Kinetic Study Of Wood Chips Decomposition By Tga

3 downloads 0 Views 258KB Size Report
May 29, 2009 - suitable raw material that can be converted to chemicals, fuels, gases, energy or heat [2]. In large scale ... decomposition process kinetics determination is one of the key problems. Pyrolysis as one ... Le-We-4, 178p.pdf. 178–1 ...
Slovak Society of Chemical Engineering Institute of Chemical and Environmental Engineering Slovak University of Technology in Bratislava

PROCEEDINGS 36

th

International Conference of Slovak Society of Chemical Engineering

Hotel Hutník Tatranské Matliare, Slovakia May 25 – 29, 2009

Editor: J. Markoš

ISBN 978-80-227-3072-3

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

Kinetic study of wood chips decomposition by TGA Lukáš Gašparovič, Zuzana Koreňová, Ľudovít Jelemenský1 Institute of Chemical and Environmental Engineering,, Faculty of Chemical and Food Technology, Slovak University of Technology, 812 37, Bratislava, Radlinského 13, Slovakia

Abstract Pyrolysis of the wood chips mixture and main wood compounds such as hemicellulose cellulose and lignin was investigated by thermogravimetry. The investigation was carried out in inert nitrogen atmosphere with temperatures ranging from 20°C to 900°C for four heating rates 2K/min, 5K/min, 10K/min and 15K/min. Hemicellulose, cellulose and lignin were used as main compounds of biomass. TGA and DTG temperature dependencies were evaluated. Decomposition processes proceed in three main stages: water evaporation, active and passive pyrolysis. The decomposition of hemicellulose and cellulose takes place in the temperature range of 200°C – 380°C and 250°C - 380°C, while lignin decomposition seems to be ranging from 180°C up to 900°C. Isoconversional method was used to determine kinetic parameters such as activation energy and pre-exponential factor mainly in the stage of active pyrolysis and partially in the passive stage. It was found that, at the end of the decomposition process, the value of activation energy decreases. Reaction order does not have a significant influence on the process because of the high value of the pre-exponential factor. Obtained kinetic parameters were used to calculate simulated decompositions at different heating rates. Experimental data compared with the simulation were in good accordance at all heating rates. From the pyrolysis of hemicellulose, cellulose and lignin is clear that the decomposition process of wood is dependent on the composition and concentration of main compounds.

1. Introduction Continuously increasing amount of the world’s consumed chemicals, fuels and energy results in rapid depletion of fossil fuels reserves which are used as a raw material in the production of the above mentioned products. This is the reason why science is recently oriented at a more effective utilization of renewable energy sources, like biomass. It is possible to produce the same products from biomass like from fossil fuels. One of the greatest advantages of renewable sources is that they don’t contribute to the amount of CO2 in the atmosphere, because the amount of CO2 released to the atmosphere by combustion is compensated by the amount of CO2 consumed by plants during photosynthesis. It is said that renewable sources are CO2 neutral [1]. Another advantage is that, in the time of crisis, the biomass seems to be a suitable raw material that can be converted to chemicals, fuels, gases, energy or heat [2]. In large scale processes of biomass conversion to valuable products via thermal treatment, the decomposition process kinetics determination is one of the key problems. Pyrolysis as one of the thermochemical processes (without the presence of oxygen) involves many partial processes which can’t be exactly described. To investigate the pyrolysis decomposition kinetics, thermogravimetric analysis is often used [3]. Thermogravimetry is used to describe 1

Corresponding Author: [email protected]

