical and chemical properties of biomass particles
and residence times of both solid and gas phases in reactors hold the key to the desired performance of
the pyrolysis process. Finally, the importance of catalytic effects in pyrolysis has also been critically
analyzed, resulting in recommendations for further research in this area especially on selection of
catalysts for optimal product yields under varying operating conditions.
& 2015 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1082
2. Biomass characteristics and its implications in pyrolysis operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1083
2.1. Biomass composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1083
2.2. Thermo-chemical degradation of biomass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1083
2.3. Biomass degradation products and uses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1084
3. Modelling of pyrolysis processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1084
3.1. Kinetic models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1084
3.1.1. Lumped models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1084
3.1.2. Distributed models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1086
3.2. Particle models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1087
3.3. Reactor models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1088
3.3.1. Phenomenological models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1088
3.3.2. CFD models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1089
4. Pyrolysis process parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1090

n
Corresponding author. Tel.: 61 8 9266 4687.
E-mail address: v.pareek@curtin.edu.au (V. Pareek).

http://dx.doi.org/10.1016/j.rser.2015.04.193
1364-0321/& 2015 Elsevier Ltd. All rights reserved.
1082 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096

4.1. Type of feedstock. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1090

4.2. Heating rate and temperature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1090
4.3. Volatiles residence time and pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1090
4.4. Particle size, shape and orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1091
4.5. Reactor conguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1091
4.6. Physico-chemical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1091
4.6.1. Thermal conductivity and emissivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1091
4.6.2. Permeability and density. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1091
4.6.3. Specic heat capacity and heat of reaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1092
4.6.4. Particle shrinkage and moisture content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1092
4.6.5. External heat transfer coefcient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1092
5. Catalytic pyrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1092
5.1. Process features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1092
5.2. Biochar catalytic effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1092
6. Summary and future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1093
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1094
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1094

1. Introduction
There is a worldwide drive for reducing reliance on fossil fuels,
the burning of which contributes signicant amount of carbon
emissions leading to global warming [1]. Also, there is a shift from
non-renewable energy sources to bio-energy due to continuous
depletion of fossil fuels. Furthermore, since biofuels are derived
from biomass, they signicantly decrease emissions of harmful
gases such as SOx and NOx [2]. Examples of commonly used
biomass include plant matter such as forest residues (dead trees,
branches and tree stumps), yard clippings, wood chips and
municipal solid waste. The benet of using biomass arises due to
its renewable nature which comes from the ability to utilize the
emitted CO2 for growing next generation of biomass through
photosynthesis cycle. Consequently, in recent years, the use of
biomass-derived fuels has been steadily increasing, and they
currently contribute around 13% of the worlds energy supply [3].
Biomass may be converted to biofuels using a number of
physical, thermo-chemical and bio-chemical processes as shown
in Fig. 1. The physical processes use densication techniques
including crushing, heat and pressure applications for biomass
conversion into biofuels. The bio-chemical processes use enzymes
and micro-organisms to convert biomass into desirable energy
products. On the other hand, the thermo-chemical processes use
heat energy and chemical catalysts for decomposition of biomass
into high value energy products.
In the direct combustion of biomass, the primary product is the
thermal energy, which may be used as the energy source for the
production of electricity, or for the combined heat and power (CHP)
production plants. The biomass gasication process is used for
production of bio-fuels such as green gasoline and also electricity.
Pyrolysis process is used for production of bio-fuels, chemicals and
charcoal with electricity and CHP generation using turbines, engines
and boilers. The hydrothermal liquefaction process may be used for
direct conversion of biomass into crude oil for generation of heat
and power or, after upgrading using hydrotreating or hydroproces-
sing, as premium liquid transport fuels.
As shown in Fig. 1, pyrolysis is a thermo-chemical decomposi-
tion process in which organic material such as biomass is con-
verted into a carbon-rich solid and a volatile matter by heating in
the absence of oxygen [4]. The solid product of this process is
known as the biochar or char, and is generally high in carbon
content. The volatile fraction of this process is partly condensed to
a liquid fraction called tar or bio-oil along with a mixture of the
non-condensable gases. The bio-oil is stored and further used for
energy production. The gases can be utilized for providing heat
energy to the pyrolysis reactor. The pyrolysis products are formed
from both primary decomposition of the solid biomass as well as

Crushing/
Solid Fuel
Grinding

Physical Drying Solid Fuel

Pelletization/Briquetting Solid Fuel

Combustion
Heat energy
( in presence of excess air)

Gasification Syngas and

(in presence of partial air and steam) Fuel gas
Biomass Thermo-
Pyrolysis Bio-oil, Bio-
chemical char and Fuel
(in inert atmosphere) gas

Hydrothermal Liquefaction Bio-oil

(in presence of high pressure water)

Hydrolysis-
Bioethanol
Bio- Fermentation
chemical
Anaerobic Digestion Biogas

Fig. 1. Biomass to bioenergy conversion pathways.

 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
Fig. 2. Schematic diagram of chemical and physical processes inside biomass particle during pyrolysis [153].
secondary reactions of volatile condensable organic products into
low-molecular weight gases, secondary tar and char [5]. One of the
signicant benets of pyrolysis is that it can be conducted at
temperatures lower (normally in range of 673973 K) than those
required in gasication (4973 K) [6] and combustion ( 41173 K)
[7] processes. It has been found that the presence of metallic
compounds in biomass causes further reduction in decomposition
temperature during pyrolysis process [8]. The pressure require-
ment is also much lower in pyrolysis process (0.10.2 MPa) as
compared to hydrothermal liquefaction (1025 MPa) [9] of bio-
mass to generate bio-oil.
The operation of a pyrolysis process depends on several
parameters such as the feedstock and reactor conditions that lead
to formation of products in different proportions as shown in
Fig. 2. Most of the initial research in this area can be attributed to
pyrolysis of coal [1015]. Although biomass has composition
similar to coal, it varies in content of constituents such as
holocellulose and lignin, which signicantly affects the degrada-
tion mechanism. Furthermore, there are differences in density and
structure (like brosity) of coal and biomass, which mandates
different operating conditions such as solid residence time, tem-
perature, etc. Despite these challenges, the potential of biomass for
generating a range of useful products using pyrolysis has pro-
moted signicant research in this area. This review discusses the
kinetics, particle and reactor scale models for studying biomass
pyrolysis, and other relevant aspects with respect to their practical
applications. This study also focuses on the effect of important
parameters such as temperature, particle size, density, etc. on the
biomass degradation mechanisms. Finally, it gives an overview of
catalysts for primary and secondary cracking reactions, as well as
the catalytic effect of bio-char. The process integration models of
biomass pyrolysis for biofuel production have been discussed
elsewhere [16].
2. Biomass characteristics and its implications in pyrolysis
operations
Biomass is any mixture of hydrocarbon material consisting of
carbon, hydrogen, oxygen with small contents of sulphur and
nitrogen. It includes various natural and derived materials, such as
woody and herbaceous species, wood wastes, bagasse, sawdust,
agricultural residues, bio-solids, grass, aquatic plants, etc. [17].
1083
2.1. Biomass composition
Biomass is generally composed of three main groups of natural
polymeric materials: cellulose (around 50% on dry basis), hemi-
cellulose (1030% in woods and 2040% in herbaceous biomass on
dry basis) and lignin (2040% in woods and 1040% in herbaceous
biomass on dry basis). Other typical components are grouped as
extractives (generally smaller organic molecules or polymers like
protein, acids, salts) and minerals [17] (inorganic compounds like
alkali metals mainly potassium, calcium, sodium, silicon, phos-
phorus and magnesium and also chlorine in herbaceous biomass
[18]). The contents of these inorganic compounds vary from being
less than 1% in woods to 15% in herbaceous biomass and feedstock
and up to 25% in agricultural and forestry residues [19].
2.2. Thermo-chemical degradation of biomass
The thermo-chemical decomposition of biomass can be carried
out in number of ways such as pyrolysis, gasication and combus-
tion, by varying the operating conditions such as vapour residence
time, heating rate and reactor conguration. Pyrolysis may be used
to produce a range of products in the form of solid, liquid and
gaseous fuels. The solid product of biomass pyrolysis the biochar
has also found applications in carbon sequestration via soil
management systems [20]. On the other hand, hydrothermal
liquefaction of biomass focuses on generation of liquid fuel only.
Finally, gasication mainly focuses on the production of fuel gas
and other gaseous products such as syngas, major application of
which is in the FischerTropsch synthesis to produce synthetic
liquid fuels, while combustion leads to the production of heat and
electric power. Hence, it is clear that the pyrolysis process offers
much broader opportunities for converting biomass into useful
chemicals than any of the other thermal conversion processes.
The pyrolysis process for biomass degradation may be broadly
classied into two major categories depending on the operating
conditions. The fast pyrolysis is characterized by high heating
rates and short vapour residence times. This generally requires a
feedstock consisting of small particles along with a provision to
remove the vapours quickly to avoid further contact with the hot
solid particles. There are a number of different reactor congura-
tions for fast pyrolysis including ablative systems, uidized beds,
stirred or moving beds and vacuum pyrolysis systems [21]. Fig. 3
shows a fast pyrolysis uidized bed reactor. It normally operates at
moderate temperatures of around 773 K. On the other hand, the
slow pyrolysis is characterized by a gentle heating of relatively
larger solid particle for longer vapour residence times, and usually
1084 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096

a lower temperature than fast pyrolysis, typically 673 K. At
industrial scales, large retorts (batch or continuous), agitated drum
kilns, rotary kilns and screw pyrolyzers are used for processing of
feed at given conditions [21]. Fig. 4 shows a screw pyrolyzer for
slow pyrolysis. In pyrolysis, any change in the operating conditions
has a signicant effect on the product yields of gas, tar and char. A
typical comparison of product yields on dry basis for different
operating modes is shown in Table 1 [22].
2.3. Biomass degradation products and uses
The pyrolysis process leads to production of a range of
products. The fuel gas produced in the process is utilized for
producing solvents such as acetone and methanol, hydrocarbons
and electricity. The bio-oil produced is used for producing differ-
ent chemicals such as levoglucosan and acetic acid, upgraded
motor fuel and electricity. Also, the char produced may be used as
either a slurry fuel, a soil improving agent or as the activated
carbon in industrial applications [23].
3. Modelling of pyrolysis processes
Due to increasing interests in the use of biomass for production
of gaseous and liquid products via pyrolysis, signicant research
has been reported on developing numerical models to better
understand this process. Most of the research efforts have been
on understanding the reaction kinetics of the pyrolysis process.
Some studies have also been conducted for reactor design and
modelling using particle scale models with due consideration to
other mechanisms such as secondary cracking reactions occurring
simultaneously inside the reactor. For developing economic and
efcient pyrolysis reactors, improved understanding of transport
phenomena inside the reactor as well as the reaction kinetics of
pyrolysis process is required.

