Biomass pyrolysis is a fundamental thermochemical conversion process that is of both industrial and ecological importance. In order to designing and operate with industrial biomass conversion systems (gasification or pyrolysis reactors), understanding of solid state pyrolysis kinetics is imperative. The upsurge of interest in simulation and optimization of the reactors for thermochemical processes requires appropriate models that help to achieve a better understanding of the reactions in the corresponding processes . In this sense, a better knowledge of the kinetics concerning to the thermal decomposition of the lignocellulosic materials is required .
The thermal decomposition of biomass proceeds via a very complex set of competitive and concurrent reactions and thus the exact mechanism for biomass pyrolysis remains a mystery . The chemistry and complication of the process which involves hundreds of intermediates have paved the way for development of numerous kinetic models in the past .
A great portion of publications have presented contradictory results, which induced a great deal of pessimism about the applicability of reaction kinetics for the evaluation of thermoanalytical curves . The cause of the problem must be searched mainly in the application of oversimplified kinetic equations for processes composed from several chemical, physical, and physicochemical subprocesses. Careless experimental work and poor mathematical evaluation techniques have also contributed to the wrong performance of the reaction kinetics in this field .
The present review, include the main concepts related to pyrolysis of lignocellulosic materials, also includes studies concerning primary and secondary decomposition.
Pyrolysis behavior of biomass
Pyrolysis is a complex process in which organic matter is thermally decomposed in the absence of externally supplied oxidizing agents. When lignocellulosic materials are exposed to inert high-temperature atmosphere, they degrade into hundreds of species generally classified, from the practical point of view, in char (the rich non-volatiles solid residue) and volatile products (low molecular weight gaseous species, in addition to all condensible, aqueous and high molecular weight organic compounds or tars . These chemical degradations are catalogued as primary reactions, mainly endothermic.
The proportions of the product yields depend on process parameters such as (peak) temperature, heating rate, pressure, fuel particle size and fuel composition, including the presence or absence of catalytically active substances.
The heating rate of the biomass particles is the most important parameter for pyrolysis with regard to the product yield distribution. It is difficult to maintain higher heating rates in laboratory conditions, those usually achieved in gasification or pyrolysis reactors . Slow pyrolysis (heating rates in the order of 10Â°C/min) is applied for maximum char coal yields , fast or even flash pyrolysis (heating rates up to 104Â°C/min) provide maximum yields of pyrolysis oils. On the other hand, Akita and Kasestudied the pyrolysis of Î±-cellulose at TGA heatingrates ranging from 0.23 Â°C/min to 2.4 Â°C/min. They applied the nthorder reaction mechanism to determine the kinetic parameters includingreaction order, activation energy and frequency factor. Their conclusionwas that the kinetic parameters were independent of the heating rate atleast for the lower values of heating rate.
The effect of particle sizeis important parameter for pyrolysis with regard to the product yield distribution.For an example, small samples give less char coal then larger samples. The sample size influence on the char yield is explained by the residential time of the volatiles, which react with the char layer when flowing out the particle to form char coal .The long residence time of the vapor phase inside large particles explains the formation of higher char coal yield. It takes longer time for the volatiles to leave a large particle than a small . Lower char coal yield for small sample masses (powder and single particle samples), could be explained by the bigger surface area that interacts with the pyrolysis medium. Formed volatile products leave the sample without undergoing secondary cracking reactions . Also, for small sizes, above certain temperatures, the time for total conversion become shorter than times needed for the reactor (TGA) to attain the final temperature . In the case of larger particles, secondary cracking reactions could be dominant, leading to additional char and tar formation.
Both the yield and the quality of the charcoal product are strongly influenced by the peak temperature of the pyrolysis process . As the peak temperature increases above 200 oC, the solid pyrolytic residue changes from "toasted" wood to "torrefied" wood to "pyrochar" to conventional char . The process of "torrefactionn" involves heating the biomass substrate to a peak temperature between 200 and280 oC.The solids residence time also affects the yield and quality of the char product, but usually this time is selected after the peak temperature is determined . When the biomass substrate is heated to somewhat higher temperatures, but not exceeding about 350 oC, a "pyrochar" is produced . The term "pyrochar" is employed herein broadly to include char as well as partially carbonized woody material that has been pyrolyzed at least sufficiently to destroy its fibrous character . This material is formed in about 50% yield and has lost the fibrous character of the biomass feedstock and has a volatile matter content of 35% or more . Above 350oC conventional char, having a volatile matter content of less than 35%, is formedfrom thebiomass sample . Several authorsindicate that it is difficult to control the peak temperature in this regimebecause of the exothermicity of the pyrolysis reactions in industrial-scale reactors (the peak temperatureusually is not defined within narrow limits) .
Improved yields of char are obtained when pyrolysis is conducted at elevated pressures. Mok et al. (1992) found that an increase in pressure from 0.1 to 1.0 MPa (at constant purge gas velocity) increased the char yield to 41%.
In thermogravimetric analyses of lignocellulosic materials, two or three peaks appear. These peaks can be assigned to each of biomass components (extractives, cellulose, hemicelluloses and lignin) indicating that their basic identity is maintained.
Lignocellulose materials are composed of holocellulose, lignin, extraneous materials or extractives, and ash. The holocellulose, or carbohydrate fraction, consisting of cellulose and the hemicelluloses, composes from 70% to 85% of most woody biomass . Cellulose, a linear macromolecule of anhydro-Î²-glucopyranose units with elemental formula (C6H1005)n, is the main component of the cell wall . Because of cellulose crystalline structure, it is not readily hydrolyzable. The hemicelluloses are easily hydrolyzed to simple sugars. Lignin, composing from 15% to 30% by weight of woody biomass, has a complex aromatic structure composed of phenyl propane units . Plant lignin is almost insoluble, and its chemical isolation is accompanied by marked changes in its molecular structure .