178–1

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

the kinetic process of non-isothermal pyrolysis, and information concerning this issue are included in the review by Colomba Di Blasi [4]. Various types of biomasses from residual wood, through agricultural waste (straw, kernel) to sewage sludge were tested. Because of three main components of biomass: cellulose, hemicellulose and lignin [4-7], the biomass decomposition proceeds through three main decomposition regions (moisture evolution not included) which belong to these components [7, 8]. Because hemicellulose is only a blend of several sugar monomers, it is not possible to extract pure substance. It is common to use xylan instead of hemicellulose [9, 10], which is always the sugar monomer with the highest representation. Some publications dealing with this problem have demonstrated that decomposition regions of hemicellulose and cellulose are in ranges of around 220-315°C and 315-400°C, respectively with maximum weight losses at 268°C and 355°C [11]. Lignin decomposition is in the range from 180°C up to 900°C with an unclear maximum weight loss. The solid residue which is left after pyrolysis is dependent on the origin of the used biomass. The process of thermal decomposition can be influenced by operating parameters e.g. temperature, heating rate and pressure, or by biomass properties e.g. size and shape of particles, composition, moisture content etc. The heating rate affects the behavior of the conversion curve. The increase of heating rates results in slight changes of the conversion curve position towards higher temperatures. This shift of conversion curve also affects the differential thermogravimetric curve [12]. Size and shape of particles have an influence on the heat transfer into the particle, on the rate of phase changes, or gas escape from particle. To ensure a good heat transfer to the particle and minimal temperature gradient, the sample should be closely in contact with the heat transfer surface [4, 13]. The reaction rate constant in the form of Arrhenius equation is often used to describe the kinetic pyrolysis decomposition. Parameters like activation energy, pre-exponential factor and reaction order have been determined from experimental data for various feed stocks [7, 14-16] using isoconversional or optimization methods. The goals of this work are to determine kinetic parameters during pyrolysis decomposition of wood chips, which are a mixture of unknown various kinds of wood, by an isoconversional method. Cellulose, hemicellulose and lignin have the same characteristic range of thermal decomposition in state of pure substance or in a mixture (that means in wood). Therefore the effort was to make up the overall thermal decomposition of wood chips by thermal decomposition of individual wood compounds by a superposition of them.

2. Materials and Methods 2.1 .Preparation of samples Samples of wood for thermogravimetric analysis were taken from the residual processing wood. Due to the above mentioned particle size effect of the thermogravimeter, wood chips with diameter of about 50 mm were milled into smaller chips with the diameter of about 1-2 mm. Milled chips were stored in a small glass bottle to prevent the contact with air moisture. Samples of hemicellulose, cellulose and lignin were used as well.

178–2

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

2.2. Elementary analysis Elementary composition and high heating value of used wood chips are presented in Table 1. Channiwala et al (2002) proposed a formula for high heating value calculation [17]: HHV[MJ/kg]=0.3491C+1.1783H+0.1005S-0.1034O-0.0151N-0.0211A

(1)

Table 1: Characterization of wood sample composition Sample Ash(%) wood chips

5,93

a

Composition wt % b C(%) H(%) O(%) FC(%)

VM(%)

48,23

69,29

6,09

39,75

24,78

HHV [MJ/kg] 19,78

FC – fixed carbon, VM-volatile matter, HHV-high heating value a the rest after combustion b calculated from difference

2.3. Measurement methods Pyrolysis of wood chips and the above mentioned main compounds of wood was performed using the thermogravimetric equipment STA 409 PC Luxx (Netzsch). Vertical TG/DSC holder was used. To achieve pyrolysis conditions, nitrogen atmosphere was used. Nitrogen was used like carrying and protective gas that protects the micro balance against possible pollutants. The volume flow of nitrogen was set up to 60 ml/min N2 for the carrying gas, which swept away product gases, and 10ml/min N2 for the protective gas, which protected balance against pollutants. Due to the buoyancy effect, correction measurements were carried out. Thermogravimetric measurements were performed at four heating rates: 2 K/min, 5 K/min, 10 K/min and 15 K/min. Al2O2 crucibles were used. The furnace space had to be inertized for 30 minutes to rid of all remants of oxygen. Weight of the sample was between 11-14 mg. This amount is enough to create a good contact area between the crucible and the sample. The temperature decomposition ranged from laboratory temperature of around 20°C to 900°C. At the end of the heating process, isothermal mode was set to 10 min to ensure that the process is over. During the heating, mass of the sample was recorded. To measure the actual sample and furnace temperatures, thermocouples type S were used. Typical thermogravimetric behavior of the decomposition is shown in Figure 2

