Research ArticleKinetic Study and Thermal DecompositionBehavior of Lignite Coal
Mehran Heydari Moshfiqur Rahman and Rajender Gupta
Department of Chemical and Materials Engineering University of Alberta Edmonton AB Canada T6G 2V4
Correspondence should be addressed to Rajender Gupta rajenderguptaualbertaca
Received 5 December 2014 Accepted 16 April 2015
Academic Editor Deepak Kunzru
Copyright copy 2015 Mehran Heydari et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited
A thermogravimetric analyzer was employed to investigate the thermal behavior and extract the kinetic parameters of Canadianlignite coal The pyrolysis experiments were conducted in temperatures ranging from 298K to 1173 K under inert atmosphereutilizing six different heating rates of 1 6 9 12 15 and 18Kminminus1 respectively There are different techniques for analyzingthe kinetics of solid-state reactions that can generally be classified into two categories model-fitting and model-free methodsHistorically model-fitting methods are broadly used in solid-state kinetics and show an excellent fit to the experimental data butproduce uncertain kinetic parameters especially for nonisothermal conditions In this work different model-free techniques suchas the Kissinger method and the isoconversional methods of Ozawa Kissinger-Akahira-Sunose and Friedman are employed andcompared in order to analyze nonisothermal kinetic data and investigate thermal behavior of a lignite coal Experimental resultsshowed that the activation energy values obtained by the isoconversional methods were in good agreement but Friedman methodwas considered to be the best among the model-free methods to evaluate kinetic parameters for solid-state reactions These resultscan provide useful information to predict kinetic model of coal pyrolysis and optimization of the process conditions
1 Introduction
During the past few decades petroleum has been the mainsource of liquid fuels On one hand petroleum reservesare declining on the other hand coal reserve is the mostabundant fossil fuel known in the world [1] Coal is asource of fuel for more than half of the worldrsquos power plantsfor electricity generation Coal and coal-derived fuels havebeen used in residential commercial and industrial appli-cations The amount of coal deposits estimated worldwideis approximately ten times larger than that for the othercarbonaceous resources The availability of coal resourceswas a main contributor to the economic growth of manycountries such as the US China India and Australia [2]Coal appears to hold the most promise of all the possiblealternatives for short-term development to meet the nationalrequirements of energy Coal and coal products play a majorrole in fulfilling the energy demands of our society [3]Direct liquefaction indirect liquefaction and gasificationare examples of existing processes for coal conversion into
energy products Therefore coal is of significant industrialand economic importance both as an energy source and asan industrial feedstock [4] In large-scale processes of coalconversion to valuable products through thermal treatmentdetermination of the kinetic parameters in the decompositionstage is one of the key problems Many unresolved problemsface a designer of coal combustors and gasifiers includingthe complex physical and chemical behavior of coal and theuncertainty regarding the kinetics of the chemical reactionsduring thermal decomposition [5] The design of processesfor pulverized coal requires that the various stages occur-ring during the thermal decomposition be understood inorder to provide optimum operating conditions This greateremphasis on more efficient utilization of coal combinedwith its chemical complexity raises the need for a betterunderstanding of the pyrolysis process Pyrolysis is themethod for obtaining liquid from coal by rejecting carbonand thereby increasing the hydrogen-to-carbon ratio of rawcoal Pyrolysis takes place as coal is treated at elevatedtemperatures in the absence of oxygen and during this
Hindawi Publishing CorporationInternational Journal of Chemical EngineeringVolume 2015 Article ID 481739 9 pageshttpdxdoiorg1011552015481739
2 International Journal of Chemical Engineering
pyrolysis a series of reactions occurs This is done in theabsence of oxygen so that undesirable combustion reactionscannot take place [6] The main products of pyrolysis are gastar and char The products of pyrolysis that is the amountand the composition of volatile material depend primarilyon the coal type reaction temperature pressure heating rateresidence time and particle size