Hemicellulose begins to decompose at about 225â°C, cellulose decomposes in temperature range of 325-350â°C, lignin decomposes in temperature range 250-500â°C . Lignin thermally decomposes over a broad temperature range, because various oxygen functional groups from its structure have different thermal stabilities, their scission occurring at different temperatures . The cleavage of the functional groups gives low molecular weight products, while the complete rearrangement of the backbone at higher temperatures leads to 30- 50 wt% char and to the release of volatile products .Lignin is the major source of char, whereas the carbonhydrates (cellulose and hemicelluloses) are the precursors of the volatile products (tar and gases) .
The thermal decomposition of biomass consist of two kind of reactions, primary and secondary reactions. Primary reactions are the fast initial decomposition of biomass components (cellulose, hemicelluloses, lignin and extractives). Products of primary reactions are volatile gases, tar and char coal . Secondary reactions occur when temperature is above 700 - 800 oC and this reaction occurs between primary reaction product . Vapors (tars and volatiles) formed by primary decomposition of biomass components can be involved in secondary reactions in the gas phase, forming soot, or at hot surfaces, especially hot char surfaces where a secondary char coal is formed.After tar evolution from the solid phase, tars vapors are subject to secondary tar reactions. Products of tar decomposition are gases and char coal. This tars decompositions is probably catalyzed by the char coal (formed by primary reactions) . Tar conversion by secondary tar reactions already occurs in the pores of the "mother" fuel particle (intraparticle) as well as in the gas phase and on surfaces outside the particle (extraparticle) , Figure 1.The importance of secondary reactions increases with longer residence time of vapor phase (volatile gases and tar) and lower heating rate. With long residence time and low heating rate, formation and escape of vapor phase will be slower and contact between vapor phase and char coal will be extended.
Regarding to above mentioned, char coal contains both "primary" char coal and "secondary" char coal .
Figure 1Formation of char 
In general, higher charcoal yields can be obtained from sample with higher ash contents. The mineral matter and trace elements (such as Ca, K, Na, Mg, and Fe), catalyze biomass thermal decomposition reactions. Minerals in biomass, particularly the alkali metals and alkali earth metals serve as catalysts for char formation . The catalytic activity of a mineral matter and trace elements depends on: chemical form, amount, inclusion size.There are at least two competitive mechanisms for the formation of char: one initiated and catalyzed by the mineral matters and the other due to the secondary reactions of tar .
The reaction kinetics of biomass pyrolysis is important to the design and control thermochemical conversion of biomass to char coal, liquid and fuel gases. Also, the upsurge of interest in the simulation and optimization of the reactors for thermochemical processes requires appropriate models that integrate different operational conditions and varied feedstocks and help to achieve a better understanding of the reactions in the corresponding processes . Thermogravimetric analysis (TGA) is a high-precision method for the study of the pyrolysis at low heating rates, under well-defined conditions in the kinetic regime. It can provide information on the partial processes and reaction kinetics .
Thermogravimetric studies showed that each kind of biomass had unique pyrolysis characteristics, by virtue of the specific proportions of the components present in it . Even the same chemical species may have differing reactivity if their pyrolysis is influenced by other species in their vicinity . The biomass components react independently and, therefore, the thermal behaviour of biomass is also reflected by the individual behavior of the biomass components. The assumption of a distribution on the reactivity of the biomass components (chemical species) frequently helps in the kinetic evaluation of the pyrolysis of complex organic samples .
Degradation kinetics of lignocellulosic fuels was studied in either dynamic or static conditions. Static conditions are achieved by maintaining the selected constant temperatures in the pyrolyzing chamber. During dynamic conditions, biomass particles submitted in pyrolyzing chamber experience an increase in temperature with time according to an assigned heating rate.In the static analysis, tests are carried out according to two different methodologies to attain the isothermal stage . In the first methodology, the small dynamic stage consists of very slow heating rates to avoid spatial gradients of temperature, while in the second methodology, very fast, external, heat-transfer rates to keep short the first dynamic stage are used . In the first methodology the weight loss is not negligible during heating and the subsequent interpretation of the data may be lacking an important part of the whole process . In the second methodology the results can be affected by heat transfer limitations. This can be avoided if an accurate control of the sample temperature is accomplished.
What should we expect from a good kinetic model? The answer of this question depends obviously on the interest of the investigator and on the properties of the studied samples . Nevertheless, Várhegyi(2007) a few general criteria can obviously be listed :
1. Description of the behavior of the samples in a wide range of experimental conditions.
2. Prediction of the behavior outside the domain of the given set of observations.
3. Characteristics that can reveal similarities and differences between the samples.
4. A deeper insight into the processes taking place
Pyrolysis kinetics, coupled with the description of transport phenomena, produce advanced computational tools for the design and optimization of chemical reactors applied for thermochemical conversion of wood and biomass . In this section, after a brief presentation of the problems encountered in carrying out measurements of weight loss under a pure kinetic control , the literature results on the chemical kinetics of wood and biomass are reviewed.
Biomass pyrolysis involves numerous extremely complex reactions and end up with large number of intermediates and end products, devising an exact reaction mechanism and kinetic modeling for biomass pyrolysis is extremely difficult, hence, pyrolysis models are modeled on the basis of visible kinetics . From a theoretical point of view, an endless variety and complexity of reactions forming a network can be assumed in biomass pyrolysis . Hence even today it is difficult to develop a precise kinetic model taking into account all the parameters concerned.
Thee chemical degradations of biomass during pyrolysis are catalogued as primary and secondary reactions.
Fundamentals of thermal analysis
When lignocellulosic materials are exposed to inert high-temperature atmosphere, they degrade into hundreds of species generally classified, from the practical point of view, in char (the rich non-volatiles solid residue) and volatile products (low molecular weight gaseous species, in addition to all condensible, aqueous and high molecular weight organic compounds or tars . These chemical degradations are catalogued as primary reactions, mainly endothermic.
Several studies (see [64-67]) suggest that primary decomposition rates of biomass can be modeled taking into account the thermal behavior of the main components and their relative contribution in the chemical composition.