3. Kinetic parameters estimation Mass loss data from the thermogravimetric analysis can be recalculated into conversion which is defined as follows: m −m α= 0 (2) m0 − m∞ where m0 is the initial mass of sample, m is the actual sample mass and m∞ is the residual mass after pyrolysis. The conversion represents the amount of sample which was decomposed.

178–3

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

To estimate kinetic parameters, the isoconversional method was used. This method has been widely utilized by several authors when describing decomposition of biomass [16, 18], waste petroleum refinery sludge [12], etc. The rate of decomposition is a function of temperature and conversion: dα = f (T , α ) dt

(3)

It is possible to rewrite the right hand side of (3) by two functions, where the first is dependent on temperature and the second one is a function of conversion. dα = k ( T ) g (α ) dt

(4)

The temperature dependent function k (T ) in (4) is usually expressed by Arrhenius equation:

 −E  k (T ) = A exp    RT 

(5)

where A is the pre-exponential factor, E is the activation energy and R is the gas constant. There are various possibilities how to express conversion function g (α ) . In this equation, the form used was as follows: g (α ) = (1 − α )

n

(6)

where n is the reaction order. Combining (4), (5) and(6), the decomposition kinetic equation is obtained: dα n  −E  = A exp   (1 − α ) dt  RT 

(7)

Actual temperature under non-isothermal conditions can be expressed as:

T = T0 + β t

(8)

where β is the heating rate and t is time. The isoconversional method is used for the description of more complex processes where lots of chemical reactions are running simultaneously, however, their mechanisms are not exactly known. According to the isoconversional method, the kinetic parameters as the pre-exponential factor and activation energy are not constant for the whole decomposition process, but they are dependent on conversion [19]. Results of the mathematic description of a process by the isoconversional method are Aα and Eα dependencies as functions of conversion which is related to the mechanism of reactions. For the isoconversional method, equation (7) is rewritten to:

 − E (α )  dα n = A (α ) exp   (1 − α ) dt  RT 

178–4

(9)

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

Equation (9) is solved for the given conversion:

 − Eα  n  dα   (1 − α )   = Aα exp   dt α  RTα 

(10)

 dα  where   is the decomposition rate, Tα is the temperature in Kelvin, Aα is the apparent  dt α pre-exponential factor and Eα is the apparent activation energy at the given value of conversion α . The logarithm form of equation (10) is shown by equation (11) which has a pattern of linear regression: Eα  dα  ln   = ln Aα + n ln (1 − α ) − RTα  dt α

(11)

The isoconversional calculation principle of Aα and Eα is described in Figure 1 [20]: For predefined values of conversion, from experimental data at four heating rates, decomposition rate (in logarithm form) and temperature were selected. It follows that for specified values of conversion there are four decomposition rates, where each belongs to one temperature. According to equation (11), these points should lie on the same line. The first two terms on the right hand side of equation (11) can be expressed as an intercept of a straight E line. The term − α represents a slope. Because the value of the intercept is dependent on RTα both the pre-exponential factor and the reaction order, for the determination of the preexponential factor, reaction order must be predetermined. The influence of this parameter on the decomposition process is mentioned in the results and discussion part. j =1 TGA data at heating rate β j j = j +1

defining of conversion α1 , α 2 ....α m take from exp erimental data dα d α1 d α 2 1 1 1 , ,.... m and , ,.... dt dt dt T1 T2 Tm i =1

 dα  to obtain a dependence ln  like   dt αi , β j a function of

1 for α i , β j , j = 1, 2,3, 4 Tαi ,β j

i = i +1 estimation of Aαi and Eα i from equation (10 ) at given reaction order n E = f (α ) , An = g (α ) for given reaction order n