Vapour, gas and char product Quencher 3.1. Kinetic models

Flue gas
A kinetic or reaction model refers to different types of reactions
occurring during degradation of biomass. These models are
Biomass Cyclone divided into two categories, namely the lumped and distributed
models depending on the reaction mechanism of the pyrolysis
Pyrolysis reactor Char
process.
Pyrolysis
Feeder liquid
Motor 3.1.1. Lumped models
In these models, the different reaction products of biomass and
Fluidizing gas
Pyrolysis liquid
individual components of biomass are lumped into three product
storage classes: gas, tar and char, for which several kinetic schemes have
Combustor
been proposed for primary degradation of biomass as well as
Pyrolysis gases secondary decomposition of volatile products (mainly tar or
Air higher molecular weight hydrocarbons). The lumped models are
Fig. 3. Schematic of a fast pyrolysis uidized bed reactor conguration [154]. further divided to study individual components as well as single
homogeneous species decomposition of biomass.
Specic models for decomposition of major components of
biomass (cellulose, hemicellulose and lignin) also exist. Bradbury
Biomass Hot-sand
et al. [24] proposed a three reaction model for the pyrolysis of
cellulose at low pressures by modifying the original Broido model
[25]. This model [24] assumed that an initiation reaction leads to
formation of active cellulose which subsequently decomposes by
two competitive rst-order reactions, one yielding volatiles and
Vapour and aerosol to condenser the other char and a gaseous fraction. The original model [25]
stipulated that the dehydration reactions occur at low tempera-
tures (around 493 K) causing formation of anhydrocellulose which
Auger driver further converts to char and gas exothermically, while depolymer-
ization reactions occur at higher temperatures (around 553 K)
causing formation of levoglucosan which leads to formation of
volatile tars endothermically. A three-reaction scheme for pyro-
lysis of cellulose based on two aspects: low pressures and high
Auger reactor temperatures favour cracking reactions of active cellulose to gases,
while low temperatures favours cross-linking and aromatization of
Biochar and sand active cellulose into char [26]. According to the model given by
Diebold [27], dehydration and chain cleavage (decarboxylation) of
primary and active cellulose lead to formation of char, water and
Fig. 4. Schematic of a slow pyrolysis screw reactor with heating medium [154]. carbon oxides, while vaporization of active cellulose lead to

Table 1
Product yield for different types of thermal degradation processes [22].

Mode Conditions Liquid (%) Char (%) Gas (%)