The pyrolysis of wood and related lignocellulose substances is frequently described by a single reaction:
The rate of mass loss depends on mass and temperature according to following equation:
Î±, t, T and k defines the reacted fraction, time, absolute temperature in K, and the rate constant, respectively .
The variation of the temperature-dependent reaction rate constant is approximated by the Arrhenius rate expression:
wherek(T) is the temperature-dependent reaction rate constant, A is the frequency factor (pre-exponential factor), R is the universal gas constant, and Ea is the activation energy of the reaction.
It should be noted that every kinetic model proposed employs a rate law that obeys the fundamental Arrhenius rate expression.
The function is approximated by:
where () is the remaining fraction of volatile material in the sample and n is the reaction rate.
If the original mass is , the final mass after reaction has finished (relatively char rate) is and the mass at any time is m, than a fraction reacted (conversion fraction), Î±, is defined as:
v is the mass of volatiles present at any time t, and vf is the total mass of volatiles evolved during the reaction.
From equations 1, 2, and 3, the following equation can be written:
For the determination of the kinetic parameters (E, A, n), in literature can be find several methods. These methods can be classified into three categories: integral, differential and special methods .
The devolatilization dynamics of biomass pyrolysis are frequently expressed as a first order decomposition process (White, Catallo et al. 2011). Assuming a first order reaction, equation (5) can be written:
Dynamic thermogravimetry is often carried out at constant heating rate :
When the natural logarithm of equation (5) is taken and the resulting equation is rearranged, one obtains the traditional and often applied differential method :
By using experimental values for Î± and as a function of temperature, a plot of versus should ideally give straight line with a slope of (-E/R), with an intercept oflnA, Figure 2.
Figure 2Application of Arrhenius equation - Arrhenius plot
Integrating equation (5):
On the right side of the equation (9) temperature containing integral has no exact solution . For solution of temperature containing integral, in the literature, several expansions and semi-empirical approximations have been suggested. For an example:approximation of Kravelen, Broido and Viliams asymptotic expansion and Doyle`s approximation.
Table 1 gives a survey of kinetic data for different biomass species where a single step reaction model have been used.
Table 1 A survey of kinetic data for biomass pyrolysis
log A [logs-1]
Akita and Kase
TGA and DTA,
N2 and vacuum
TGA , 200-460oC, 3oC/min,vacuum
Lewellen et al. 
Fairbridge et al. 
Rogers et al. 
Whatman filter paper
Cooley et al. 
TGA, 200-600oC, He, *1oC/min,**2oC/min,
Várhegyi et al. 
TGA, 200-400oC, Ar
Grønli et al. 
Balci et al. 
dynamic 5-100 oC/min
TGA, static 1.2106 oC/min
TGA, dynamic 20oC/min
TGA, dynamic 10oC/min
TGA, dynamic 120oC/min
Williams et al. 
Várhegyi et al. 
RajeswaraRao et al. 
S. Singh et al.
Refuse derived fuel
25 oC/min, N2
Biomass (Pine wood waste)
The activation energy (Table 1),ranges from83130 to 343260 kJ/mol for cellulose, from 125 to 260 kJ/mol for hemicellulose, from 37 to 125 kJ/mol for lignin and from 60 to 230240kJ/mol for wood. The reason for this diversity may be attributed to different experimental conditions, e.g.: sample size, measurement temperature, heating rate and atmosphere . Also the reason for this differences, can be caused by different extraction procedures and to lack of accuracy caused by the approximations used in the different computational methods .
Special methods are generally based on particular couples of experimental data, e.g. data from different heating rates, or data evaluated from graphical plots . The special methods give worst accuracy.
Today, with developed software and computers, there is no need for simplifying approximations, if Î± and is known (results from TGA experiments), the kinetic parameters (E, A, n) can be calculated by non - linear curve fitting of equation (5), .
On an indicative basis, in thermogravimetry (slow heating rates for a sufficiently small mass of the sample, so that a kinetic control is established), primary degradation of biomass starts at about 225 °C, fast rates are attained at about 300 °C and the process is practically terminated at 425-475 °C [23, 36, 37]. The decrease in weight ofbiomass is caused by the release of volatiles, or devolatilization, during thermal decomposition of biomass. Thermogravimetric curves, measured for dynamic or isothermal conditions, are source of information for the formulation of global or semi-global mechanisms of thermal decomposition of biomass. This information can include the effects of reaction parameters and sample properties .
Biomass fuels and residues contain a wide variety of pyrolyzing species. Even the same chemical species may have differing reactivity if their pyrolysis is influenced by other species in their vicinity .
According to the results of wide number researches presented in literature[12, 23, 24, 38, 39], the concept of biomass pyrolysis may be represented as a combination of the individual decomposition of hemicellulose, cellulose, and lignin. The major constituents of cellulose are polymer glucosan, hemicellulose are polysaccharide producing biomass sugars, and lignin constituents are multi-ring organic compound .
For heating rates at sufficiently slow or moderate temperatures, several zones appear in the weight loss curves, which can be associated with component dynamics.
The lower temperature peaks represents the decomposition of hemicellulose (decompose at 225-300°C), higher peaks representing the decomposition of cellulose (decompose at 325-375°C. Lignin decomposition occurs throughout the whole temperature range, but the main area of weight loss occurs at higher temperatures (decompose at 250-500 °C), which means that lignin is mainly responsible for the flat tailing section .
As the heating rate is increased, given that the range of the degradation temperatures of components is relatively narrow, the different peaks in the degradation rate tend to merge and the characteristic process temperatures tend to become progressively higher .
The mineral matter and trace elements (such as Ca, K, Na, Mg, and Fe), catalyze biomass thermal decomposition reactions. Minerals in biomass, particularly the alkali metals and alkali earth metals serve as catalysts for char formation . The catalytic activity of a mineral matter and trace elements depends on: chemical form, amount, inclusion size. Asfurthermore, if temperatures are sufficiently high, significant degradation rates are simultaneously attained by all the components .