Figure 1: Block diagram for the estimation of

178–5

Aα and Eα

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

4. Results and discussion 4.1. Pyrolysis of wood chips The changes of conversion and decomposition rate during wood chips pyrolysis are shown in Figure 2. The moisture and adsorbed water content released up to 160°C was excluded. The mass of water was subtracted from the overall mass of the sample, and the initial mass of the sample was obtained m0 .The main decomposition region (active pyrolysis) is in range from approximately 450 K to 770 K. Above 770 K there is only a small drop of mass (passive pyrolysis). Most of hemicellulose and cellulose in wood is decomposed in the first stage of active pyrolysis. In this stage, the decomposition rate reached two maximums, at 562 K and at 618 K. These maximums are in literature assigned to the maximum of hemicellulose and cellulose decomposition. Lignin is decomposed in both stages. These behaviors are also presented in Figure 8. The stage of passive pyrolysis can be allocated to lignin decomposition in which thermal degradation passes through the whole temperature range with a very low decomposition rate.

0,0008

1,0

(dα/dt)C, max

0,8 0,0006

0,0004 0,4

0,0002

water evaporation

α[%]

0,6

-1

dα/dt [s ]

(dα/dt)H, max

0,2

0,0 0,0000 400

600

800

1000

1200

Temperature [K]

Figure 2: Typical □ TGA and ○ DTG dependence during the pyrolysis

4.2. Effect of heating rate It is known that the heating rate affects both location of the TGA curve and maximum decomposition rate. Effect of the heating rate is shown in Figure 3 and Figure 4. As can be seen in Figure 3, there is a shift in conversion lines caused by various heating rates. At higher heating rates, individual conversions are reached at higher temperatures. The maximums of the decomposition rate are also slightly shifted towards higher temperatures. This fact can be a consequence of heat and mass transfer limitations. It means that temperature in the furnace space can be a little higher as the temperature of particle and the rate of devolatilization is higher than the release of volatilities. Because of the heat transfer limitation, temperature

178–6

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

gradients may exist in the particle. Temperature in the core of a particle can be a bit lower than temperature on the surface, and different devolatilization processes or releasing rates can occur. This is the reason, why it is necessary to have a small particle, homogenous sample and heat transfer surface between the sample and the crucible as large as possible.

100

80 100 90

60

Shifting of conversions

80

60

α[%]

α[%]

70

40

50 40

Heating rate raising

30 20 10

20

0 500

550

600

650

700

Temperature [K]

0 400

600

800

1000

1200

Temperature [K]

Figure 3: Pyrolysis TGA dependencies of wood chips at various heating rates: □2 K/min, ○5K/min, ◊ 10 K/min, 15 K/min.

0,0030

0,003

0,0025 Heating Rate raising

0,002

-1

dα/dt [s ]

0,0020

-1

dα/dt [s ]

0,001

0,0015 0,000 450

500

0,0010

550

600

650

700

750

Temperature [K]

0,0005

0,0000 400

600

800

1000

1200

Temperature [K]

Figure 4: Pyrolysis DTG dependencies of wood chips at various heating rates: □2 K/min, ○5K/min, ◊10 K/min, 15 K/min.

4.3. Isoconversional method

178–7

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

According to the above mentioned determination of kinetic parameters, the experimental data were processed in order to obtain parameters like the activation energy and pre-exponential factor. From the set of data at different heating rates, the isoconversional lines for predefined conversion were calculated, which can be seen in Figure 5. Predefined conversions were in the range from 1% to 94% with the changing step of 0,5%. Because at higher temperatures no significant changes in conversion occur, the isoconversional lines were not very precise.