The heating rate can affect the behavior of the conversioncurve The increase of heating rates results in slight changesin the conversion curve and maximum decomposition ratetowards higher temperatures [7] Pyrolysis kinetics of coal isimportant because it is the initial step ofmain coal conversionprocesses such as liquefaction gasification and combustionin which coal particles undergo major physical and chemicaltransformations For a better understanding of pyrolysisseveral researchers investigated thermal decomposition ofcoal by thermogravimetric analysis (TGA) As coal has beenused as a fuel since the beginning of industrial developmentit has been among the earliest materials to be subjected tothermal analysis
To investigate the kinetics of the decomposition processTGA is often used In TGA the weight change of the sampleis observed as it is heated usually at a constant heatingrate under a controlled atmosphere such as nitrogen air orother gases The record of weight loss with respect to thetime or temperature is termed a thermogravimetric (TG)thermogramWhen the rate of weight loss (the first derivativewith respect to time) is recorded as a function of time or tem-perature it is called a differential thermogravimetric (DTG)thermogram The DTG has been used to study the kineticsof thermal decomposition reactions of a variety of solidsincluding coal Much of this work is based on the assumptionthat thermal decomposition is describable by an overall first-order reaction and follows the Arrhenius-type equation Thekinetics of the thermal behavior of a material can be deter-mined by the application of a kinetic model to the rates ofmass degradationThemain advantages of TGA for the studyof coal pyrolysis are simplicity in implementation and uti-lization and good repeatability [8] A large number of studieshave reported on thermogravimetric and differential thermalanalysis in an attempt to explain kinetics of thermal decom-position of coal and to obtain qualitative information on coalpyrolysis The literature reviews on these subjects regardingthermal analysis are present from Howard [9] Lawson [10]Anthony and Howard [11] Hathi [12] and Khawam [13] Themain differences in the thermobalances used for the studies ofHonda (1915) Guichard (1926) Vallet (1932) Rigollet (1934)Dubois (1935) Longechambon (1936) and Jouin (1947) werementioned by Hathi [12] and Khawam [13] These ther-mobalances recorded mass versus temperature or time VanHeerdan and Huntjens studied the rates of decomposition ofDutch coals on a thermobalance that recorded mass loss datacontinuously over the temperature range 200ndash550∘CAmath-ematical equation in the form of the Arrhenius equation wasconsidered to explain the rate of coal decomposition Theyconcluded that the decomposition process is first order withregard to the fraction of unreacted coal They observed thatinitial devolatilization is fast removal of moisture and oxidesof carbon the middle devolatilization is slow and contains
the removal of the major volatile matter from coal and thefinal devolatilization is a slow process for liberating the gasfrom residuals [12] Scaccia et al investigated the pyrolysisof low-rank Sulcis coal by thermogravimetric techniques(TGDTG) in the temperature range ambient to 1000∘C atthree different heating rates From thermogravimetric resultsit was established that coal pyrolysis involved three mainstages water evaporation devolatilization of thermally labileand more stable volatiles and char formation [14]
The knowledge of kinetic parameters is essential formodeling the reactor and optimization of the process con-ditions There are various methods for evaluating kineticparameters from nonisothermal thermogravimetric analysis(TGA) and the most common of them can be classifiedinto two major types model-fitting and model-free [14ndash17]In the model-fitting method different models are fit to theexperimental data and the model giving the best statisticalfit is selected as the model from which the activation energy(119864119886
) and frequency factor (119860) are evaluated Historicallymodel-fitting methods were broadly used because of theirability to directly calculate the kinetic parameters from thethermogravimetric analysis results However these methodshave several drawbacks the most important one being theirinability to uniquely select the appropriate reaction model[13] Furthermore comparing the results of these models inthe literature can be difficult especially for nonisothermaldata since a wide range of kinetic parameters have been deter-mined for the coal pyrolysis process This led to the declineof these methods in favor of isoconversional (model-free)methods which can estimate the activation energy withoutevaluating the reaction model [13] The greatest advantagesof this model are its simplicity and avoidance of errorsrelated to selecting specific reaction models Isoconversionalmethod is called model-free method because of its ability todetermine the activation energy for different constant extentsof conversion without considering any particular form ofthe reaction model These methods require several kineticcurves to perform the analysis and thus are sometimes calledmulticurve methods [18] These methods can calculate theactivation energy at different heating rates on the same valueof conversion The terms ldquomodel-freerdquo and ldquoisoconversionalrdquoare sometimes used interchangeably however not all model-free methods are isoconversional For example the Kissingermethod is a model-free method but is not isoconversionalbecause it does not calculate activation energy at differentconstant extents of conversion but instead assumes constantactivation energy [13]