The term ''pseudo - component'' is more appropriate as it is impossible to avoid overlap between the different components in the measured weight loss curves . In other words, although for each zone a main contributor can be identified as hemicellulose, cellulose and lignin, respectively, the simultaneous participation of the other components cannot be avoided with an extent that depends on the biomass characteristics and the severity of the conversion conditions .
Finally the mass loss or mass loss rate can be described by models assuming biomass as the sum of pseudo - components. The pseudo - components as components of the biomass decompose in similar way and in similar temperature ranges. This idea was firstly introduce by Orfao et al. (1999), who defined three pseudo - components for describing the primary thermal decomposition of pine and eucalyptus woods. Later Manyà et al. (2003),Mészáros et al. (2004b)and Diaz (2006) showed satisfactory results when several partial reactions for corresponding pseudo - components were assumed in the decomposition of a wide variety of biomass materials.
A difficulty in kinetic analysis also exists in separating the effects of chemistry and transport phenomena . One of the key points, in relation to involve of heat and masstransfer processes in kinetic analysis, is the sample size. Sample size during pyrolysis cause spatial gradients of temperature (a process taking place under non-negligible effects of internal heat transfer) or significant differences of temperatures between the sample and the controlling thermocouple, especially when these are not in close contact (non-negligible external heat transfer resistance) .
For an example, small samples give less char coal then larger samples. The sample size influence on the char yield is explained by the residential time of the volatiles, which react with the char layer when flowing out the particle to form char coal .The long residence time of the vapor phase inside large particles explains the formation of higher char coal yield. It takes longer time for the volatiles to leave a large particle than a small . Lower char coal yield for small sample masses (powder and single particle samples), could be explained by the bigger surface area that interacts with the pyrolysis medium. Formed volatile products leave the sample without undergoing secondary cracking reactions . Also, for small sizes, above certain temperatures, the time for total conversion become shorter than times needed for the reactor (TGA) to attain the final temperature . In the case of larger particles, secondary cracking reactions could be dominant, leading to additional char and tar formation.
When coupled with the description of transport phenomena, chemical kinetics should be able to predict:
2. product distribution, as the operating conditions are varied .
Pyrolysis kinetics, coupled with the description of transport phenomena, produce advanced computational tools for the design and optimization of chemical reactors applied for thermochemical conversion of wood and biomass .
The numerous pyrolysis models for description of the primary decomposition process are based either on a single-step global reaction models or on several or onmultiple-step models (several competitive parallel reactions).
A simplified description of primary decomposition processes, usually adopted for isothermal conditions or fast heating rates, is based on a single-step global reaction process. In this case, weight loss curves are often associated with additional measurements concerning the yields of the three product classes, in order to evaluate the related formation rates .
In the multi-component reaction mechanisms each reaction takes into account the dynamics of pseudo-components in the measured curves of weight loss. Devolatilizationreactions are essentially considered, with only a very few exceptions where both devolatilization and charring are included .
The kinetic models make use of an Arrhenius dependence on temperature, thus introducing the parameters activation energy and pre-exponential factor, and a linear or power law dependence on the component mass fraction, which may lead to additional parameters (the exponents) .
All of them constitute derivations of a summative model for pyrolysis, firstly proposed by Shafizadeh and McGinnis (1971) and still widely accepted today.
The single-step global models
The single-step global models (global decomposition), is used to describe primary and secondary solid degradation by means of experimentally measured rates of weight loss. The dependence of product yields on reaction conditions cannot be predicted, as a constant ratio between volatiles and char is assumed.The global decomposition is used to predict the overall rate of devolatilization (volatiles release) from the biomass sample (i.e., mass loss). This mechanism does not separately predict the production of condensible and gas from volatile products. The corresponding experimental studies have been mostly carried out with small particles, employing thermogravimetric systems. Single-step global models have provided reasonable agreement with experimentally observed kinetic behaviour .
Cellulose is the most widely studied substance in the field of wood and biomass pyrolysis. Temperature variations are associated with three main chemical path- ways, as postulated by the outstanding contributions in cellulose pyrolysis given by the groups led by Broido and Shafizadeh.
The first kinetic model with predictive capabilities that captured some of the complexity of cellulose pyrolysis was developed by Broido and Nelson (1975). Broido and Nelson (1975) described how heat pretreatments at 230-275 oC caused cellulose char yields to vary from 13% (no heat pretreatment) to over 27% . The results of primary thermal decomposition of cellulose are used to rationalize the competitive reaction model displayed in Figure 3.
Figure 3Scheme proposed by Broido and Nelson (1975) for cellulose pyrolysis
In Figure 3 cellulose substrate reacts at elevated temperatures and decomposes by two competitive mechanisms, producing either volatiles (without char), with rate constant k1 or solid intermediates (char) and low molecular weight volatiles, with rate constant k2.
The temperature dependence of each of the rate constants is approximated by the Arrhenius equation:
Shafizadeh and his co-workers [41, 42], in order to take into account the reactivity of the condensable fraction, and the corresponding formation of char and gas by secondary reactions, they proposed a parallel reactions mechanism for the primary and secondary decomposition that allows to predict the evolution of each main product fraction, Figure 4. This typology of models is known as "Broido-Shafizadeh models", and is based on a distributive approach of the process .
Figure 4 Pyrolitic mechanism proposed by Shafizadeh and Chin (1977)
Shafizadeh and his co-workers undertook a kinetic study of cellulose pyrolysis in vacuum by batchwise heating of 250 mg samples of cellulose at temperatures ranging between 259 and 407 oC.At low temperatures, an initial process, corresponding to a reduction in the degree of polymerization and the formation of the so-called 'anhydrocellulose' or 'active cellulose', is postulated . Product distribution from cellulose pyrolysis indicates that both char and gas yields decrease as the reaction temperature is increased, whereas, in primary wood pyrolysis, both liquid and gas yields continuously increase at the expense of char . They emphasize the role of vapor-solid interactions in the formation of char. From Figure 4, secondary vapor-solid interactions are the main source of char formed during cellulose pyrolysis. The residence time of the volatiles in the cellulose during the pyrolysis reaction largely influences the extent of char formation . Pyrolysis of levoglucosan is known to give some residual char, and it has even been suggested that char formation is not a primary step but is a result of repolymerization of volatile material .