β4 β3 β2 β1

Figure 5: Isoconversional lines for predefined conversion increasing from right to left

The slope of each isoconversional line was calculated, and consequently, the activation energy was obtained. The pre-exponential factor can be obtained from the intercept of the isoconversional line. However it depends on the reaction order (see equation (11)). Dependencies of Aα and Eα as functions of conversion are shown in Table 2. Table 2: Calculated kinetic parameters for wood chips Conversion (%) Eα (kJ/mol)

5

10

20

30

40

50

60

70

80

90

204,95

199,54

209,60

214,19

215,94

204,07

200,94

196,49

144,72

131,56

n=0 3,18E+16 3,77E+15 9,74E+15 7,65E+15 4,39E+15 2,66E+14 9,16E+13 1,86E+13 1,09E+08 1,16E+05

Aα (s-1)

n=1 3,35E+16 4,18E+15 1,22E+16 1,09E+16 7,31E+15 5,31E+14 2,29E+14 6,19E+13 5,43E+08 1,16E+06 n=2 3,53E+16 4,65E+15 1,52E+16 1,56E+16 1,22E+16 1,06E+15 5,73E+14 2,06E+14 2,72E+09 1,16E+07

178–8

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

The pre-exponential factor seems not to be varying greatly with the change of the reaction order. Due to this fact it does not depend if there is the zeroth, first or second reaction order. The term ln (1 − α ) in equation (11) is always incomparably lower than all intercepts

ln Aα + n ln (1 − α ) and it has no significant influence on the value of the pre-exponential factor. The activation energy does not change during the process, however, nearing the end of the process when the conversion is above 80%, it starts to decrease. Up to conversion values of 70% – 80% similar type of bonds C-H, C-O and C-C were destructed. At higher conversions than 80%, new bond destructions might occur. Now, when kinetic parameters for each conversion are known, simulation data can be calculated via equation (10) with respect to the reaction order. In order to solve this equation, ordinary differential equation solver ode15s which is a standard component of MATLAB was used. Like input parameters, initial condition (initial conversion is zero) and time interval ( t = 0, tend ) were used. The solver performed a non-uniform time discretization and solved the ordinary differential equation implicitly. Kinetic parameters Aα and Eα were fitted by polynomial regression and backward calculated for actual conversion in differential equation. Comparison of TGA and DTG curves of the experimental and simulation data are shown in Figure 6 and Figure 7, respectively.

100

80

α[%]

60

40

20

0 400

600

800

1000

1200

Temperature [K]

Figure 6: Pyrolysis TGA dependencies of wood chips at various heating rates: experiment: □2 K/min, ○5K/min, ◊10 K/min, 15 K/min and simulation: solid 2 K/min, dashed 5 K/min, dot 10 K/min, dash-dot 15 K/min.

178–9

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

0,0030

0,0025

-1

dα/dt [s ]

0,0020

0,0015

0,0010

0,0005

0,0000 400

600

800

1000

1200

Temperature [K]

Figure 7: Pyrolysis DTG dependencies of wood chips at various heating rates: experiment: □2 K/min, ○5K/min, ◊10 K/min, 15 K/min and simulation: solid 2 K/min, dashed 5 K/min, dot 10 K/min, dash-dot 15 K/min.

It can be seen from both figures that experimental and simulation data are in good accordance. Values of the decomposition rates at the transition area and maximums in Figure 7 are maintained as well. This satisfactory accordance makes the isoconversional method an acceptable method for the pyrolysis kinetic estimation.

4.4. Pyrolysis of wood chips vs. cellulose and lignin All wood types, either softwood or hardwood, consist mainly from three main compounds: hemicellulose, cellulose and lignin. It is assumed, in case of no interactions between the compounds, that the decomposition behavior of wood is a linear combination of the main compounds and can be described by mass weighted average:

 dm   dm   dm   dm  = wC     + wH   + wL    dt  wood  dt C  dt  H  dt  L

(12)

 dm  where   and w are the mass loss in time and the weight fractions. Pyrolysis  dt  decompositions of hemicellulose, cellulose and lignin are shown in Figure 8. From this behaviors it is possible to see that hemicellulose starts to decompose first at the maximum decomposition rate of about 560 K. Maximums of lignin and cellulose lie close to each other, at 614 K and 620 K. Decomposition of hemicellulose and cellulose is quite sharp while decomposition of lignin goes through a wide temperature range. These three dependencies were used to describe wood decomposition. For optimization the minimization function (13) was used.