Fast Temperature  773 K, short vapour residence time  1 s 75 12 13

Moderate Temperature  773 K, moderate vapour residence time  1020 s 50 20 30
Slow Temperature  773 K, very long vapour residence time  530 min 30 35 35
Gasication Temperature 41023 K, moderate vapour residence time  1020 s 5 10 85
 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
Fig. 5. Overall decomposition of cellulose [27].
V1 V2
A B
B Char
Fig. 6. Overall decomposition of hemicellulose [29].
Lignin Char + Gas
Lignin
Tar
Lignin Gas + Reactive Vapours
Fig. 7. Overall degradation of lignin [30].
formation of primary vapours which further undergo cleavage,
dehydration, decarboxylation and decarbonylation reactions to
form secondary gas (CO, H2, CO2, CH4 and higher olens) and
secondary tar (phenolic or polycyclic aromatic tars) (Fig. 5). Lin
et al. [28] studied the mechanism of conversion of cellulose into
levoglucosan and anhydrosugar polymers, which further undergo
series of reactions such as polymerization, dehydration and con-
densation for formation of gases and char. However, the proposed
reaction pathway contradicted the scheme of Diebold [27] in terms
of energy requirement and product formation due to majority of
char formed from repolymerization of volatile anhydrosugars and
fragmented species instead of active cellulose in their model.
A model describing the kinetics of isothermal decomposition of
hemicellulose (or xylan) in inert atmosphere using a two-stage
mechanism is shown in Fig. 6. In this model [29], A is hemi-
cellulose (or xylan), B is intermediate reaction product with a
reduced degree of polymerization, and V1 and V2 are volatiles
formed during reactions.
A semi-global, multi-step model for primary lignin pyrolysis is
given in Fig. 7. According to this model [30], at low temperatures
(below 773 K), dehydration reactions caused the formation of char
and gas, while at higher temperatures (above 773 K), lignin
monomers were formed which may lead to secondary degradation
and condensation reactions; at very high heating rates, lignin
resulted into formation of gas and reactive vapours that probably
led to formation of secondary char by condensation.
Caballero et al. [31] studied the kinetics for thermal degrada-
tion of lignin as a function of distribution of solid decomposition. A
lumped model for primary and secondary pyrolysis of Kraft lignin
was also developed to analyse the cracking of tar with different
retention times in a catalytic reactor [32]. The model suggested
V1
A B
B D
Fig. 8. Multi-component mechanism of biomass decomposition [48].
1085
that with increase in the retention/reaction time, there was an
increase in the amount of lower molecular weight components
with reduction in the higher molecular weight compounds.
A number of kinetic models [3337] have also been developed
by considering biomass as a homogeneous species. In these
models, only the primary thermo-chemical degradation of bio-
mass are considered. Shazadeh and Chin model [33] discussed
the detailed chemistry of formation of char, liquid and gaseous
products from wood pyrolysis by decomposition of cellulose,
hemicellulose and lignin. It was analyzed that hemicellulose
component is the least stable and decomposes at 498598 K
(598648 K for cellulose and 523773 K for lignin). It was also
examined that cellulose and hemicellulose components produce
volatile pyrolysis products, while lignin predominantly forms a
charred residue [33]. However, the studies of Yang et al. [38]
showed that lignin has a higher temperature range (4331173 K)
for decomposition, and the main gaseous products (CO2, CO and
CH4) and organics formed by pyrolysis of cellulose, hemicellulose
and lignin were same with different yields for each component.
Thurner and Mann [34] studied a model for investigating the
kinetics of wood pyrolysis and identifying the composition of
pyrolysis products. Based on the modelling and experimental
analysis, it was found that gas consists mainly of CO2, CO, O2 and
C3 compounds with trace amounts of CH4, C2H4, and C2H2, and
tar consists mainly of levoglucosan [34]. Di Blasi and Branca [37]
investigated that variation in the product yields and kinetic rates
was mainly because of the effect of different heating rates,
operating temperatures and experimental setups.
The models were also proposed by considering biomass as a
heterogeneous mixture of different components. In these models,
the negligible effect of secondary tar degradation mechanism on
product yields was considered. A number of studies have been
performed for calculating the kinetic parameters for devolataliza-
tion of pseudo-components (mainly considering three component
mechanism for cellulose, hemicellulose and lignin) of various
biomass species in different temperature and heating rate ranges
using thermo gravimetric techniques and non-linear least squares
algorithm [3943]. Few models [4447] were proposed for con-
sidering multi-component decomposition mechanism of biomass
pyrolysis. Branca and Di Blasi [48] studied the multi-step reaction
mechanism of isothermal wood degradation and a kinetic model
was proposed considering degradation of extractives, hemicellu-
lose, cellulose and lignin in different reaction zones. As per this
model (Fig. 8), A is wood, B and D are intermediate solid-phase
reaction products, and V1, V2 and V3 are volatiles generated in
three stages with C as nal charred residue. Radmanesh et al. [49]
proposed a three-independent parallel reaction model for
V2 V3
D
C
1086 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
complete devolatilization of biomass components using a thermo-
gravimetric analyser and gas chromatograph techniques and
analyzed the effect of heating rate on yield of gases and tars.
The models [5052] were also studied for homogeneous
decomposition of tar produced by primary degradation of bio-
mass. The decomposition of tar in vapour phase, both inside the
particle and in reactor environment was considered. Liden et al.
[50] proposed a kinetic model for production of organic liquids
from ash pyrolysis of wood. In this scheme (Fig. 9), it was
assumed that wood decomposes according to two parallel reac-
tions yielding gas with char and liquid tar which further decom-
poses by secondary homogeneous reactions.
Boroson et al. [51] offered a kinetic scheme for secondary
reactions of condensable vapours through vapour phase homoge-
neous cracking (extra-particle cracking) in temperature range of 773
1073 K. Koufopanos et al. [52] model included a kinetic scheme for
the primary and secondary reactions describing pyrolysis of small
particles of main biomass components, namely cellulose, hemicellu-
lose and lignin by summing up their individual reaction rates. This
model (Fig. 10) indicated that the biomass decomposes to volatiles,
gases and char; the volatiles and gases may further react with char to
produce volatiles, gases and char of different compositions. Srivastava
et al. [53] modied this kinetic mechanism [52] for detailed study of
order of reactions during biomass pyrolysis. It was concluded by the
authors [53] that pyrolysis is faster for 0th order reaction and nal
pyrolysis temperature is higher for 1st order of reaction because time
and temperature have a linear relationship. Fagbemi et al. [54]
studied the thermal cracking of tar in the temperature range of
673 to 1173 K and found that the decrease in CO2 concentration with
temperature was due to simultaneous increase in CO (heterogeneous
CO2-solid reactions) and H2 (dehydrogenation of C2, C3) concentra-
tion. Morf et al. [55] analyzed the homogeneous secondary cracking
of tar in temperature range of 773 to 1273 K. According to this study
[55], the tar cracking resulted in production of gases like CO, CH4 and
H2 for temperatures greater than 923 K, and also favoured the
conversion of primary tar (mainly oxygenates such as acetol) into
secondary (mono-aromatics such as phenols) and tertiary tar com-
pounds (poly-aromatics such as napthalene) at higher temperatures.
Baumlin et al. [56] overcame the problems associated with assump-
tions of considering plug ow behaviour for secondary cracking
reactions of pyrolysis vapours, leading to incorrect study of reaction
temperatures and gas residence times [55,57]. It was found that on
increasing the temperatures, there was an increase in the mole
fractions of H2 and CH4, with corresponding decrease in CO2, and CO
staying nearly constant [56].
When compared with homogeneous cracking, the heteroge-
neous cracking of tar on char matrix in presence of inert metals or
minerals has not been discussed in detail in the literature. Only a
Tar Gas
Wood
Gas + Char
Fig. 9. Primary and secondary decomposition of wood [50].
Biomass
(Volatile + Gases)1 (Char)1
Fig. 10. Combined mechanism for primary and secondary decomposition of biomass [52].
few studies [5860] have been performed for analyzing the effect
of char on tar decomposition reactions. For extra-particle hetero-
geneous cracking of tars on char surface, Boroson et al. [58]
performed experiments in temperature range of 673873 K in
xed char beds and showed that there was conversion of some
amount (around 35%) of total tar (considerably aromatic in nature)
on char surfaces, causing formation of mainly gases (like CO and
CO2) and some amount of additional char (coke). Morf [59] also
conducted experimental studies for heterogeneous cracking of tars
on a xed char bed with temperature in range of 7231233 K.
It was found that the increase in concentration of non-
condensable gases is higher as compared to homogeneous crack-
ing in the same temperature range. Abu El-Rub [60] studied the
catalytic activity of biomass chars for tar reduction in gasier
product stream using a xed char bed in temperature range of
9731173 K. It was concluded that during the process, tar mole-
cules adsorb on char particle active sites to convert into gases and
coke (deposit on char surfaces) according to gasication and
polymerization reactions, respectively. However, no kinetic model
was proposed in these studies for heterogeneous cracking of tars.
3.1.2. Distributed models
In distributed models, it is considered that the pyrolysis
products are formed by innite number of independent parallel
reactions having different activation energies given using Gaussian
distribution function. Chen et al.s [61] model was based on a coal
pyrolysis model known as FG-DVC (Functional GroupDepoly-
merization, Vaporization, Cross-linking) for examining the bio-
mass pyrolysis in terms of gas evolution and tar and char
formation. However, the model did not include the impact of
secondary reactions, heating rate and mineral effect on kinetics of
biomass pyrolysis.
Niksas [62] reaction scheme (Fig. 11) was based on the depoly-
merization of a mixture of chain macromolecules into fragments that
volatilize into tar, partially decompose into gases, or recombine with
the nascent char matrix. This reaction model for biomass devolatiliza-
tion is also known as bio-FLASCHAIN (bio-FC) and was earlier used for
coal devolatilization. However, this mechanism is much more difcult
to implement because of substantial gaps in the supporting database
required for validating this model.
The Chemical Percolation Devolatilization (CPD) model [63,64]
was an extension of coal CPD model for three main biomass
components, i.e., cellulose, hemicellulose (with extractives) and
lignin, on basis of their chemical structure and behaviour during
transformation under different mechanisms. In this model, the
secondary decomposition of tar clusters to yield light gases was
not considered leading to deviation in results when compared to
experimental data. Rostami et al. [65] proposed a new model based
on distributed activation energy mechanism for biomass pyrolysis. It
was found that accuracy of the model may be improved by making
few changes in kinetic parameters which were obtained from
experimental data [65]. Few other mechanisms were proposed for
understanding thermo-chemical decomposition of different biomass
samples using distributed models [6668]. These studies utilized the
thermogravimetric analysis techniques for model validation with
experimental data. The non-linear least squares algorithm utilized in
(Volatile + Gases)2 + (Char)2
 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
Fig. 11. Distributed activation energy based reaction mechanism [62].
Fig. 12. Mass loss rate curves for different biomass samples (Heating rate40 K/
min) [68].
these models [6668] for calculation of the model parameters was
based on mass loss rate of biomass samples with respect to
temperature or time, such as shown in Fig. 12.
3.2. Particle models
In the above kinetic models, it was assumed that the degrada-
tion of biomass takes place either at very slow heating rates for
avoiding spatial gradients of temperature or at very high heating
rates for avoiding initial heating period. Both the schemes were
utilized to study the static degradation of biomass i.e., degradation
taking place isothermally during pyrolysis process. However, both
the mechanisms have some drawbacks. In former case, the weight
loss cannot be neglected during heating period, and including this
information is often not possible. In the latter case, the results are
affected by heat transfer limitations because accurate control of
sample temperature is not possible. Therefore, these reaction
models have to be further coupled with transport equations for
1087
studying the overall thermo-chemical degradation process includ-
ing heat and mass transfer effects.
Different particle models [6971] have been proposed by
modifying the Bamford model [72] to understand the physical
and chemical processes in pyrolysis of different types of biomass.
The main process in these models has been described as the
transfer of heat to the solid particles by radiation and convection
[72]. Inside the solid particle, the main heat transfer process before
thermal decomposition is conduction. The effect of convective
heat transfer between the gas and solid, and enthalpy changes
associated with the reaction and phase changes during heat
transfer to the solid are also important [73,74]. In majority of
these models, the basic assumptions were:
1. A global Arrhenius pyrolysis mechanism with regard to bio-
mass consumption.
2. Local thermal equilibrium between solid and volatiles.
3. The ideal gas law applied.
4. Darcy ow considered inside the porous medium.
Chan et al. [75] analyzed that for active devolatilization, the
mass transfer by diffusion is negligible in comparison with the
hydrodynamic ow. Koufopanos et al. [52] found that the
endothermic heat of pyrolysis is prevalent at low conversions,
while exothermic heat at high conversions. Di Blasi [76] proposed
a model to include the effect of particle shrinkage on pyrolysis
with primary and secondary reactions [77]. It was found that
particle shrinkage affects both primary and secondary reaction
paths due to different heat transfer conditions interior to particle
and reduced volatile residence times, respectively. Di Blasi [78]
also used a 2-dimensional (2-D) model for examining anisotropic
behaviour of biomass. Based on the results, it was concluded that
2-D conguration was faster with higher volatiles yield when
compared with the 1-D conguration. Janse et al. [79] described
the relevant physical and chemical processes inside a single
particle which dominate the product distribution and conversion
times of wood particles for ash pyrolysis conditions. Bryden and
Hagge [80] model included a detailed analysis of pyrolysis of a
moist, shrinking biomass particle. In this model, the combined
1088 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
impact of both moisture [81] and particle shrinkage [82] on
pyrolysis times and product yields were taken into account. The
effect of shrinkage due to char was considered [82] but not due to
volume occupied by volatiles [76]. It was concluded that pyrolysis
can be divided into three regimes: thermally thin, thermally thick
and thermal wave. The combined impact of shrinkage and moist-
ure content was considered to be mostly conned in the thermal
wave regime. Galgano and Di Blasi [83] modelled the radiative
decomposition of biomass using shrinking unreacted-core model
[84]. The progress of degradation of moist wood particle is shown
in Fig. 13. As per the model [83], the drying stage comprised of