Nevertheless, their model clearly indicates charto be a primary pyrolysis product resulting from the decomposition of the solid phase alone .
For years, Broido's experimental findings have not been reproduced or confirmed by any subsequent researchers (see [11, 43-45].
Varhegy and his co-workers (see executed experiments and analyzed the results by modern kinetic techniques to examine the validity of the Broido-Shafizadeh model. The thermogravimetric analyses of Avicel cellulose were done, involving prolonged thermal pretreatments of small samples (0.5-3 mg). The weight loss curves were simulated by modern numerical techniques using the Broido-Safizadeh and other related models . The solid residue yields were significantly lower than
those estimated by Broido and his colleagues. Varhegyi et al. (1994) attribute this difference to the very large samples employed in Broido's work, which enhanced the pyrolyticvapor-solid interactions that lead to char formation. Results showedno evidence to support the inclusion of the initiation step present in the Broido-Shafizadeh model (step i in Figure 4), but they did strongly confirm the role of parallel reactions in the decomposition chemistry.
Most of the work done in this field has been reviewed by Antal and Várhegyi (1995). Antal and Várhegyi (1995), analysed pyrolysis of samples of pure, ash free cellulose (i.e., Avicel PH-105, Whatman CF-11, Millipore ash-free filter pulp, and Whatman #42) at low tomoderate heating rates. Conclusion of analysis is that the pyrolysis of a small sample of many different cellulosic substrates can be adequately described by an irreversible, single-step endothermic reaction that follows a first order rate law at both low and high heating rates.
However, single-step global models is limited by the assumption of a fixed mass ratio between pyrolysis products (i.e., volatiles and chars), which prevents the forecasting of product yields based on process conditions [1, 40].
Multi-component devolatilization mechanisms
As well as for cellulose, wide interest in the primary pyrolysis of whole biomass has appeared in the literature (the pyrolysis of hemicelluloses and lignin). Várhegyi et al. (1989s, and 1989b) performed severalthermogravimetric experiments using: Avicel cellulose, 4-methyl-Pglucurono-D-xylan (hemicellulose) and sugar cane bagasse, in the presence and absence of catalysts (inorganic salts). The three major DTG peaks were observed during the experiments resulted from decomposition of cellulose, hemicellulose, and lignin (main constituents of lignocellulosic materials). Thermogravimetricanalysis showed a distinct DTG peak resulting from the decomposition of cellulose, than a lower DTG peak at lower temperature rangeresulting from hemicellulose pyrolysis, and an attenuated shoulder that can be attributed to lignin decomposition. Várhegyi et al. (1989b) showed that the mineral matter present in the biomass samples can highly increase the overlap of the partial peaks in DTG curves. Sometimes the first peaks merge into one very broad peak .
Várhegyi et al. (1989a, 1989b and 2004) showed that pretreatments have influence on pyrolysis behaviour oflignocellulose materials. Thermal pretreatment destroys the hemicellulose component of the lignocellulose material but doesn't enhance the char yield.Várhegyi, Grønli et al. 2004 evidenced the ability of pretreatments to separate merged peaks, to displace reaction zones toward higher temperatures, decrease the char yield and increase peak reaction rates [46, 47]. The water washing, as one of pretreatments type, is preferred because it results in less hydrolysis and solubilization of the holocellulose. Also the acid washes appeared to decrease the measured activation energy of cellulose pyrolysis .
As it is said, generally, from the thermogravimetric analysis can be seen that temperature domains of moisture evolution and hemicellulose, cellulose and lignin decomposition more or less overlap each other. Considering this and also the results from experiments with biomass different pretreatments, it can be concluded that general biomass pyrolysis behaves as a superposition of the independent kinetics of the primary components (hemicellulose, cellulose, and lignin).
The inability to predict the kinetic behavior of biomass under different process conditions has encouraged researchers for developing complex multi-component models.
It assumes that the true reaction system is too complex to be characterized in any fundamental way, so the reaction is described in terms of pseudo species, which are themselves complex materials or mixtures . Absolute concentration is not important, as all species are characterized in terms of the fraction of their initial or final value .
The basic building block for all reactions is a pseudocomponent reaction :
wherex is the fraction of the initial material unreacted, f(x) is a mathematical function of the unreacted initial material, yiis the ith product of the reaction, and . The simplest case is that of a pseudo-first-order reaction, for which f(x) )x. Other more complex functions will be discussed later. The yivalues represent, for example, a partitioning into gaseous, liquid, and solid products. The pseudocomponents reactions can be presents as :
wherej represents the jth component of x, , yijis the ith product of reaction component j, ,and .
One of first researches who introduce this idea was Orfao et al. (1999). They noted that thermal decomposition of xylan and lignin could not be modelled with acceptable errors by means of simple reactions (minimum deviations were 15% and 10%, respectively). Orfao et al. (1999)defined three pseudocomponents for describing the primary thermal decomposition of pine and eucalyptus woods and pine bark. The pyrolysis of lignocellulosic materials was successfully modelled by a kinetic scheme consisting of three independent first-order reactions of three pseudo-components. The first and the second pseudo-components correspond to the fractions of hemicellulose and cellulose which are reactive at low temperatures and the third includes lignin and the remaining fractions of the carbohydrates . Reasonable agreement was obtained between the activation energies calculated for the other pseudo-components and reported values .