  dm    dm   dm   dm    MF = ∑   −  wC    + wH   + wL     dt  H  dt  L   T0   dt  wood   dt C Tend

178–10

2

(13)

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

1,2

-dm/dt [mg/min]

1,0

0,8

0,6

0,4

0,2

0,0 500

600

700

800

Temperature [K]

Figure 8: Pyrolysis decomposition of hemicellulose (dots line), lignin (full line) and cellulose (dashed line) at the heating rate of 5 K/min

Three weight fractions were changing to reach a minimum of minimization function. The comparison between decomposition of wood chips and weighed decompositions sum of hemicellulose, cellulose and lignin is shown in Figure 9.

0,6

0,5

-dm/dt [mg/min]

0,4

0,3

0,2

0,1

0,0 500

600

700

800

Temperature [K]

Figure 9: Pyrolysis decomposition of wood chips (full line) and weighed decomposition sum of hemicellulose, cellulose and lignin (dashed line) at the heating rate of 5 K/min

From Figure 9 it is clear that there is only a small difference between the presented behaviors, but overall processes are in good accordance. If the assumption of no interaction is valid, then the pyrolysis decomposition behavior of wood biomass is dependent on the composition and concentration of the main compounds.

178–11

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

5. Conclusion In this work, thermal decomposition of wood chips, hemicellulose, cellulose and lignin were studied in inert nitrogen atmosphere. From TGA and DTG it was found that decomposition proceeds through three stages of weight loss. The first stage is allocated to water evaporation in the range to 160°C. In the second stage, the mass loss is a consequence of thermal decomposition of wood compounds from 160°C to 430°C. The last, third stage represents a slow and long decomposition of lignin up to 900°C. Kinetic parameters of wood chips decomposition such as the apparent activation energy and pre-exponential factor were obtained by the isoconversional method. Activation energy during the main decomposition stage is in the range 190 – 217 kJ/mol depending on the conversion. At higher conversions, the activation energy started decreasing rapidly as the whole process slowed down. The future intentions are to obtain kinetic data by the least square optimization method and to compare them with the data from this work. If a good accordance is achieved, kinetic parameters could be used to describe pyrolysis process in a small laboratory reactor. Comparison of wood chips, hemicellulose, cellulose and lignin DTG represents a potential for the description of the decomposition process of any biomass via decomposition characteristic of the main compounds of biomass with respect to the composition and concentration.

Acknowledgement This work was supported by the Slovak Research and Development Agency under the contract LPP-0230-07. The presented contribution was created as a part of project SK00023 financed by the Norwegian Financial Mechanism, Financial Mechanism of EEA and the State Budget of the Slovak Republic (www.eeagrants.com).

List of symbols DTG

Differential thermogravimetry

MF

Minimization function

TGA

Thermogravimetry analysis

A

Pre-exponential factor

C

subscript for cellulose

E

Activation energy

H

subscript for hemicellulose

k

Reaction rate constant

L

subscript for lignin

m

Weight

[s-1] [J mol-1] [s-1]

[kg]

178–12

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

m0

Initial weight

[kg]

m∞

Final weight

[kg]

n

Reaction order

[1]

R

Gas constant

[J mol-1 K-1]

t

Time

[s]

T

Temperature

[K]

T0

Initial temperature

[K]

w

Weight fraction

α

Conversion

[1]

β

Heating rate

[K min-1]

178–13

36th International Conference of SSCHE May 25–29, 2009, Tatransk´e Matliare, Slovakia

Le-We-4, 178p.pdf

References 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12.

13.