three zones in the particle domain, i.e., moist core, dry wood and
char layer. Based on the analysis, it was found that the wood
drying and conversion times are linearly dependent on initial
moisture content. It was also concluded that the initial moisture
content affects the process variables such as pyrolysis temperature
and biomass reactivity.
Babu and Chaurasia [85] proposed a model for examining the
impact of shrinkage on particle size, pyrolysis time, product yields,
specic heat capacity and Biot number. According to the authors,
the shrinkage of particle favoured the attainment of larger yields
of volatiles and gases produced by primary reactions at the
expense of other two products. Yuen et al. [86] formulated a 3-D
model for pyrolysis of wet wood by incorporating the anisotropic
properties of wood and internal convection of gases with a single
Heat transfer by
conduction, convection
and radiation
Water vapour, gases and
tar outflow by diffusion
and convection
Moist Dry Char
wood wood
Fig. 13. Schematic for showing decomposition progress in a moist wood particle
(cross section of particle) (adapted and modied from Ref. [83]).
Fig. 14. Mass loss (daf or dry ash free basis) prole for sawdust particles with
different shapes and similar mass/volume (reactor temperature 1625 K, gas
temperature 1310 K, equivalent particle diameter 0.32 mm) [91].
kinetic reaction for the overall devolatilization process. Some other
simplied models [87,88] for degradation of biomass considering
heat transfer with primary and secondary chemical reactions have
also been developed.
Sreekanth and Kolar [89] provided a 2-D model for thermal
decomposition of single wood particle in a Fluidized Bed Reactor
(FBR). A comparison was made with 1-D models with respect to
thermal wave propagation inside particle due to anisotropic effect.
Park et al. [90] studied the degradation mechanism of wood to
analyse the effect of endo/exothermicity of the reactions occurring
during the process. The model did not account for the particle
shrinkage and diffusive ux for gaseous species. It was suggested
that at high temperatures, thick wood particle may split by a
combination of high internal pressure and weakened structure. Lu
et al. [91] used their model to analyse the effect of particle shape
and size on rate of biomass devolatilization. The authors [91]
contradicted the results of Janse et al.[79] by stating that spherical
particles have slower rates of heat and mass transfer as compared
to other aspherical particles of same volume/mass (Fig. 14). It was
also found that the spherical particles produce less volatile yields
relative to other shapes [91]. Peters [92] developed a model for
analyzing the pyrolysis rate of different biomass samples for
different heating rates and particle geometries. However, the effect
of particle shrinkage and moisture content were not taken into
account [91,92].
3.3. Reactor models
The single particle models can be used for reactor modelling by
taking extra-particle processes into account, such as the effect of
residence time of the gas phase and the solid phase on primary
decomposition and secondary tar cracking mechanism with other
hydrodynamics considerations inside the reactor. Phenomenolo-
gical as well as CFD models have been used to study the biomass
pyrolysis in various reactor congurations for analyzing the effect
of process parameters on the reactor hydrodynamics and the
product yields. Below we briey discuss both of these approaches:
3.3.1. Phenomenological models
For slow or conventional pyrolysis, rotary kiln [93] and xed
bed [94] reactor congurations have been studied. Di Blasi et al.
[95] proposed a 2-D mathematical model for pyrolysis of wheat
straw in a convective heated pyrolyzer by considering it to be a
xed bed reactor due to long residence time of solids and high
solid to gas density. Yang et al. [94] examined segregated solid
wastes in a packed-bed pyrolyzer for analyzing the nal yields of
gas, tar and char in temperature range of 623973 K and heating
rate around 10 K/min. It was found that the nal quantity of char is
30100% more in packed bed pyrolyzer as compared to Thermo-
gravimetric Analyser (TGA) studiespossibly due to tar cracking
and repolymerization, both inside the pyrolyzing particles and on
activated outer surface of other particles.
For fast pyrolysis, FBR is a prominent option for effective heat
and mass transfer during the process. A number of simplied
models to study the devolatilization of biomass in uidized bed
environment have been proposed [9698]. The biomass fast
pyrolysis has also been studied both experimentally and numeri-
cally in rotating cone [99] and entrained ow [100] reactors.
Difelice et al. [101] studied the devolatilization of a single biomass
particle in a FBR. It was found that biomass particles oat or sink
depending on the bulk density of the bed particles, and based on
this, the feeding point for biomass should be decided. Di Blasi
[102] proposed a model for pyrolysis of cellulose and wood [103]
in an FBR, where a single particle model [104] developed earlier
was coupled with an external heat transfer model [105]. The
 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
model also accounted for the effect of bed hydrodynamics on
particle conversion. The single particle and reactor model results
were compared with experimental results in terms of gas, tar and
char yields (Fig. 15). It was found that at higher bed temperatures
(greater than 900 K), the activity of secondary reactions in the
freeboard zone is not negligible, and the volatile residence time is
smaller than thoses calculated using the bubble velocity of two-
phase theorypossibly due to the net ux addition by the fast
devolatilization rate.
Luo et al. [106] discussed ash pyrolysis of wood in an FBR and
the effect of main operational parameters on product distribution.
According to modelling [106] and experimental results [107],
maximum pyrolysis oil production occurs at dense bed tempera-
ture of around 773 K. It was examined that wood particle needs to
be of moderate size because the ne particles are carried easily by
gas ow leading to short residence time and incomplete decom-
position, whereas the larger particle sizes reduce the overall oil
production due to slower heating rate. Kaushal and Abedi [108]
developed a 1-D model for biomass pyrolysis using a two- step
kinetic model with the two-phase hydrodynamic model (bubble
and emulsion phase) for an FBR. It was found that the cracking and
gas phase reactions are important at temperatures higher than
773 K; and the bed hydrodynamic and mass transfer governs the
pyrolysis rate at these higher temperatures.
3.3.2. CFD models
In last few years, CFD modelling has proven its utility as a
powerful tool for analyzing the reactor performance for different
kind of processes. Multi-phase CFD modelling of pyrolysis process
is essentially a combination of Eulerian and Lagrangian appr-
oaches. These modelling approaches are further sub-divided into
different classes, such as EulerEuler and EulerLagrangian.
Table 2 describes applicability of these models at different scales.
Fig. 15. Particle and reactor model results comparison (for yield of gas (YG), tar (YT)
and char (YC); reactor temperature Tr in Kelvin; volatiles residence time tc in
seconds) with experimental data [102].
Table 2
Description of different CFD models [155].
Name Gas phase
Discrete bubble model (Model 1) Lagrangian
Two-uid model (Model 2) Eulerian
Unresolved discrete particle model (Model 3) Eulerian (unresolved)
Resolved discrete particle model (Model 4) Eulerian (resolved)
Molecular dynamics (Model 5) Lagrangian
1089
Among these models (Table 2), Models 13 are can be suitable
for macro to large-scale uidized bed studies with Two-Fluid
Model (TFM, EulerianEulerian) [109] and Discrete Particle Model
(DPM, EulerianLagrangian) [110] being used more widely. Dis-
crete Bubble Model (DBM) [111] is still in its early stages of
development and has not been used extensively. The EulerEuler
modelling approach considers different phases (gassolid, gas
liquid, liquidsolid) as inter-penetrating continuum, with each
phase occupying xed volume fraction inside the reactor. Con-
servation equations are derived for each phase based on Navier
Stokes relationships, while some additional closure laws are
required to describe particle phase constitutive relations, such as
given by kinetic theory of granular ows [112114] for solid
phases. According to this theory, granular particles are considered
as molecular gases, but with energy losses due to collisions
between grains. Furthermore, kinetic energy associated with
uctuating motion of particles is dened as granular temperature
of particles. The solids pressure and viscosity are considered as a
function of granular temperature while dening particles motion
in a dense bed. However, EulerLagrangian modelling approach
considers uid phase as a continuum, while dispersed phase is
considered by tracking a given number of particles, bubbles or
droplets through the continuous phase using Newtons laws of
motion. Some constitutive relations are required, such as for
accounting energy dissipation by particle collisions using hard-
sphere (by empirical coefcient of restitution and friction) [115] or
soft-sphere (empirical spring stiffness and friction coefcient)
[116] approach. The EulerLagrangian approach of hydrodynamic
modelling is better in comparison to EulerEuler approach for
obtaining detailed information about motion of particles in bed,
but it has drawback in terms of large computational requirements
to keep macroscopic ow time scale in range with time scale of
particle-particle interactions. Therefore, for industrial scale sys-
tems, EulerEuler is still considered a better option for analyzing
the hydrodynamics.
CFD models have been proposed for examining biomass ther-
mal decomposition in the presence of air/steam [117,118] and inert
gas environment [119121]. The main focus of these studies
remains on uidized bed reactors because of their inherent
benets of excellent heat and mass transfer rates with uniform
product distribution. A detailed model for study of pyrolysis in
uidized bed was proposed by Lathouwers and Bellan [122].
However, this model had drawbacks in terms of fullment of
energy requirement of reactor and achievement of high tempera-
tures of around 498 K for biomass feed. The model was used to
optimize the tar yield as a function of operating parameters [123].
Papadikis et al. [119] studied the pyrolysis of biomass using Euler
EulerLagrangian approach [124,125] and compared different drag
models, namely Gidaspow, Syamlal OBrien, and Wen-Yu, based on
their impact on heat transfer, degradation rate, product yields and
char residence time during biomass pyrolysis [126]. The effect of
particle size on heat transfer coefcient has also been studied
[127]. The main drawback of these studies [119,126,127] was that
only 1 or 2 reacting particles were considered, and the simulation
time has been limited to 3 to 5 s of total real time of process.
Solid phase Gassolid coupling Scale
Eulerian Drag closures for bubbles Industrial (10 m)
Eulerian Gassolid drag closures Engineering (1 m)
Lagrangian Gas-particle drag closures Laboratory (0.1 m)
Lagrangian Boundary condition at particle surface Laboratory (0.01 m)
Lagrangian Elastic collisions at particle surface Mesoscopic ( o 0.001 m)
1090 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
Fig. 16. Simulated product yields for different biomass feedstocks (particle diame-
ter 0.5 mm, reactor temperature773 K, gas velocity 3x min. uidization velo-
city) [128].
Oevermann et al. [117] presented an EulerLagrangian model for
wood gasication in bubbling uidized bed reactor. Gerber et al.
[118] discussed an EulerEuler model for wood gasication in
bubbling uidized bed using char as the bed material. However,
the mechanism chosen for secondary tar reactions in reactor
environment did not account for the catalytic effect of char bed
on cracking step, and hence, these reactions were considered in gas
phase only [118]. Xue et al. [120] formulated an EulerEuler model
for the fast pyrolysis of biomass in uidized bed. The reactor
hydrodynamics was coupled with reaction kinetics and biomass
transport based on a simplied single-particle model. It was found
that in 2-D models, the grid sensitivity did not have a signicant
impact on the apparent biomass density, vertical velocity, average
temperature and product yields [128]. However, in comparison, the
3-D model resulted into some variation in these parameters near
bed surface and biomass feed location but without any substantial
change in the product yields. The simulated product yields of
different types of biomass were calculated and compared for same
operating conditions (Fig. 16). Bruchmuller et al. [121] analyzed
thermo-chemical decomposition of biomass using CFD-DEM
approach. The modelling results for a 3-D domain were analyzed
for bed hydrodynamics with different inlet distributor plates as well
as for the effect of uidization velocity, temperature and moisture
content on product yield and composition.
4. Pyrolysis process parameters
The thermo-chemical decomposition of biomass in inert atmo-
sphere is dependent on various process parameters such as feed-
stock type, operating conditions and physico-chemical properties
of biomass, which ultimately affect the biomass conversion time or
pyrolysis rate with product distribution and quality. Below we
briey discuss some of these important process parameters:
4.1. Type of feedstock
Cellulose, hemicellulose and lignin with other components
such as extractives (generally smaller organic molecules or poly-
mers) and minerals (inorganic compounds) are present in differing
proportions in different biomass types and these proportions
inuence the product distributions on pyrolysis [129]. At pyrolysis
temperatures, the main biomass components contribute to pro-
duct yields broadly as follows: cellulose and hemicellulose com-
ponents provide the volatile pyrolysis products, while lignin
predominantly forms a charred residue [33]. However, according
to Di Blasi and Branca [37], the holocellulose (cellulose and
hemicellulose) converts mainly to liquids (tars), while the char
and gas products are due to lignin decomposition. Extractives
contribute to liquid and gas products either through simple
volatilization or decomposition; and minerals, particularly the
alkali metals (especially sodium, potassium and calcium), gener-
ally remain in the char and have a catalytic effect on pyrolysis
reactions leading to increased char yields depending on other
conditions [130], in addition to the effect of ash contribution to
char yield [131]. This elemental contribution in ash also degrades
the quality of pyrolysis oil due to some catalytic reactions, and its
removal affects the process by causing an increase in pyrolysis
liquid and decrease in gas products [107].
4.2. Heating rate and temperature
A rapid heating and cooling of primary vapours are required to
minimize the extent of secondary reactions that reduce the liquid
yield and have a negative impact on its quality [33]. In addition,
the slow heating leads to higher char yields [132].
The temperature also has a signicant impact on char yields
and properties. Higher temperatures result into lower char yield in
all pyrolysis reactions. The primary reason for this is of the
stripping of more volatile material from the char at higher
temperatures, thus causing reduction in the yields. For example,
the char yield decreases from 31% to 17% with increase in the
temperature from 638 K to 879 K [90]. However, there is also an
adverse impact of lower temperatures, which may lead to incom-
plete decomposition of biomass leading to higher amount of
unpyrolyzed solid in char content. The temperature also affects
the char composition with chars produced at higher temperatures
having higher carbon contents [34,129]. The solid product char
contains more than 85 wt% carbon for temperatures greater than
773 K [133]. Liquid yields increase on increasing the pyrolysis
temperatures up to a maximum value, usually at 673823 K, but is
highly dependent on other operating conditions. For a variety of
feedstock types, it has been studied that the maximum yield of
pyrolysis oil is obtained at temperatures of around 673823 K with