Later,Manyà et al. (2003) the thermal decompositions of sugarcane bagasse and waste-wood samples studied using thermogravimetric analysis. First, an irreversible first order reaction model was assumed for each pseudocomponent, but results showed that the model simulated curves do not fit well to the experimental data. Manyà et al. (2003) with kinetic study presented that pyrolysis of lignin is better described by a third-order reaction rate law. The reformulation of the ligninkinetic model, and its subsequent implementation in the summative model (for the thirdpseudocomponent), has allowed one to reach a good agreement between simulated andexperimental data . Later, Mészáros et al. (2004b) and Diaz (2006) showed satisfactory results when several partial reactions for corresponding pseudocomponents were assumed in the decomposition of a wide variety of biomass materials.
The goal of the kinetic evaluation is to obtain better, more informative results from the experiments. In the attempt to better identify the zones associated with the devolatilization of the biomass components and their overlapped kinetics, different T(t) heating programs have been employed . Mészáros et al. (2004b) increased the information content of the experiments by involving successive nonisothermal steps (stepwise heating programs) into their study. The wider range of the experimental conditions reveals more of the chemical inhomogeneities of the biomass components . Linear and stepwise heating programs were employed to increase the amount of information in the series of experiments . Employing non isothermal experiments, not only indetification of pseudo-components or zones were possible to made (hemicellulose, cellulose and lignin), but also, the contribution of extractives or more than one reaction stage in the decomposition of components, especially hemicellulose and lignin, could be also taken into pyrolysis kinetic analysis account.
Experimental measurements of the pyrolyticbehavior of biomass have been the focus of extraordinary interest in the research community, but practical problems associated with these measurements have often been overlooked. The most important errors are connected to problems of temperature measurements and to the self-cooling/self-heating of samples due to heat demand by the chemical reaction . A consequence of these limitations is that the single step activation energy measured at high heating rates is almost always lower than its true value . Another consequence is that weight loss is reported at temperatures much higher than it actually occurs .
All mentioned, are possible reasons for gross disagreements in the literature concerning the kinetics of pyrolysis.
For example, Antal and Várhegyi (1995) concluded that the pyrolysisof a small sample of pure cellulose is characterized by an endothermic reaction governed by a first-order rate law with a high activation energy (ca. 238 kJ/mol). Almostimmediately after the paper was published, these conclusions were contradicted by the findings of Milosavljevic and Suuberg (1995), claim that the cellulose thermal degradation can be well described by a two-stage mechanism: the first at a low-temperature range with high activation energy (218 kJ/mol) and the second at a high-temperature range with reduced activation energy (140-155 kJ/mol).Antalet. al (1998)measured the rates of pyrolysis of the same cellulose employed by Milosavljevicand Suuberg (1995) in Antal`s laboratory equipment. Also, the kinetics of other cellulose samples was studied to learn if different pure celluloses evidence markedly different pyrolysis behavior.The mass used for samples by Milosavljevic and Suuberg(30 mg) causes diffusion effects and, subsequently, an increase in the residence time for the vapor fraction, which promotes secondary reactions . Also, the thermal lag (between the thermocouple lecture and the real temperature of the sample) accentuates the compensation effect . This phenomenon causes an erratic estimation for the kinetic parameters . If heat transfer effects cannot be neglected, then the kinetic model may not be adequate for describing the behavior of the process involved, and must be combined with heat transfer equations . It is difficult to combine a realistic modeling of the heat transfer phenomena with complex chemical kinetic models . An alternative way is the empirical assessment of systematic errors. To specify the serious trouble that supposes the experimental error, Grønli et al. (1999) coordinated the realization of a round-robin kinetic study for the cellulose pyrolysis (Avicel PH-105) in eight European laboratories .
Results confirmed the theories of Antalet. al (1998), but also alerted the scientific community about the convenience of carrying out this experiment (under standard conditions) in order to be able to quantify their own experimental errors [47, 53].
The least understood aspect of pyrolysis is the interaction of the nascent, hot pyrolysis vapors (volatiles, tar) with the decomposing solid, which vapors must traverse during their escape to the environment. Secondary decomposition is interactions among primary volatiles and the solid residue. Tars produced during the decomposition of the virgin biomass can decompose further. At high temperatures and given sufficiently long residence times, secondary reactions of primary tar vapors also become active . Secondary reactions may occur in the pores of the particles, while undergoing primary degradation, homogeneously in the vapor phase and heterogeneously over the char surfaces and the extra-particle surfaces, and include processes such as cracking, partial oxidation, re-polymerization and condensation .
The most cited mechanism for description of secondary pyrolysis reaction simply consists of two competing reactions reported by Antal (1983),Figure 4. The first reaction produces more permanent gases by cracking the reactive volatile matter to smaller, less reactive species . The second reaction produces refractorycondensable materials, which may be a tar or some combinationof water-soluble organic compounds .
Figure 4A global mechanism for the secondary reactions of vapor-phase
tarry species as proposed by Antal (1983) 
Di Blasi (2008), explained that the existence of the second reaction is inferred from the gas yield data, which display an asymptotic behavior that is strongly dependent on temperature . Higher temperatures result in dramatic increases in the asymptotic yields of all the light permanent gases produced. The temperature-dependent asymptotes require the existence of the second reaction in order to explain the disappearance of carbon atoms in the gas phase when the gas phase temperature is reduced. The thermal stability of tars for temperatures below 500oC is a key issue in the fast pyrolysis processes aimed at bio-oil production [23, 54]. The kinetics of secondary tar reactions is important in biomass gasification. The amount of tar produced and its composition depend on the type of gasifier and the process conditions. In principle, producer gas with a low tar content can be obtained if a high-temperature zone can be created where the volatile products of pyrolysis are forced to reside sufficiently long to undergo secondary gasification .
Numerous factors influence the final yield and quality of charcoal from biomass, including the substrate composition, the heating rate, the final (peak) temperature of pyrolysis, the pressure and flow of the surrounding gaseous environment, the presence of catalysts (both natural and exotic), and the possibility of autocatalysis by volatile pyrolysis products .