Isoconversional methods are helpful tools for the analysisof solid-state kinetics Theoretically they include many ben-efits and applications However practically they have somedisadvantages especially regarding reproducibility when per-forming a series of runs at different heating rates in whichtheir fluctuation may enhance experimental errors Thus fornonisothermal experiments each run must be conductedunder the same experimental conditions (sample weightpurge gas rate and sample size) so the only variable is theheating rate In order to obtain accurate results with high res-olution curves low ranges of heating rates can be consideredfor the experiments
International Journal of Chemical Engineering 3
Numerous recent studies on the TGA pyrolysis of coal[19ndash21] and coal-biomass blends [22ndash24] are available in theliterature and most of them are based on model-fitting tech-niques There are a few reports relating to thermal decompo-sition behavior of coal based on model-free techniques [14]Moreover most of the previous studies have been performedon coal-biomass blends in order to determine the kinetics ofcopyrolysis of coal and biomass mixtures To the best of ourknowledge there is very little information regarding pyrolysisof coal itself based on model-free methods
The aim of the present work is to study the pyrolysiskinetics of Canadian lignite coal by means of thermogravi-metric analysis (TGA) within the temperature range of 298ndash1173 K at different heating rates under nitrogen atmosphereThe effect of the heating rate on decomposition will also bestudied In this study different model-free methods such asthe Kissinger and the isoconversional methods of OzawaKissinger-Akahira-Sunose and Friedman are employed andcompared in order to analyze nonisothermal kinetic data andinvestigate thermal behavior of a Canadian lignite coal Thekinetic parameters of the coal decomposition processwill alsobe determined These results may provide helpful informa-tion for pyrolysis researchers to predict a kineticmodel of coalpyrolysis and optimization of the process conditions
2 Materials and Methods
21 Sample Preparation It is estimated that approximatelyhalf of the coal resources of the world are low-rank coalsuch as lignite and subbituminous coal [25] Lignite coal isabundant in Canada and plays an important role in energyproduction It was thus chosen as the experimental samplein the present study Canadian lignite coal was obtainedfrom Poplar River Mine located in southern SaskatchewanCanadaThe bulk coal sample was crushed by means of a jawcrusher and ground in a ball mill and blended to homogenizethe coal and reduce the particle size between 106 and 150120583mThe coal sample was received wet with 32 moisture contentandwas dried in vacuum oven at 80∘C for 8 h until a moisturecontent of 12 was achieved The sample was submitted toboth proximate analysis according to the ASTM D7582 byMacro Thermogravimetric Analyzer and ultimate analysisaccording to ASTMD3176 in Elemental Vario MICRO CubeThe results of the proximate and ultimate analysis (CHNS) aswell as higher heating value of the sample used are presentedin Table 1 Higher heating value of coal was also calculatedwith Channiwala and Parikh formula [26]
HHV = 03491C + 11783H + 01005S minus 01034O
minus 00151N minus 00211A (MJkg) (1)
where C H S O N and A are the mass fractions of carbonhydrogen sulfur oxygen nitrogen and ash respectively
22 Experimental Method The TGA experiments were per-formed using a thermogravimetric analyzer TGAndashSDTQ600 at the coal research center of University of AlbertaAbout 10mg of fine coal particle size between 106 and
Table 1 Characteristics of the coal sample
Proximate analysis (wt) Ultimate analysis (wt daf)Moisture 1278 C 4463Volatile matter 4124 H 468Ash 1983 N 066Fixed carbon 2615 S 057
Olowast 4946HHV (MJkg) 1602
daf = dry and ash-free basis lowastObtained by difference
150 120583m was placed in a small Alumina crucible for eachrun and heated from 298K to the maximum temperatureof 1173 K at six different heating rates of 1 6 9 12 15 and18 Kminminus1 respectively under nitrogen atmosphere with aflow rate of 100mLmin During the heating variation of theweight loss and its derivative with respect to the time andtemperature was collected automatically by the instrumentand determined through the TA universal analysis softwareThe experiments were repeated under identical conditions tocheck the reproducibility of the results