a consistent decrease in char yield and corresponding increase in
gas yield [132]. Above this temperature range, secondary reactions
causing vapour decomposition become more dominant and the
condensed liquid yields were reduced [90]. Gas yields are gen-
erally low below the peak temperature (773 K) for liquid yield;
above this, gas yields increase strongly by higher temperatures, as
the main products of vapour decomposition are gases [134]. The
best pyrolysis oil quality in terms of caloric value and hydrogen
over carbon ratio is achieved at the maximum yield temperature
i.e., around 773 K [135]. For temperature ranging between 723 and
823 K, the water content is about 3035 wt% of total liquid yield
and for higher reaction temperatures, there is considerable
increase in water content of liquid fraction of pyrolysis product
due to secondary cracking reactions leading to decrease in heating
value of pyrolysis oil [133]. However, the higher heating values of
non-condensable gases are obtained at reaction temperatures
greater than 723 K in pyrolysis, due to increased production of
CO and CH4 as compared to CO2 [107].
4.3. Volatiles residence time and pressure
The residence time of volatiles is dependent on gas ow rate
through the reactor which ultimately affects the contact time
between primary vapours and hot char, and thus affects the
severity of secondary reactions and also volatile product proper-
ties [50,136]. Lower oil yields at prolonged vapour residence time
are due to cracking and polymerization reactions of vapours to
gases and solids, respectively. However, it was found that the loss
 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
of tar yield is limited to around 10% for vapour residence time
from 1 to 5 s at temperatures around 500 K. Short volatiles
residence times below 1 s affects the biomass decomposition
process by causing incomplete depolymerization of lignin due to
random bond cleavage and inter-reaction of lignin macromolecule,
nally resulting in less homogeneous liquid product [137].
The pyrolysis process is also signicantly affected by the operat-
ing pressure, which when increased leads to smaller specic
volume of volatile causing higher intra-particle residence time
which favours their decomposition while escaping the biomass
particle. This also leads to higher concentration (partial pressure) of
volatiles, thus increasing the decomposition reaction rate through
secondary reactions [5]. It has been found that higher pressures
lead to higher char fractions [25]. Alves and Figueiredo [129]
showed that the chars formed under higher pressure conditions
resulted into higher xed-carbon yields. This effect is useful in
maximizing the carbon sequestration potential of biochars.
4.4. Particle size, shape and orientation
Particle size has a signicant effect on the heat and mass
transfer phenomena of pyrolysis process. Larger particles imply
larger thermal gradients and also the uid residence times are
sufciently long leading to secondary reactions [75]. The increas-
ing particle size also causes reduction in the liquid yields due to
secondary reactions activity leading to increases in gas yields for
external temperatures greater than 800 K [103]. Bradbury et al.
[25] analyzed that small biomass size leads to smaller char
fraction. However, experimental studies [133] show that there is
no signicant inuence of the increase in particle size (0.717 mm)
on the product yields. However. It did decrease the heating rate
from 1000 K/s to 1.5 K/s; and there was a marginal decrease in the
liquid yields and around 5% decrease in the gas yields with
corresponding increase in the char yields when the particle size
was increased from 17 to 20 mm. Increasing the particle size also
led to increase in the water content (40 to 55 wt%) of pyrolysis
liquid and decrease in the carbon content (78.5 to 75 wt%) of the
solid char product [133].
Particle shape also inuences the pyrolysis process. Park et al.
[90] stated that spherical particles have lesser char yield and
conversion time when compared to slab-shaped and cylindrical
particles. However, the studies of Lu et al. [91] showed that
spherical particles have the smallest surface to volume ratio,
leading to slower rate of heat and mass transfer and higher
conversion time in comparison to other aspherical particles. At
small particle diameters (typically less than 0.2 mm), the rate of
reaction becomes dominant, and the different particle shapes
show nearly equal conversion times [79].
Grain orientation is an important parameter in biomass pyr-
olysis due to the anisotropic behaviour of biomass. Roberts [74]
pointed that the permeability of ow along the grains was 104
times that across the grain and thermal conductivity along the
grains was twice that across the grains. Chan et al. [138] concluded
that s perpendicular grain heating decreased the tar yield with
corresponding increase in the char, water (dehydration reaction)
and gas yields. They attributed this to lower thermal conductivity
(almost one-third as compared to parallel or tangential grain
direction) but not due to increase in residence time owing to
reduction in porosity. Di Blasi [76] proposed that the secondary
reactions occur to a larger extent for perpendicular grain heating
as compared to parallel grain heating.
4.5. Reactor conguration
Different reactor congurations have been studied for the
thermal decomposition of biomass in absence of air. The most
1091
important factor which inuences reactor selection is the medium
of heat transfer to biomass particles inside the reactor during
degradation process. In ablative type, heat transfer is due to the
contact of biomass particles with the hot surface. This process uses
larger particles of biomass and is affected by the rate of heat
supply to the reactor [137]. The heat transfer gas or carrier gas is
not required, but supplying heat to biomass is problematic because
of high heat losses. The major challenge of this conguration are
achieving long residence times for the biomass particles to allow a
high degree of conversion, higher degree of char attrition and
higher carry-over of carbon into pyrolysis liquid product [136]. In
the vacuum pyrolysis, the heat is transferred by direct contact with
hot surface. This process uses short residence time of volatiles and
larger biomass particles. However, this method has poor heat and
mass transfer rates and requires larger scale equipment [136].
Circulating Fluidized Bed (CFB) uses heat source (uidizing or
carrier gas and solid like sand) for transferring heat to the biomass
particles by both convection and conduction. This process has heat
transfer limitations and requires particle sizes less than about
3 mm for good liquid yields [137]. CFB conguration has certain
disadvantages related to non-uniformity of biomass particle resi-
dence times, requirement of solid recycling of partially reacted
feed and of post treatment of pyrolysis liquid to reduce char
content due to carryover, high char attrition, ash build-up in
circulating solids causing cracking of organic molecules in volatile
products and loss of pyrolysis oil yield [136]. Bubbling Fluidized
Bed (BFB) uses same mode of heat transfer to biomass particles as
that of CFB, but with variation in contribution of convective and
conductive heat transfer. In this process, pyrolysis oil quality is
enhanced with high concentration of desirable chemicals and
minimum carryover of char (micro-carbon) into liquid product.
There is also a rapid elution of char from the reactor; thereby
reducing the chances of volatile cracking [136].
4.6. Physico-chemical properties
There are other parameters such as physical and chemical
properties of biomass which can inuence the pyrolysis process.
These are discussed as below:
4.6.1. Thermal conductivity and emissivity
With increase in the thermal conductivity (0.030.4 W/m/K) of
biomass, there is often an increase in the nal tar yield and
decrease in the gas yield, keeping char yield constant with some
reduction in the conversion time. This is due to increase in the rate
of heat transfer, which consequently increases the ow velocity
and reduces the intra-particle residence times and secondary
reactions activity [139]. Babu and Chaurasia [140] concluded that
any increase in the emissivity leads to considerable increase in the
secondary reaction products at the expense of primary reaction
products. This also reduces the pyrolysis time due to increase in
the temperature inside the biomass particle.
4.6.2. Permeability and density
With increase in the char permeability, there is a decrease in
mass ow resistance and increase in ow velocity causing reduc-
tion in the secondary reaction activity that increases the tar yield
and decrease the gas yield. However, the conversion time does not
change with the char permeability [139]. On the other hand, any
increase in the density, there is decrease in the primary tar yield
and increase in the char and gas yields because of the lower
temperatures during primary pyrolysis which favours the charring
reactions. There is also an increase in conversion time on increase
in density due to lower particle heating rates. However, if activity
of secondary reactions is considered, the tar yield remains
1092 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
constant because of the increase in the primary degradation rates
or volatiles release rate with increase in density, ultimately
balancing the effect of lower primary degradation temperature
[139].
4.6.3. Specic heat capacity and heat of reaction
Any increase in specic heat capacity of biomass results into
larger thermal capacity and higher conversion time. There is also a
reduction in the secondary reaction activity and products with
increase in the specic heat because of reduction in intra-particle
residence time [139]. However, Babu and Chaurasia [85] found
that specic heat capacity does not affect the primary and
secondary reaction product yields. Conversion time is also not
affected by any change in the specic heat capacity as the
energetics of secondary reactions is not important [85]. The char
forming reactions are considered exothermic in nature, while tar
evaporation as endothermic; therefore, the extent of these reac-
tions depend on other operating conditions of the process like
pressure and ow rate [141]. It was also observed that cellulose
decomposition is an endothermic process at lower conversions
(below 573 K), while lignin decomposition exothermic at higher
conversion [52].
4.6.4. Particle shrinkage and moisture content
On increasing the shrinkage level, the tar content increases [76]
and the char content decreases [103]. This may be attributed to
higher temperatures at the primary reaction front (reaction
temperature) leading to higher reaction rates for tar formation,
and lower rates of tar degradation. The higher heat transfer rates
of the shrinking particle also reduce the pyrolysis or conversion
time of biomass degradation [85,103]. Biomass naturally contains
some moisture which exists in two basic forms i.e., bound or
hygroscopic water, and free or capillary water. Bound water is
bonded to the hydroxyl groups of the major constituents of
biomass including cellulose, hemicellulose and lignin. After occu-
pation of all available sites, the medium is at its Fiber Saturation
Point (FSP), which is around 30% of dry weight for different types
of wood. The free water (above FSP), however, is present in the
liquid form in the lumens or voids of the wood and is held by weak
capillary forces [142]. When the biomass pyrolysis is conducted at
higher temperatures, the water vapour mixes with the volatiles
formed from pyrolysis and ows out of the solid thereby affecting
the heat and mass transfer processes during pyrolysis. With
increase in the moisture content of biomass, the rise in surface
temperature slows down because of energy consumption in the
evaporation reaction, but at the same time, it can cause some
changes in biomass structure leading to changes in the porosity,
permeability and uid ow. But overall, any increase in the
moisture content increases the pyrolysis time with negligible
effect on product yield [81]. The combined impact of the particle
shrinkage and moisture content results in overall increase in the
pyrolysis time with substantial increase in the tar yield and
decrease in the gas yield for large biomass particles [80].
4.6.5. External heat transfer coefcient
Di Blasi [103] studied different external heat transfer correla-
tions (like innite heat transfer condition, Agarwals model, Ranz-
Marshall and whole-bed correlations), and found that both con-
version time and char yield increases with decrease in the value of
external heat transfer coefcient. Babu and Chaurasia [140] con-
cluded that the concentration of volatiles increases with increase
in the convective heat transfer at lower operating temperatures.
Kersten et al. [132] concluded that any increase in external heat
transfer coefcient for smaller particles (13 mm) cause reduction
in the conversion time with no effect on the product yield.
However, for larger particles, no signicant reduction in the
conversion time is observed because of change in the controlling
step from external heat transfer to internal heat transfer due to
low thermal conductivity of biomass.
5. Catalytic pyrolysis
The presence of catalyst during pyrolysis process results into
cracking reactions and upgradation of biomass pyrolysis products,
depending on the operating conditions. Catalyst type and reactor
conguration play an important role in the production of primary
products during pyrolysis process. The primary pyrolysis vapours
can also be catalytically cracked over different catalysts to give
liquid and gaseous fuels. In particular, oxygenates in the pyrolysis
oil, which reduce the specic energy content of pyrolysis oil can be
decreased using zeolite type catalysts [17]. The application of
different catalysts in biomass pyrolysis process and upgradation
of pyrolysis products has been discussed by Bridgewater [143].
Kinetic models have been proposed for analyzing the biomass
catalytic pyrolysis and also the catalytic reactions of vapour
products formed during the thermal conversion process [144
146]. This study focuses on in-situ catalytic upgrading of biomass
pyrolysis products. The effect of different catalysts on ex-situ
upgrading of pyrolysis oil has been discussed elsewhere [147].
5.1. Process features
There are a number of important ways in which catalysts
modify pyrolysis processes.
1. Decomposition temperature of biomass components may sig-
nicantly decrease with a suitable catalyst.
2. Catalysts affect network of reaction, i.e., deoxygenation which
allows in situ upgrading of bio-oil by reducing oxygenated
organic compounds. The presence of catalysts also reduces
polymerization precursors (like multifunctional phenols) for
stabilizing bio-oil.
3. Decarboxylation, decarbonization and dehydration reactions
may release more CO, CO2 and H2O when catalysts are
employed.
4. Catalysts promote formation of coke due to dehydration reac-
tions (mainly due to high acidity of catalysts).
The effect of different catalysts on biomass pyrolysis and
product distribution is shown in Table 3.
5.2. Biochar catalytic effect
In biomass pyrolysis, the intra-particle secondary reactions can
occur both homogeneously in gas phase as well as heteroge-
neously on solid char matrix, and the extra-particle secondary
reactions in reactor environment can include vapour phase crack-
ing as well as heterogeneous conversion on the surface of
biomass-derived chars [5]. There are natural catalysts in biomass
which substantially inuence products of pyrolysis process [134].
The presence of metallic minerals in biomass contributes to
change in outlet composition of pyrolysis products for given
operating conditions [130]. Boroson et al. [58] found that tar
formed by primary pyrolysis of wood is very reactive in the
presence of wood char causing formation of gaseous components
and char (or coke). Di Blasi [77], and Gronli and Melaaen [148]
examined the conversion of tar into char molecules due to
repolymerization step during the biomass particle degradation.
Bridgwater et al. [137] also analyzed that the hot char is catalyti-
cally active and favours the cracking of vapours during fast
 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096 1093