MacKay and Roberts (1982), give comprehensive pyrolysis analysis of many lignocellulosic (biomass) species. Samples werepyrolyzed under argon at 15Â°C/min to 500Â°C. Variation in the total mass and carbon yields among precursors was found to be due to variation in composition, i.e. distribution of the main organic components (lignin, holocellulose and extractives) . Also, they revealed a range in charcoal yields from 25.9 to 35.2% which they were able to relate to the lignin, holocellulose, and extractive content of the feedstock . Biomass species with high lignin content were found to offer higher charcoal yields. The char yield from lignin (53%) was found to be three times higher that of cellulose (18%) because of the higher initial carbon content in lignin (63 vs 44%) as well as the higher carbon yield (76 vs 40%) .
Often it is assumed that the char yield can be increased by reducing heating rate. Unfortunately, this assumption is not true. Thermogravimetric studies reported by Várhegyiet a1. (1988),(see [44, 56]), revealed no influence on the charcoal yield from bagasse when the heating rate was decreased from 80 to 10 oC/min. A decrease in heating rate from 2 to 0.5 oC/min resulted in no significant change in the charcoal yield at 541 oC, however, a small increase in yield was detected between 20 oC/min and 2 oC/min .
Both the yield and the quality of the charcoal product are strongly influenced by the peak temperature of the pyrolysis process . As the peak temperature increases above 200 oC, the solid pyrolytic residue changes from "toasted" wood to "torrefied" wood to "pyrochar" to conventional char . The solids residence time also affects the yield and quality of the charcoal product, but usually this time is selected after the peak temperature is determined . When the biomass substrate is heated to somewhat higher temperatures, but not exceeding about 350 oC, a "pyrochar" is produced . The term "pyrochar" is employed herein broadly to include charcoal as well as partially carbonized woody material that has been pyrolyzed at least sufficiently to destroy its fibrous character . This material is formed in about 50% yield and has lost the fibrous character of the biomass feedstock and has a volatile matter content of 35% or more . Pyle (1976), uses the onset of exothermicity, which he claims will occur at about 50% weight loss, to define the peak temperature for theformation of "pyrochar" . The pyrolysis reactions become exothermic when the percentage volatile matter contained in the pyrochar reaches 35-45% .
Improved yields of charcoal are obtained when pyrolysis is conducted at elevated pressures. Mok et al. (1992) found that an increase in pressure from 0.1 to 1.0 MPa (at constant purge gas velocity) increased the char yield to 41%.
Additional improvements can be realized when the pyrolysis vapors remain in contact with solids at the peak temperature until pyrolysis is complete. Várhegyi et al. (1988), a small sample of Avicel cellulose hermetically sealed in a crucible with a pinhole in the top of the crucible. The decomposition occurs in the presence of the vapors, and these vapors spend a longer time at a higher partial pressure in the hot zone above the sample . Results from this analysis described an increase in the char yield from 5 to 19 wt % when the pyrolysis was conducted in "covered" (with pinhole) versus open crucibles. Any restriction of the ability of the pyrolyticvapors to escape from the vicinity of the char product increases the ï¬xed-carbon yield. This perspicuous finding revealed the role of secondary reactions involving the interactions of pyrolytic volatile matter with the solid sample in the formation of char .The similar experiment were performed by Wang et al. (2011), used open versus closed crucibles to accentuate the impact of vapour phase conditions on char yields. The closure of the crucible substantially enhances the char yield. These observations corroborate earlier work and reveal the importance of secondary reactions involving vapour phase species in the formation of charcoal. Conditions that improve or prolong the contact of vapor-phase pyrolysis species with the solid serve to enhance the char yield .
In addition to the differences in the temperature and residence times of the vapor phase, the presence of reactive species, such as steam (also from primary decomposition), and char may have an important impact on the tar decomposition rates .For instance, freshly formed can cause the heterogeneous conversion of about 14% of the primary tar product. The catalytic role exerted by charcoal on tar conversion is also recognized in biomass gasification. To exploit this feature, new reactor design schemes have been proposed.
Concerning kinetic modelling, in former case case, the rate of tar cracking is generally described by a global reaction, with a rate linearly dependent on the mass concentration of the vapour phase tar () and the usual Arrhenius dependence on temperature. In alternative, the cracking rate is linearly dependent only on the reactive fraction of the primary tar ().
The concerning kinetic modeling, Di Blasi and Russo (1994) presented an approach describing the kinetics according to a competitive reaction scheme . Secondary decomposition was included as tar cracking to light hydrocarbons, though tar polymerization to char was ignored .
Babu and Chaurasia (2003)developed a mathematical model to describe the pyrolysis of a single solid particle of biomass. Mathematical modelcouples the heat transfer equation with the chemical kinetics equations. The pyrolysis rate has been simulated by a kinetic scheme involving three reactions (primary and secondary): two parallel reactions and a third for the secondary interactions between the volatile and gaseous products and the char . The model developed can be utilized to predict the temperature and concentration profiles for different types of biomass for a wide range of particle dimensions and temperatures .
Branca and Di Blasi performed several studies (see [23, 59-61]) on the low-temperature devolatilization of pyrolysis liquids produced from different fuels and variable heating conditions confirm the importance of polymerization versus cracking reactions. For fast pyrolysis liquids and a devolatilization process carried out under the conditions of thermal analysis, secondary char retains about the half of the initial carbon content of the liquid . Moreover, high yields are obtained especially for liquids produced from cellulose, indicating the important role played by sugars and not only by the products of lignin decomposition .
Semi-global kinetic mechanisms of solid degradation explicitly include primary and secondary reactions, for which pyrolysis products are lumped into three groups (tar, gas and char) . Semi-global mechanisms allow the effects of reaction selectivity and volatile residence time to be predicted on product distribution. Extensive applications have allowed the dynamics of the decomposition process (temperature, species concentration, pressure and velocity yelds) and global characteristics (conversion time, product yields) to be predicted for widely variable heating conditions (slow/fast pyrolysis and chemical/heat transfer control) . However, model validation is still limited because reliable input data and/or extensive measurements to be used for comparison purposes are lacking .