14. 15. 16. 17. 18. 19. 20.

Lv, P., et al., A study on the economic efficiency of hydrogen production from biomass residues in China. Renewable Energy, 2008. 33(8): p. 1874-1879. McKendry, P., Energy production from biomass (part 1): overview of biomass. Bioresource Technology, 2002. 83(1): p. 37-46. Brown, M.E., Introduction to Thermal Analysis. second ed. Techniques and Applications, ed. J. Simon. Vol. 1. 2001. Di Blasi, C., Modeling chemical and physical processes of wood and biomass pyrolysis. Progress in Energy and Combustion Science, 2008. 34(1): p. 47-90. Mohan, D., C.U. Pittman, and P.H. Steele, Pyrolysis of Wood/Biomass for Bio-oil: A Critical review. Energy & fuels ISSN 0887-0624 CODEN ENFUEM 2006. Orfăo, J.J.M., F.J.A. Antunes, and J.L. Figueiredo, Pyrolysis kinetics of lignocellulosic materials--three independent reactions model. Fuel, 1999. 78(3): p. 349-358. Kastanaki, E., et al., Thermogravimetric studies of the behavior of lignite-biomass blends during devolatilization. Fuel Processing Technology, 2002. 77-78: p. 159-166. Gronli, M.G., G. Varhegyi, and C. Di Blasi, Thermogravimetric Analysis and Devolatilization Kinetics of Wood. Industrial & Engineering Chemistry Research, 2002. 41(17): p. 4201-4208. Couhert, C., J.-M. Commandre, and S. Salvador, Is it possible to predict gas yields of any biomass after rapid pyrolysis at high temperature from its composition in cellulose, hemicellulose and lignin? Fuel, 2009. 88(3): p. 408-417. Couhert, C., J.-M. Commandré, and S. Salvador, Failure of the component additivity rule to predict gas yields of biomass in flash pyrolysis at 950 °C. Biomass and Bioenergy, 2009. 33(2): p. 316-326. Yang, H., et al., Characteristics of hemicellulose, cellulose and lignin pyrolysis. Fuel, 2007. 86(12-13): p. 1781-1788. Choudhury, D., et al., Non-isothermal thermogravimetric pyrolysis kinetics of waste petroleum refinery sludge by isoconversional approach. Journal of Thermal Analysis and Calorimetry, 2007. 89(3): p. 965970. Stenseng, M., A. Jensen, and K. Dam-Johansen, Investigation of biomass pyrolysis by thermogravimetric analysis and differential scanning calorimetry. Journal of Analytical and Applied Pyrolysis, 2001. 58-59: p. 765-780. Kumar, A., et al., Thermogravimetric characterization of corn stover as gasification and pyrolysis feedstock. Biomass and Bioenergy, 2008. 32(5): p. 460-467. Munir, S., et al., Thermal analysis and devolatilization kinetics of cotton stalk, sugar cane bagasse and shea meal under nitrogen and air atmospheres. Bioresource Technology, 2009. 100(3): p. 1413-1418. Park, Y.-H., et al., Pyrolysis characteristics and kinetics of oak trees using thermogravimetric analyzer and micro-tubing reactor. Bioresource Technology, 2009. 100(1): p. 400-405. Channiwala, S.A. and P.P. Parikh, A unified correlation for estimating HHV of solid, liquid and gaseous fuels. Fuel, 2002. 81(8): p. 1051-1063. Biagini, E., A. Fantei, and L. Tognotti, Effect of the heating rate on the devolatilization of biomass residues. Thermochimica Acta, 2008. 472(1-2): p. 55-63. Šimon, P., Isoconversional methods. Journal of Thermal Analysis and Calorimetry, 2004. 76(1): p. 123132. Koreňová, Z., Optimization of thermal decomposition conditions of waste tires, Disertation Thesis, in Department of Chemical and Biochemical Engineering. 2008, Slovak University of Technology: Bratislava. p. 117.

178–14