23 Kinetic Analysis There are a number of approaches formodelling the complex pyrolysis process The simplest is theempirical model which employs global kinetics where theArrhenius expression is used to correlate the rates of massloss with temperature The pyrolysis process of coal can beexpressed by the following reaction
Coal 119896997888rarr Volatiles + Char (2)
The general expression for the decomposition of a solidsample is
119889119909
119889119905= 119896 (119879) 119891 (119909) (3)
where 119909 is the degree of conversion which represents thedecomposed amount of the sample at time 119905 and is definedin terms of the change in mass of the sample
119909 =(119898119894
minus 119898119905
)
(119898119894
minus 119898119891
)
(4)
where 119898119894
is the initial mass 119898119891
is the final mass and 119898119905
isthe mass at time 119905 of the sample analyzed by TGA 119891(119909) is afunction of 119909 depending on the reaction mechanism 119896(119879) isthe rate constant at temperature 119879 which generally obeys theArrhenius equation
119896 (119879) = 119860 exp(minus119864119886
119877119879) (5)
where119860 is the preexponential factor (minminus1) Equation is theactivation energy (kJmolminus1) 119877 is the universal gas constant(J Kminus1molminus1) and 119879 is the absolute temperature (K)
Substitution of (5) into (3) gives the general expression tocalculate the kinetic parameters
119889119909
119889119905= 119891 (119909)119860 exp(
minus119864119886
119877119879) (6)
4 International Journal of Chemical Engineering
There are various possibilities to express the conversionfunction 119891(119909) for the solid-state reactions Most of theprevious authors used the conversion function as follows
119891 (119909) = (1 minus 119909)119899
(7)
where 119899 is the reaction order here it is considered first orderCombining (6) and (7) the kinetic equation of decomposi-tion is obtained as follows
119889119909
119889119905= 119860 exp(
minus119864119886
119877119879) (1 minus 119909)
119899
(8)
Under nonisothermal conditions inwhich samples are heatedat constant heating rates the actual temperature under thiscondition can be expressed as
119879 = 1198790
+ 120573119905 (9)
where 1198790
is the initial temperature 120573 is the linear heating rate(∘Cmin) and 119879 is the temperature at time 119905 Nonisothermalmethods are usually common in solid-state kinetics becausethey require less experimental data in comparison to isother-mal methodsThe following expression can be considered fornonisothermal experiments
119889119909
119889119879=119889119909
119889119905sdot119889119905
119889119879 (10)
where 119889119909119889119879 is the nonisothermal reaction rate 119889119909119889119905 is theisothermal reaction rate and 119889119879119889119905 is the heating rate (120573)Substituting (8) into (10) gives
119889119909
119889119879=119860
119861exp(minus119864119886
119877119879) (1 minus 119909)
119899
(11)
Equation (11) represents the differential form of the non-isothermal rate law In this study the data fromnonisothermalexperiments are considered to calculate kinetic parametersbased on model-free methods such as Kissinger and the iso-conversional methods of Ozawa Kissinger-Akahira-Sunoseand Friedman and compared in order to analyze and toinvestigate thermal behavior of a Canadian lignite coal
24 Model-Free Methods The kinetic analysis based onmodel-free methods allows the kinetic parameters to beevaluated for different constant extents of conversion withoutevaluating any particular form of the reaction model Thetemperature sensitivity of the reaction rate depends on theextent of conversion to products This is partly a result ofthe heterogeneous nature of solid-state reactions such as coalpyrolysis it also arises somewhat because many solid-statereactions follow complex mechanisms including multipleseries and parallel stages with different activation energiesModel-fitting methods are applied to extract a single setof Arrhenius parameters for an overall process and are notcapable to show this type of complexity in the solid-statereactions Model-free methods are able of addressing theaforementioned drawbacks of themodel-fittingmethodsTheability of model-free methods to show this type of reactioncomplexity is therefore a critical step toward the ability toexplain mechanistic conclusions from kinetic data