The majority of the generalized models correspond to extremely complicated schemes and suffer from a high number of undefined parameters, lack in the interrelation with the biomass composition, or they have not been validated for a sufficiently wide range of experiment conditions.
Distributed activation energy model (DAEM)
The complex composition of biomass materials, the conventional linearization techniques of the nonisothermal kinetics are not suitable for the evaluation of the TGA experiments. The biomass fuels and rawmaterials contain a wide variety of pyrolyzing species. Even the same chemical species may have a different reactivity if its pyrolysis is influenced by other species in its vicinity . The assumption of a distribution in thereactivity of the decomposing species frequently helps the kinetic evaluation of the pyrolysis of complex organic samples[38, 63].
The chemical complexity of both the biomass and the related pyrolysis products motivate the introduction of kinetic models based on kinetic laws different from those presented above.
The distributed activation energy model (DAEM) is the best way to represent mathematically the physical and chemical inhomogeneity of a substance .
Distributed activation energy models have been used for biomass pyrolysis kinetics since 1985,when Avni et al.  applied a DAEM for the formation of volatiles from lignin. Later this type of research was extended to a wider range of lignocellulose materials. Saidi et al. (see), employed DAEM-based kinetic models in establishing an actual combustion model of a burning cigarette. A three-dimensional model for a puffing cigarette was constructed using the principles of the conservation of mass and momentum. To do this, an average temperature-time history of a burning cigarette was derived using existing experimental data for the temperature distribution in a cigarette. Várhegyi et al.wasstudied decomposition of two tobacco blends by thermogravimetry-mass spectrometry (TGA-MS) at slow heating programs under well-defined conditions. The kinetic evaluation was based on a distributed activation energy model (DAEM). The complexity of the studied materials required the use of more than one DAEM reaction . The resulting models describe well the experimental data and are suitable for predicting experiments at higher heating rates. Várhegyi et al. [66, 67], Becidan et al. , Trninic et al. ,based DAEM kinetic studies on the simultaneous evaluation of experiments with linear and stepwise temperature programs. The model parameters obtained in this way allowed accurate prediction outside of the domain of the experimental conditions of the given kinetic evaluations . The determination of the unknown model parameters and the verification of the model were based on the leastsquares evaluation of series of experiments . This approach led to favorable results and allowed predictions outside the experimental conditions of the experiments used in the parameter determination [64, 67].
The distributed reactivity is usually approximated by a Gaussian distribution of the activation energy due to the favorable experience with this type of modeling on similarly complex materials . According to this model, the sample is regarded as a sum of Mpseudocomponents, where M is usually between 2 and 4 . Here pseudocomponent is the totality of those decomposing species which can be described by the same reaction kinetic parameters in the given model . The reactivity differences are described by different activation energy values. On a molecular level, each species in pseudocomponentj is assumed to undergo a first-order decay . The corresponding rate constant k and mean lifetime Ï„ are supposed to depend on the temperature by an Arrhenius formula:
If Î±j(t,E) is the solution of the corresponding first-order kinetic equation at a given E and T(t) with conditions Î±j(0,E) = 0 and Î±j(âˆž,E) = 1:
The density function of the species differing by E within a given pseudocomponent is denoted by Dj(E).Dj(E) is approximated by a Gaussian distribution with mean E0,j and width-parameter (variation)Ïƒj. The overall reacted fraction of the jthpseudocomponent, Î±j(t), is obtained by integration:
The normalized sample mass, m, and its derivative are the linear combinations of Î±j(t) and dÎ±j/dt, respectively:
where a weight factorcj is equal to the amount of volatiles formed from a unit mass of pseudocomponentj.
The pyrolysis of lignocellulosic materials is the result of complex interactions among many
physical and chemical processes.
Understanding and modelling of pyrolysis process is central base for understanding behaviour of lignocellulose materials not only during pyrolysis process but alsoduring gasification and combustion processes.
The complexity of pyrolysis phenomena are primarily due to:
1. complexities of lignocellulose materials composition, which include the presence of long and complex organic molecules and their characteristic decomposition reactions, the presence of moisture, and the type of lignocellulose materials which is considered.
2. heating rate effects, classifying pyrolysis into slow and fast regimes.
3. residence time effects, which result in auto-catalysis of secondary reactions.
The data handling and the criteria used to determine the 'best' kinetics parameters that reproduce the experimental results is a crucial target of the kinetic modeling.
A variety of mathematical modelling techniques available for the analysis of pyrolysis were discussed in this paper.
The main troubles found on analyzing the pyrolysis kinetics are:
Much of the disagreement in the literature concerning to the data obtained by TGA work has been due to the influence of varied systematic error associated with the characteristics of the originating equipment and the experimental procedure [2, 52].
2. At the time being the understanding of the effects of pretreatments remains largely qualitative. The various industrial applications (e.g. the production of charcoal, activated carbon and liquid fuels) require a clear understanding of the variations in the original composition and the effects of the pretreatments, since the yields and the composition, structure, and other properties of the products are highly influenced by the properties of the feedstock .
3. The traditional employment of linear heating programs in thermogravimetric experiments has not guaranteed an in-depth characterization of the various pyrolytic processes occurring in the material . The chemical heterogeneities of the biomass components and the physical heterogeneity of the plant materials increase the overlap between the independent kinetics . Continued effort on increasing the information content of the experiments, by coupling some other heating approaches or/and analysis of the evolved products, is still necessary .
4. Mathematical models of biomass pyrolysis coupling chemical processes (including secondary reactions) and transport phenomena are rarely available in the literature. There is still a lack of fundamental research to improve design methods and modeling involving primary and secondary reactions, leading to optimal reactor design and process . Being pyrolysis the initial step in all the thermal conversions of biomass