Table 3
Effect of different catalysts on biomass pyrolysis products.

Publication Catalyst Biomass Reactor conguration Products Findings

Comparison of the products ZSM-5 zeolite Rice husk Fluidized bed reactor with condensation Oxygenated oil The oxygen containing species in oils
from the pyrolysis and and off-line gas analysis system compounds (such convert into H2O, CO and CO2 in
catalytic pyrolysis of rice as phenols, presence of catalysts. Increase in
husks [156] cresols) and aromaticity of the oil compounds due to
gaseous products catalytic activity. Also, the heavy
oxygenated compounds in the oil
convert into lower molecular weight
oxygenated species.
Yields of hydrogen-rich ZnCl2, Na2CO3 Cotton Stainless steel cylindrical reactor with Gaseous products Gas yield increases with temperature.
gaseous products via and K2CO3 cocoon electrically heated tubular furnace (CO, CO2, Olens, H2-rich gaseous products increase with
pyrolysis from selected shell, tea O2, H2, parafns), ZnCl2, but pyrolytic gas decreases.
biomass samples [157] factory liquid and char Na2CO3 is better for cotton coccon shell
waste, olive and tea waste than K2CO3.
husk
Catalytic pyrolysis of biomass Cr2O3, and Rice straw, Batch pyrolysis assembly-reactor part Gaseous products Cr2O3 is better than other metal oxides
for hydrogen rich fuel gas some metal sawdust (pyrolysis reactor, cracking reactor and (H2 containing gas, for total as well as H2 containing gas
production [158] oxides (MnO, heating furnace), condenser and CO2, CO), tar, water yield.
FeO, Al2O3, purication part (condenser and dryer) and char
CaO, CuO) and gas storage part (gas tank or water
tank)
Kinetic description of the HZSM-5 Sawdust Conical spouted bed reactor with water Gas, liquid (light With increase in catalyst mass, yield of
catalytic pyrolysis of zeolite condensers and lter in downstream and heavy gas increases, while liquid decreases
biomass in a conical fraction), and charnotably with slight decrease in amount
spouted bed reactor [159] of char. Decarboxylation,
decarbonylation and dehydration
reactions dominate the process
producing CO2, CO, H2O and coke.
Pyrolysis oil is more stable than liquid
from thermal pyrolysis and can easily be
used for further processing such as
steam reforming.
Catalytic fast pyrolysis of Fresh (FC) and Corncob Fluidized bed reactor with condensers, Gas, liquid and FC and SC remarkably decreases the oil
biomass in a uidized bed spent (SC) FCC lters, gas collection bag char fraction, while increase water, coke and
with fresh and spent non-condensable gas yield. Catalyst with
uidized catalytic cracking strong acidity should be avoided for
(FCC) catalysts [160] higher oil fraction and less coke. Most of
oil fraction is light oil fraction, with
reduced multifunctional components of
phenols (polymerization precursors).
Comparison of non-catalytic HZSM-5 Corncob Fluidized bed reactor with condensers, Gas, liquid and HZSM-5 catalyst considerably decreases
and catalytic fast pyrolysis zeolite lters, gas collection bag char the oil yield, while increase water, coke
of corncob in a uidized and non-condensable gas yield. The O2
bed reactor [161] content in heavy oil fraction converts
mainly to CO, CO2 and H2O. And there is
an increase in aromatic hydrocarbons in
oil content due to presence of catalyst.
Catalytic pyrolysis of biomass Nickel, cobalt, Cellulose, Tubular quartz micro-reactor and a semi- Hydrocarbons, The substituted catalysts are better than
for biofuels production iron, and lignin and continuous ow reactor oxygenates such as conventional catalysts for higher
[162] gallium- wood phenol and cresol, hydrocarbon yield. The deoxygenation
substituted and coke activity reduces with time due to
ZSM-5 deposition of coke on catalyst, leading to
catalyst regeneration after certain period
of cracking process.