241 Kissinger Method According to Kissinger the maxi-mum reaction rate occurs with an increase in the reactiontemperature [27] The degree of conversion at the peaktemperature of the DTG curve is a constant at differentheating rates Kissinger method is a model-free methodbut it is not isoconversional method because it assumesconstant activation energy with the progress of conversionIn Kissinger equation (12) 119879
119898
representing the peak temper-ature is expressed as
ln( 1198611198792119898
) = ln(119860119877119864119886
) minus119864119886
119877119879119898
(12)
Therefore kinetic parameters including activation energy(119864119886
) and preexponential factor (119860) can be obtained from aplot of ln(1198611198792
119898
) versus 1000119879119898
for a series of experimentsat different heating rates
242 Kissinger-Akahira-Sunose (KAS)MethodTheKissinger-Akahira-Sunose (KAS) method was based on the followingequation
ln( 1198611198792) = ln( 119860119877
119864119886
119892 (119909)) minus119864119886
119877119879 (13)
where 119892(119909) is the integral conversion function (reactionmodel) which is reported in the literature [15] For constantconversion a plot of left side of the above equation against1000119879 at different heating rates is a straight line whoseslope and intercept can evaluate the activation energy andpreexponential factor respectively
243 The Flynn-Wall-Ozawa (FWO)Method TheKissinger-Akahira-Sunose (FWO) method is based on the followingequation
ln (119861) = ln(119860119864119886
119877119892 (119909)) minus 5331 minus 1052
119864119886
119877119879 (14)
Thus for a constant conversion a plot of natural logarithmof heating rates ln(119861) versus 1000119879 obtained from thermalcurves recorded at different heating rateswill be a straight linewhose slope (minus1052(119864
119886
119877119879)) will calculate the activationenergy
244 Friedman Method This method is one of the firstisoconversional methods Using (2) and (4) and taking thenatural logarithm of each side the expression proposed byFriedman can be presented as
ln(119889119909119889119905) = ln [119860119891 (119909)] minus
119864119886
119877119879 (15)
The activation energy (119864119886
) is determined from the slope ofthe plot of ln(119889119909119889119905) versus 1000119879 at a constant conversionvalue
International Journal of Chemical Engineering 5
3 Results and Discussion
31Thermal Decomposition Process TheTG andDTG curvesof the pyrolysis of a Canadian lignite coal under nitrogenatmosphere obtained at six different heating rates of 1 6 9 1215 and 18K minminus1 are shown in Figures 1 and 2 respectivelyTheTG curves show the percentagemass loss of a coal sampleover the range of temperature from 298K to 1173 K The rateof mass loss is temperature dependent the higher the tem-perature the larger the mass loss because pyrolysis processproceeds slowly at low temperatures As shown in Figure 1the devolatilization process launches at temperature about450K and proceeds fast with elevating the temperature up to850K and then themass loss of the sample drops slowly to theultimate temperature The DTG curves of sample at differentheating rates are illustrated in Figure 2 The DTG curveexhibits three zones related to moisture evaporation primarydecomposition and secondary decompositionThe first zonerepresents elimination of moisture which occurs below 450K[28] The second region is related to main decompositionstage in the temperature range 450ndash850K for low heatingrate and 925K for high heating rate Major volatile matterat this stage liberated from coal structure that was formedby thermal decomposition some covalent bond such as etherbonds and methylene group which will form gases such ashydrogen carbon monoxide and lighter hydrocarbons [29]This region is themost significant region to examine since themajor weight loss and complicated chemical reaction such asrelease of tar and gaseous products and semicoke formationtake place in this temperature range [30 31] The third zonethat is the second pyrolysis stage where low decompositionrates are observed can be attributed to the further gasificationof the formed char due to high temperature effects On theother hand the coal sample contains high ash and the phasetransitions of the inorganics found in the mineral matterlosses of the molecular water contents of the clay mineralsand decomposition of carbonate minerals may contributeto weight loss of this step There is only a small drop ofmass observed at this stage The TGA data are normalizedfrom 0 to 1 before analysis The temperature at which thederivative of mass loss starts to increase is selected as thezero conversion point and the temperature at which themassderivative returned to the base line is chosen as end point Itis known that the heating rate affects all TGA curves and themaximum decomposition rate When heating rate increasesthe temperature of the maximum decomposition rate ofthe coal shifted toward higher temperature Figure 3 showsconversion curves versus temperature at different heatingrates The curves showed typical sigmoid shape of kineticcurves With increasing the heating rate conversion valuesreachedhigher temperatures because at the same temperatureand time a high heating rate has a short decomposition timeand the temperature required for the sample to reach thesame conversion will be higher The heat transfer limitation(thermal lag) exists between furnace and sample temperatureIt means that temperature in the particle can be a little lowerthan furnace temperature and gradient of temperature mayexist in the coal sample so in order to reduce the thermallag the coal sample should be ground to the fine particle to
300
40
50
60
70
80
90
100
Wei
ght (
)
400 500 600 700 800 900 1000 1100
Temperature (K)1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 1 Thermal behavior of Poplar coal at different heating ratesunder N
2
atmosphere
300
5
4
3
2
1
0
400 500 600 700 800 900 1000 1100
Temperature (K)
DTG
(m
in)
1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 2 DTG curves of Poplar coal at different heating rates underN2
atmosphere
increase the surface area of particle and consequently increasethe heat transfer effect between the sample surface and thecrucible as large as possible
32 Kinetic Analysis The results of TGDTG experimentaldata of coal pyrolysis obtained under nonisothermal con-dition under nitrogen atmosphere were used for kineticanalysis Different model-free methods such as Kissinger andthe isoconversional methods of Ozawa Kissinger-Akahira-Sunose and Friedman are employed in order to obtainparameters like the activation energy and preexponential
6 International Journal of Chemical Engineering
45000
02
04
06
08
10
550 650 750 850500 600 700 800 900
Temperature (K)
Con
vers
ion
(x)
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 3 Conversion curves for different heating rates for pyrolysisof lignite coal under N
2
atmosphere
factor In theKissingermethod the degree of conversion at thepeak temperature (119879
119898
) is a constant under different heatingrates The kinetic parameters using Kissinger method werefound by linear regression line which is shown in Figure 4The activation energy and preexponential factor extractedfrom the slope and intercept are 281 kJmolminus1 and 261 times1017minminus1 respectively The activation energy and preexpo-nential factor were calculated as a function of conversionby using isoconversional methods of KAS FWO and Fried-man methods The isoconversional plots of these methodsare shown in Figures 5ndash7 respectively Different range ofconversion from 005 to 09 is considered for calculating thekinetic parameters based on isoconversional method Theactivation energies from the slope and preexponential factorsfrom the intercept of three different isoconversional methodswere obtained and listed in Table 2 It can be observed fromTable 2 that the values of activation energies are not similar atdifferent constant extents of conversion because most solid-state reactions are not simple one-stepmechanism and followa complex multistep reaction The thermogravimetric dataanalysis by isoconversional technique may reveal complexityof the solid-state reactions such as coal pyrolysis [14] Itmeansthat in the pyrolysis process of coal the activation energy isa function of conversion Figure 8 shows the dependence ofthe activation energy on extent of conversion The activationenergy rises from about 130 kJmolminus1 at low conversion tonearly 350 kJmolminus1 at 75 conversion and it subsequentlydrops to about 300 kJmolminus1 near the end of reaction Theinitial activation energy valuewas lowdue to cleavage of someweak bonds and elimination of volatile components fromthe coal matrix because at the beginning of the process allthe strong bonds are not cleaved Therefore more activationenergy is required to decompose these stable moleculesWith the progress of pyrolysis process the value of activation
138 140 142 144 146 148
Kissinger
R2 = 09858
y = minus33803x + 36586
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
ln(120573
1000Tm (Kminus1)
T2 m
)
Figure 4 Kissinger plot of lignite coal pyrolysis at different heatingrates
12 14 16 18 20 22
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus95
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
minus140
ln(120573T
2)
1000T (Kminus1)
Figure 5 KAS plots of lignite coal pyrolysis at different values ofconversion
energy increased up to conversion of 75 with breaking ofsome strong covalent linkages For higher conversion valuesabove 75 the activation energy gradually decreases Thereason arises from the fact that during the decompositionprocess at high temperature with high conversion whenmostof the stable bonds are broken less stablemolecules which areeasier to break are present so less energy barrier is requiredfor decomposition at this step and the value of activationenergy decreases with progress of conversionThe arithmeticmeans of the activation energy calculated by KAS FWO andFriedmanmethod are 282 275 and 283 kJmolminus1 respectively