pyrolysis process. The experiments were conducted for studying
tar cracking using biomass char as a catalyst in different tempera-
ture ranges, and it was concluded that both char and thermal
cracking are responsible for conversion of tar into gas, secondary
tar and char [149,150]. With these studies [149,150], it was also
found that pore structure of char particle and mineral content are
important factors for biomass char activity towards tar conversion
through gasication and polymerization reactions. Experimental
studies have also been performed to analyse the low temperature
catalytic cracking effect of char on biomass pyrolysis [151] and oil
renery products [152]. The catalytic activity of biochar has been
compared with other conventional catalysts for tar reduction, and
the in-situ cracking of tar compounds [60]. Based on this study
[60], it was found that biochar is a very promising catalyst for
solving the tar utilization problems in gasication systems and
there is requirement to perform more experimental and modelling
studies in this direction.
6. Summary and future work
A number of mathematical and CFD modelling studies have been
conducted for understanding the effect of different operating condi-
tions such as biomass composition, particle size, reactor temperature,
and residence time of volatiles. The models have been developed at
different scales, i.e., kinetic, particle and reactor, for analyzing the
parameters such as reaction mechanism, mixing/segregation beha-
viour of different phases. Studies have also been carried out for
catalytic cracking of biomass and catalytic upgrading of pyrolysis oil
for making its appropriate use as a source of energy.
1094 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
There is, however, uncertainty over the ability of current
pyrolysis models in optimizing product yields for different reactor
congurations. For example, kinetics models of primary degrada-
tion and secondary tar cracking reactions are often not able to
accurately predict the pyrolysis rate for different biomass samples.
One of the major reason for this drawback is that the effect of
biochar activity on secondary tar cracking mechanism have not
been explicitly accounted for in these models. Also, the distributed
activation energy models are at an early stages of development,
and need to be developed further to enhance their ability to
analyse detailed decomposition of biomass and tar into various
gaseous, liquid and solid products. Similarly, the particle scale
models are not able to account for the overall impact of different
operating parameters such as drying and particle shrinkage as a
function of thermo-physical properties of biomass. This has led to
inconsistencies in predicting the heat and mass transfer rates in
the biomass particle. Finally, the reactor scale models also do not
clearly examine the effect of above mentioned parameters, espe-
cially the in-situ catalytic activity of biochar, which signicantly
increases production of volatiles, thus lowering the residence time.
This then eventually affects the reactor hydrodynamics leading to
inaccurate predictions of product yields.
Therefore, more research and development effort is needed to
design industrial-scale pyrolysis reactors. First, a comprehensive
particle scale model needs to be formulated by considering the
overall effect of process parameters such as drying, shrinkage and
the biochar catalytic activity during biomass pyrolysis. Second, in
order to understand the mixing/segregation behaviour of biomass
and biochar particles, detailed multiphase models of the system
should be developed. These multiphase models should then be
integrated with other constitutive models to account for the effect
of operating conditions such as temperature and particle size on the
behaviour of different phases in the reactor and product yields.
Finally, multi-scale modelling and experimental studies are
required to accurately analyse the tar cracking mechanisms in the
presence of different catalysts including the in-situ catalytic effect of
biochar on the product yield in different reactor congurations.
Acknowledgments
The authors gratefully acknowledge the nancial support
provided of the Australia-India Strategic Research Funding
(GCF020010) and Australian Research Council under the ARC
Linkage Projects Scheme (ARC Linkage Project LP100200135) with
BHP Billiton Iron Ore Pty Ltd. and Ansac Pty Ltd. as the partner
organisations.
References
[1] Demirba A. Global renewable energy resources. Energy Sources Part A:
Recovery Util Environ Eff 2006;28:77992.
[2] Zhang Q, Chang J, Wang T, Xu Y. Review of biomass pyrolysis oil properties
and upgrading research. Energy Convers Manage 2007;48:8792.
[3] Demirbas MF, Balat M, Balat H. Potential contribution of biomass to the
sustainable energy development. Energy Convers Manage 2009;50:174660.
[4] Demirbas A, Arin G. An overview of biomass pyrolysis. Energy Sources
2002;24:47182.
[5] Di Blasi C. Modeling chemical and physical processes of wood and biomass
pyrolysis. Prog Energy Combust Sci 2008;34:4790.
[6] Ruiz JA, Jurez MC, Morales MP, Muoz P, Mendvil MA. Biomass gasication
for electricity generation: review of current technology barriers. Renewable
Sustainable Energy Rev 2013;18:17483.
[7] Demirbas A. Combustion characteristics of different biomass fuels. Prog
Energy Combust Sci 2004;30:21930.
[8] Raveendran K, Ganesh A, Khilar KC. Inuence of mineral matter on biomass
pyrolysis characteristics. Fuel 1995;74:181222.
[9] Toor SS, Rosendahl L, Rudolf A. Hydrothermal liquefaction of biomass: a
review of subcritical water technologies. Energy 2011;36:232842.
[10] Agarwal PK, Genetti WE, Lee YY. Model for devolatilization of coal particles
in uidized beds. Fuel 1984;63:115765.
[11] Heidenreich CA, Yan HM, Zhang DK. Mathematical modelling of temperature
response of low-rank coal particles during pyrolysis. Dev Chem Eng Miner
Process 1999;7:593610.
[12] Heidenreich CA, Yan HM, Zhang DK. Mathematical modelling of pyrolysis of
large coal particlesestimation of kinetic parameters for methane evolution.
Fuel 1999;78:55766.
[13] Heidenreich CA, Zhang DK. Measuring the temperature response of large wet
coal particles during heating. Fuel 1999;78:9914.
[14] Shen L, Zhang DK, Yan HM, Roach JR, Nguyen QD. Low temperature pyrolysis
of an australian brown coal in a uidised bed reactor. Dev Chem Eng Miner
Process 2000;8:293309.
[15] Yip K, Wu H, Zhang Dk. Effect of inherent moisture in collie coal during
pyrolysis due to in-situ steam gasication. Energy Fuels 2007;21:288391.
[16] Damartzis T, Zabaniotou A. Thermochemical conversion of biomass to
second generation biofuels through integrated process designa review.
Renewable Sustainable Energy Rev 2011;15:36678.
[17] Yaman S. Pyrolysis of biomass to produce fuels and chemical feedstocks.
Energy Convers Manage 2004;45:65171.
[18] Zabaniotou AA. Pyrolysis of forestry biomass by-products in Greece. Energy
Sources Part A: Recovery Util Environ Eff 1999;21:395403.
[19] Fitz HC, DeBellevue EB, Costanza R, Boumans R, Maxwell T, Wainger L, et al.
Development of a general ecosystem model for a range of scales and
ecosystems. Ecol Modell 1996;88:26395.
[20] Lehmann J, Gaunt J, Rondon M. Bio-char sequestration in terrestrial ecosys-
temsa review. Mitig Adaptat Strateg Global Change 2006;11:395419.
[21] Brownsort PA. Biomass pyrolysis processes: review of scope, control and
variability. UKBRC working paper 5; 2009.
[22] Brown R. Chapter 8. In: Lehmann J, Joseph S, editors. Biochar production
technology. London: Biochar for Environmental Management Earthscan;
2009.
[23] Balat M, Balat M, Krtay E, Balat H. Main routes for the thermo-conversion of
biomass into fuels and chemicals. Part 1: Pyrolysis systems. Energy Convers
Manage 2009;50:314757.
[24] Bradbury AGW, Sakai Y, Shazadeh F. A kinetic model for pyrolysis of
cellulose. J Appl Polym Sci 1979;23:327180.
[25] Broido A, Nelson MA. Char yield on pyrolysis of cellulose. Combust Flame
1975;24:2638.
[26] Agrawal RK. Kinetics of reactions involved in pyrolysis of cellulose I. The
three reaction model. Can J Chem Eng 1988;66:40312.
[27] Diebold JP. A unied, global model for the pyrolysis of cellulose. Biomass
Bioenergy 1994;7:7585.
[28] Lin Y-C, Cho J, Tompsett GA, Westmoreland PR, Huber GW. Kinetics and
mechanism of cellulose pyrolysis. J Phys Chem C 2009;113(2009):7107.
[29] Di Blasi C, Lanzetta M. Intrinsic kinetics of isothermal xylan degradation in
inert atmosphere. J Anal Appl Pyrolysis 1997;4041:287303.
[30] Antal MJ. In: Boer KW, Dufe JA, editors. Biomass pyrolysis: a review of the
Literature. Part ICarbohydrate pyrolysis, 1. Advances in Solar Energy
Boulder (CO): American Solar Energy; 1982. p. 61111.
[31] Caballero JA, Font R, Marcilla A, Conesa JA. New kinetic model for thermal
decomposition of heterogeneous materials. Ind Eng Chem Res 1995;34:80612.
[32] Adam M, Ocone R, Mohammad J, Berruti F, Briens C. Kinetic investigations of
kraft lignin pyrolysis. Ind Eng Chem Res 2013.
[33] Shazadeh F, Chin Peter PS. Thermal deterioration of wood. Wood technol-
ogy: chemical aspects. American Chemical Society; 1977. p. 5781.
[34] Thurner F, Mann U. Kinetic investigation of wood pyrolysis. Ind Eng Chem
Process Des Dev 1981;20:4828.
[35] Font R, Marcilla A, Verdu E, Devesa J. Kinetics of the pyrolysis of almond
shells and almond shells impregnated with cobalt dichloride in a uidized
bed reactor and in a pyroprobe 100. Ind Eng Chem Res 1990;29:184655.
[36] Samolada MC, Vasalos IA. A kinetic approach to the ash pyrolysis of
biomass in a uidized bed reactor. Fuel 1991;70:8839.
[37] Di Blasi C, Branca C. Kinetics of primary product formation from wood
pyrolysis. Ind Eng Chem Res 2001;40:554756.
[38] Yang H, Yan R, Chen H, Lee DH, Zheng C. Characteristics of hemicellulose,
cellulose and lignin pyrolysis. Fuel 2007;86:17818.
[39] Orfo JJM, Antunes FJA, Figueiredo JL. Pyrolysis kinetics of lignocellulosic
materialsthree independent reactions model. Fuel 1999;78:34958.
[40] Grnli MG, Vrhegyi G, Di Blasi C. Thermogravimetric analysis and devola-
tilization kinetics of wood. Ind Eng Chem Res 2002;41:42018.
[41] Many JJ, Velo E, Puigjaner L. Kinetics of biomass pyrolysis: a reformulated
three-parallel-reactions model. Ind Eng Chem Res 2002;42:43441.
[42] Mszros E, Vrhegyi G, Jakab E, Marosvlgyi B. Thermogravimetric and
reaction kinetic analysis of biomass samples from an energy plantation.
Energy Fuels 2004;18:497507.
[43] Branca C, Albano A, Di Blasi C. Critical evaluation of global mechanisms of
wood devolatilization. Thermochim Acta 2005;429:13341.
[44] Alves SS, Figueiredo JL. Pyrolysis kinetics of lignocellulosic materials by
multistage isothermal thermogravimetry. J Anal Appl Pyrolysis
1988;13:12334.
[45] Koufopanos CA, Lucchesi A, Maschio G. Kinetic modelling of the pyrolysis of
biomass and biomass components. Can J Chem Eng 1989;67:7584.
[46] Miller RS, Bellan J. A generalized biomass pyrolysis model based on super-
imposed cellulose, hemicelluloseand liqnin kinetics. Combust Sci Technol
1997;126:97137.
 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
[47] Ranzi E, Cuoci A, Faravelli T, Frassoldati A, Migliavacca G, Pierucci S, et al.
Chemical kinetics of biomass pyrolysis. Energy Fuels 2008;22:4292300.
[48] Branca C, Di Blasi C. Kinetics of the isothermal degradation of wood in the
temperature range 528708 K. J Anal Appl Pyrolysis 2003;67:20719.
[49] Radmanesh R, Courbariaux Y, Chaouki J, Guy C. A unied lumped approach
in kinetic modeling of biomass pyrolysis. Fuel 2006;85:121120.
[50] Liden A, Berruti F, Scott DS. A kinetic model for the production of liquids
from the ash pyrolysis of biomass. Chem Eng Commun 1988;65:20721.
[51] Boroson ML, Howard JB, Longwell JP, Peters WA. Product yields and
kinetics from the vapor phase cracking of wood pyrolysis tars. AIChE
J 1989;35:1208.
[52] Koufopanos CA, Papayannakos N, Maschio G, Lucchesi A. Modelling of the
pyrolysis of biomass particles. Studies on kinetics, thermal and heat transfer
effects. Can J Chem Eng 1991;69:90715.
[53] Srivastava VK, Sushil, Jalan RK. Prediction of concentration in the pyrolysis of
biomass materialII. Energy Convers Manage 1996;37:47383.
[54] Fagbemi L, Khezami L, Capart R. Pyrolysis products from different biomasses:
application to the thermal cracking of tar. Appl Energy 2001;69:293306.
[55] Morf P, Hasler P, Nussbaumer T. Mechanisms and kinetics of homogeneous
secondary reactions of tar from continuous pyrolysis of wood chips. Fuel
2002;81:84353.
[56] Baumlin S, Broust F, Ferrer M, Meunier N, Marty E, Ld J. The continuous
self stirred tank reactor: measurement of the cracking kinetics of biomass
pyrolysis vapours. Chem Eng Sci 2005;60:4155.
[57] Rath J, Staudinger G. Cracking reactions of tar from pyrolysis of spruce wood.
Fuel 2001;80:137989.
[58] Boroson ML, Howard JB, Longwell JP, Peters WA. Heterogeneous cracking of
wood pyrolysis tars over fresh wood char surfaces. Energy Fuels
1989;3:73540.
[59] Morf PO. Secondary reactions of tar during thermochemical biomass con-
version; 2001.
[60] Abu El-Rub ZYK. Biomass char as an in-situ catalyst for tar removal in
gasication systems. Enschede 2008.
[61] Chen Y, Charpenay S, Jensen A, Wjtowicz MA, Serio MA. Modeling of
biomass pyrolysis kinetics. In: Symposium (international) on combustion,
1998;27:132734.
[62] Niksa S. Predicting the rapid devolatilization of diverse forms of biomass
with bio-ashchain. Proc Combust Inst 2000;28:272733.
[63] Sheng C, Azevedo JLT. Modeling biomass devolatilization using the chemical
percolation devolatilization model for the main components. Proc Combust
Inst 2002;29:40714.
[64] Shen L, Zhang DK. A CPD modelling study of sewage sludge devolatilisation
for oil production. In: Proceedings of the third Asia-Pacic conference on
combustion, Seoul, Korea; 2001. p. 43740.
[65] Rostami AA, Hajaligol MR, Wrenn SE. A biomass pyrolysis sub-model for CFD
applications. Fuel 2004;83:151925.
[66] Becidan M, Vrhegyi G, Hustad JE, Skreiberg . Thermal decomposition of
biomass wastes. A kinetic study. Ind Eng Chem Res 2007;46:242837.
[67] Sonobe T, Worasuwannarak N. Kinetic analyses of biomass pyrolysis using
the distributed activation energy model. Fuel 2008;87:41421.
[68] Varhegyi Gb, Bobaly Bz, Jakab E, Chen H. Thermogravimetric study of
biomass pyrolysis kinetics. A distributed activation energy model with
prediction tests. Energy Fuels 2010;25:2432.
[69] Kung H-C. A mathematical model of wood pyrolysis. Combust Flame
1972;18:18595.
[70] Kansa EJ, Perlee HE, Chaiken RF. Mathematical model of wood pyrolysis
including internal forced convection. Combust Flame 1977;29:31124.
[71] Pyle DL, Zaror CA. Heat transfer and kinetics in the low temperature
pyrolysis of solids. Chem Eng Sci 1984;39:14758.
[72] Bamford CH, Crank J, Malan DH. The combustion of wood. Part I. Math Proc
Camb Philos Soc 1946;42:16682.
[73] Kanury AM, Blackshear PL. On the combustion of wood II: The inuence of
internal convection on the transient pyrolysis of cellulose. Combust Sci
Technol 1970;2:59.
[74] Roberts AF. The heat of reaction during the pyrolysis of wood. Combust
Flame 1971;17:7986.
[75] Chan W-CR, Kelbon M, Krieger BB. Modelling and experimental verication
of physical and chemical processes during pyrolysis of a large biomass
particle. Fuel 1985;64:150513.
[76] Di Blasi C. Heat, momentum and mass transport through a shrinking biomass
particle exposed to thermal radiation. Chem Eng Sci 1996;51:112132.
[77] Di Blasi C. Analysis of convection and secondary reaction effects within
porous solid fuels undergoing pyrolysis. Combust Sci Technol
1993;90:31540.
[78] Di Blasi C. Physico-chemical processes occurring inside a degrading two-
dimensional anisotropic porous medium. Int J Heat Mass Transfer
1998;41:413950.
[79] Janse AMC, Westerhout RWJ, Prins W. Modelling of ash pyrolysis of a single
wood particle. Chem Eng Process 2000;39:23952.
[80] Bryden KM, Hagge MJ. Modeling the combined impact of moisture and char
shrinkage on the pyrolysis of a biomass particle. Fuel 2003;82:163344.
[81] Bryden KM, Ragland KW, Rutland CJ. Modeling thermally thick pyrolysis of
wood. Biomass Bioenergy 2002;22:4153.
[82] Hagge MJ, Bryden KM. Modeling the impact of shrinkage on the pyrolysis of
dry biomass. Chem Eng Sci 2002;57:281123.
1095
[83] Galgano A, Di Blasi C. Modeling the propagation of drying and decomposi-
tion fronts in wood. Combust Flame 2004;139:1627.
[84] Galgano A, Di Blasi C. Modeling wood degradation by the unreacted-core-
shrinking approximation. Ind Eng Chem Res 2003;42:210111.
[85] Babu BV, Chaurasia AS. Heat transfer and kinetics in the pyrolysis of
shrinking biomass particle. Chem Eng Sci 2004;59:19992012.
[86] Yuen RKK, Yeoh GH, de Vahl Davis G, Leonardi E. Modelling the pyrolysis of
wet woodI. Three-dimensional formulation and analysis. Int J Heat Mass
Transfer 2007;50:437186.
[87] Sadhukhan AK, Gupta P, Saha RK. Modelling and experimental studies on
pyrolysis of biomass particles. J Anal Appl Pyrolysis 2008;81:18392.
[88] Damartzis T, Kostoglou M, Zabaniotou A. Simulation of the agro-biomass
(olive kernel) fast pyrolysis in a wire mesh reactor considering intra-particle
radial and temporal distribution of products. Int J Chem Reactor Eng 2009:7.
[89] Sreekanth M, Kolar AK. Progress of conversion in a shrinking wet cylindrical
wood particle pyrolyzing in a hot uidized bed. J Anal Appl Pyrolysis
2009;84:5367.
[90] Park WC, Atreya A, Baum HR. Experimental and theoretical investigation of
heat and mass transfer processes during wood pyrolysis. Combust Flame
2010;157:48194.
[91] Lu H, Ip E, Scott J, Foster P, Vickers M, Baxter LL. Effects of particle shape and
size on devolatilization of biomass particle. Fuel 2010;89:115668.
[92] Peters B. Validation of a numerical approach to model pyrolysis of biomass
and assessment of kinetic data. Fuel 2011;90:230114.
[93] Klose W, Wiest W. Experiments and mathematical modeling of maize
pyrolysis in a rotary kiln. Fuel 1999;78:6572.
[94] Yang YB, Phan AN, Ryu C, Shari V, Swithenbank J. Mathematical modelling
of slow pyrolysis of segregated solid wastes in a packed-bed pyrolyser. Fuel
2007;86:16980.
[95] Di Blasi C, Branca C, Teislev B. Development of a novel reactor for the
oxidative degradation of straw. Bioresour Technol 2004;91:26371.
[96] De Diego LF, Garca-Labiano F, Abad A, Gayn P, Adnez J. Modeling of the
devolatilization of nonspherical wet pine wood particles in uidized beds.
Ind Eng Chem Res 2002;41:364250.
[97] Jand N, Foscolo PU. Decomposition of wood particles in uidized beds. Ind
Eng Chem Res 2005;44:507989.
[98] Saastamoinen JJ. Simplied model for calculation of devolatilization in
uidized beds. Fuel 2006;85:238895.
[99] Wagenaar BM, Prins W, van Swaaij WPM. Pyrolysis of biomass in the rotating
cone reactor: modelling and experimental justication. Chem Eng Sci
1994;49:510926.
[100] Brown AL, Dayton DC, Nimlos MR, Daily JW. Design and characterization of
an entrained ow reactor for the study of biomass pyrolysis chemistry at
high heating rates. Energy Fuels 2001;15:127685.
[101] Difelice R, Coppola G, Rapagna S, Jand N. Modeling of biomass devolatiliza-
tion in a uidized bed reactor. Can J Chem Eng 1999;77:32532.
[102] Di Blasi C. Modelling the fast pyrolysis of cellulosic particles in uid-bed
reactors. Chem Eng Sci 2000;55:59996013.
[103] Di Blasi C. Modeling intra- and extra-particle processes of wood fast
pyrolysis. AIChE J 2002;48:238697.
[104] Di Blasi C. Kinetic and heat transfer control in the slow and ash pyrolysis of
solids. Ind Eng Chem Res 1996;35:3746.
[105] Agarwal PK. Transport phenomena in multi-particle systemsIV. Heat
transfer to a large freely moving particle in gas uidized bed of smaller
particles. Chem Eng Sci 1991;46:111527.
[106] Luo Z, Wang S, Cen K. A model of wood ash pyrolysis in uidized bed
reactor. Renewable Energy 2005;30:37792.
[107] Luo Z, Wang S, Liao Y, Zhou J, Gu Y, Cen K. Research on biomass fast pyrolysis
for liquid fuel. Biomass Bioenergy 2004;26:45562.
[108] Kaushal P, Abedi J. A simplied model for biomass pyrolysis in a uidized bed
reactor. J Ind Eng Chem 2010;16:74855.
[109] Pain CC, Mansoorzadeh S, Gomes JLM, de Oliveira CRE. A numerical
investigation of bubbling gassolid uidized bed dynamics in 2-D geome-
tries. Powder Technol 2002;128:5677.
[110] Kafui KD, Thornton C, Adams MJ. Discrete particle-continuum uid modelling
of gassolid uidised beds. Chem Eng Sci 2002;57:2395410.
[111] Bokkers GA, Laverman JA, van Sint Annaland M, Kuipers JAM. Modelling of
large-scale dense gassolid bubbling uidised beds using a novel discrete
bubble model. Chem Eng Sci 2006;61:5590602.
[112] Lun CKK, Savage SB, Jeffrey DJ, Chepurniy N. Kinetic theories for granular
ow: inelastic particles in Couette ow and slightly inelastic particles in a
general ow eld. J Fluid Mech 1984;140:22356.
[113] Ding J, Gidaspow D. A bubbling uidization model using kinetic theory of
granular ow. AIChE J 1990;36:52338.
[114] Syamlal M, Rogers W, OBrien TJ. MFIX documentation theory guide. Other
information: PBD; Dec 1993. p. Medium: ED; size: 49 p.
[115] Hoomans BPB, Kuipers JAM, Briels WJ, van Swaaij WPM. Discrete particle
simulation of bubble and slug formation in a two-dimensional gas-uidised
bed: a hard-sphere approach. Chem Eng Sci 1996;51:99118.
[116] Tsuji Y, Kawaguchi T, Tanaka T. Discrete particle simulation of two-
dimensional uidized bed. Powder Technol 1993;77:7987.
[117] Oevermann M, Gerber S, Behrendt F. EulerLagrange/DEM simulation of
wood gasication in a bubbling uidized bed reactor. Particuology
2009;7:30716.
1096 A. Sharma et al. / Renewable and Sustainable Energy Reviews 50 (2015) 10811096
[118] Gerber S, Behrendt F, Oevermann M. An Eulerian modeling approach of wood
gasication in a bubbling uidized bed reactor using char as bed material.
Fuel 2010;89:290317.
[119] Papadikis K, Gu S, Bridgwater AV. CFD modelling of the fast pyrolysis of
biomass in uidised bed reactors: modelling the impact of biomass shrink-
age. Chem Eng J 2009;149:41727.
[120] Xue Q, Heindel TJ, Fox ROA. CFD model for biomass fast pyrolysis in uidized-
bed reactors. Chem Eng Sci 2011;66:244052.
[121] Bruchmller J, van Wachem BGM, Gu S, Luo KH, Brown RC. Modeling the
thermochemical degradation of biomass inside a fast pyrolysis uidized bed
reactor. AIChE J 2012;58:303042.
[122] Lathouwers D, Bellan J. Modeling of dense gassolid reactive mixtures
applied to biomass pyrolysis in a uidized bed. Int J Multiphase Flow
2001;27:215587.
[123] Lathouwers D, Bellan J. Yield optimization and scaling of uidized beds for
tar production from biomass. Energy Fuels 2001;15:124762.
[124] Papadikis K, Bridgwater AV, Gu S. CFD modelling of the fast pyrolysis of
biomass in uidised bed reactors, Part A: Eulerian computation of momen-
tum transport in bubbling uidised beds. Chem Eng Sci 2008;63:421827.
[125] Papadikis K, Gu S, Bridgwater AV. CFD modelling of the fast pyrolysis of
biomass in uidised bed reactors. Part B: Heat, momentum and mass
transport in bubbling uidised beds. Chem Eng Sci 2009;64:103645.
[126] Papadikis K, Gu S, Fivga A, Bridgwater AV. Numerical optimization and
scaling of uidized beds for tar production from biomass. Energy Fuels
2010;24:213345.
[127] Papadikis K, Gu S, Bridgwater AV. Computational modelling of the impact of
particle size to the heat transfer coefcient between biomass particles and a
uidised bed. Fuel Process Technol 2010;91:6879.
[128] Xue Q, Dalluge D, Heindel TJ, Fox RO, Brown RC. Experimental validation and
CFD modeling study of biomass fast pyrolysis in uidized-bed reactors. Fuel
2012;97:75769.
[129] Alves SS, Figueiredo JL. A model for pyrolysis of wet wood. Chem Eng Sci
1989;44:28619.
[130] Nik-Azar M, Hajaligol MR, Sohrabi M, Dabir B. Mineral matter effects in rapid
pyrolysis of beech wood. Fuel Process Technol 1997;51:717.
[131] Antal MJ, Grnli M. The art, science, and technology of charcoal production.
Ind Eng Chem Res 2003;42:161940.
[132] Kersten SRA, Wang X, Prins W, van Swaaij WPM. Biomass pyrolysis in a
uidized bed reactor. Part 1: Literature review and model simulations. Ind
Eng Chem Res 2005;44:877385.
[133] Wang X, Kersten SRA, Prins W, van Swaaij WPM. Biomass pyrolysis in a
uidized bed reactor. Part 2: Experimental validation of model results. Ind
Eng Chem Res 2005;44:878695.
[134] Babu BV, Chaurasia AS. Modeling, simulation and estimation of optimum
parameters in pyrolysis of biomass. Energy Convers Manage
2003;44:213558.
[135] Scott DS, Piskorz J, Bergougnou MA, Graham R, Overend RP. The role of
temperature in the fast pyrolysis of cellulose and wood. Ind Eng Chem Res
1988;27:815.
[136] Scott DS, Majerski P, Piskorz J, Radlein D. A second look at fast pyrolysis of
biomassthe RTI process. J Anal Appl Pyrolysis 1999;51:2337.
[137] Bridgwater AV, Meier D, Radlein D. An overview of fast pyrolysis of biomass.
Org Geochem 1999;30:147993.
[138] Chan WCR, Kelbon M, Krieger-Brockett B. Single-particle biomass pyrolysis:
correlations of reaction products with process conditions. Ind Eng Chem Res
1988;27:226175.
[139] Di Blasi C. Inuences of physical properties on biomass devolatilization
characteristics. Fuel 1997;76:95764.
[140] Babu BV, Chaurasia AS. Parametric study of thermal and thermodynamic
properties on pyrolysis of biomass in thermally thick regime. Energy Convers
Manage 2004;45:5372.
[141] Mok WSL, Antal MJ. Effects of pressure on biomass pyrolysis. II. Heats of
reaction of cellulose pyrolysis. Thermochim Acta 1983;68:16586.
[142] Moghtaderi B. The state-of-the-art in pyrolysis modelling of lignocellulosic
solid fuels. Fire Mater 2006;30:134.
[143] Bridgwater AV. Production of high grade fuels and chemicals from catalytic
pyrolysis of biomass. Catal Today 1996;29:28595.
[144] Adjaye JD, Bakhshi NN. Catalytic conversion of a biomass-derived oil to fuels
and chemicals II: Chemical kinetics, parameter estimation and model
predictions. Biomass Bioenergy 1995;8:26577.
[145] Lv P, Chang J, Wang T, Wu C, Tsubaki N. A kinetic study on biomass fast
catalytic pyrolysis. Energy Fuels 2004;18:18659.
[146] Lu C, Song W, Lin W. Kinetics of biomass catalytic pyrolysis. Biotechnol Adv
2009;27:5837.
[147] Mortensen PM, Grunwaldt JD, Jensen PA, Knudsen KG, Jensen AD. A review of
catalytic upgrading of bio-oil to engine fuels. Appl Catal, A: Gen
2011;407:119.
[148] Grnli MG, Melaaen MC. Mathematical model for wood pyrolysiscomparison
of experimental measurements with model predictions. Energy Fuels
2000;14:791800.
[149] Abu El-Rub Z, Bramer EA, Brem G. Experimental comparison of biomass chars
with other catalysts for tar reduction. Fuel 2008;87:224352.
[150] Gilbert P, Ryu C, Shari V, Swithenbank J. Tar reduction in pyrolysis vapours
from biomass over a hot char bed. Bioresour Technol 2009;100:604551.
[151] Shen L, Zhang DK. Low-temperature pyrolysis of sewage sludge and putres-
cible garbage for fuel oil production. Fuel 2005;84:80915.
[152] Zhang Y, Wu J-h, Zhang DK. Cracking of simulated oil renery off-gas over a
coal char, petroleum coke, and quartz. Energy Fuels 2008;22:11427.
[153] Di Blasi C. Modeling and simulation of combustion processes of charring and
non-charring solid fuels. Prog Energy Combust Sci 1993;19:71104.
[154] Brown RC. Biorenewable resources: engineering new products from agricul-
ture. Ames, IA: Blackwell Press; 2003.
[155] van der Hoef MA, Annaland MV, Deen NG, Kuipers JAM. Numerical simula-
tion of dense gassolid uidized beds: a multiscale modeling strategy. Annu
Rev Fluid Mech 2008;40:4770.
[156] Williams PT, Nugranad N. Comparison of products from the pyrolysis and
catalytic pyrolysis of rice husks. Energy 2000;25:493513.
[157] Demirbas A. Yields of hydrogen-rich gaseous products via pyrolysis from
selected biomass samples. Fuel 2001;80:188591.
[158] Chen G, Andries J, Spliethoff H. Catalytic pyrolysis of biomass for hydrogen
rich fuel gas production. Energy Convers Manage 2003;44:228996.
[159] Atutxa A, Aguado R, Gayubo AG, Olazar M, Bilbao J. Kinetic description of the
catalytic pyrolysis of biomass in a conical spouted bed reactor. Energy Fuels
2005;19:76574.
[160] Zhang H, Xiao R, Wang D, Zhong Z, Song M, Pan Q, et al. Catalytic fast
pyrolysis of biomass in a uidized bed with fresh and spent uidized
catalytic cracking (FCC) catalysts. Energy Fuels 2009;23:6199206.
[161] Zhang H, Xiao R, Huang H, Xiao G. Comparison of non-catalytic and catalytic
fast pyrolysis of corncob in a uidized bed reactor. Bioresour Technol
2009;100:142834.
[162] French R, Czernik S. Catalytic pyrolysis of biomass for biofuels production.
Fuel Process Technol 2010;91:2532.