International Journal of Chemical Engineering 7
12 14 16 18 20 22 24
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus05
00
05
10
15
20
25
30
35
1000T (Kminus1)
ln 120573
Figure 6 FWO plots of lignite coal pyrolysis at different values ofconversion
12 14 16 18 20 22
minus20
minus25
minus30
minus35
minus40
minus45
minus50
minus55
minus60
minus65
minus70
minus75
ln(dxdt)
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
1000T (Kminus1)
Figure 7 Friedman plots of lignite coal pyrolysis at different valuesof conversion
which are close to average activation energy obtained fromthe Kissinger method (28103 kJmolminus1) The results obtainedwith KAS and Friedman methods are very close and in goodagreement [32]The kinetic data obtained for pyrolysis of coalare found to agree closely with some of the literature dataHowever the differences observed in the literature data canbe attributed to the fact that the pyrolysis characteristics ofcoal highly depend on the properties of the coal which in turndiffers based on origin of the coal [28 30 31]
The KAS and FWOmethods were originally derived withconstant activation energies so the errors associated withkinetic measurements from methods should be dependent
Activ
atio
n en
ergy
Conversion (x)
KissingerKAS
FWOFriedman
00100
150
200
250
300
350
400
02 04 06 08 10
Figure 8 The activation energy as a function of conversion usingmodel-free isoconversional technique
on the variation of the activation energy with respect toconversion This error does not appear in the Friedmanmethod [33 34] Another advantage that can be attributed toFriedman method is that the activation energies obtained bythe Friedmanmethod are independent of the range of heatingrates which can decrease the systematic error in evaluatingthe activation energy values Thus Friedman method canbe considered to be the best among the four model-freemethods in order to evaluate kinetic parameters for solid-state reactions [33 34] The kinetic parameters obtainedin this study can be useful for pyrolysis and gasificationresearchers to predict kinetic model of coal pyrolysis andoptimization of the process conditions
4 Conclusion
In this study the pyrolysis kinetics of a Canadian lignite coalwas carried out bymeans of thermogravimetric analysis (TG)in the temperature range of 298ndash1173K at six different heatingrates of 1 6 9 12 15 and 18 Kminminus1 under nitrogen atmo-sphere It was found that the main pyrolysis process occurredin the temperature range 450ndash850K In this work kineticstudy and thermal behavior of lignite coal were presentedwhere Arrhenius parameters were determined and comparedthrough four different methods of Kissinger Ozawa KASand Friedman The activation energy is calculated as afunction of conversion by using these methods and is foundto be similar Among these methods Friedman methodwas considered to be the best in order to evaluate kineticparameters for solid-state reactions such as coal pyrolysisMethods such as FWO and KAS are restricted to the use of alinear variation of the temperature and positive heating rateMoreover they are generated based onmathematical approx-imation which can enhance systematic error The advantage
8 International Journal of Chemical Engineering
Table 2 Calculated kinetic parameters for a Canadian lignite coal by three different isoconversional methods
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
pyrolysis a series of reactions occurs This is done in theabsence of oxygen so that undesirable combustion reactionscannot take place [6] The main products of pyrolysis are gastar and char The products of pyrolysis that is the amountand the composition of volatile material depend primarilyon the coal type reaction temperature pressure heating rateresidence time and particle size
The heating rate can affect the behavior of the conversioncurve The increase of heating rates results in slight changesin the conversion curve and maximum decomposition ratetowards higher temperatures [7] Pyrolysis kinetics of coal isimportant because it is the initial step ofmain coal conversionprocesses such as liquefaction gasification and combustionin which coal particles undergo major physical and chemicaltransformations For a better understanding of pyrolysisseveral researchers investigated thermal decomposition ofcoal by thermogravimetric analysis (TGA) As coal has beenused as a fuel since the beginning of industrial developmentit has been among the earliest materials to be subjected tothermal analysis
To investigate the kinetics of the decomposition processTGA is often used In TGA the weight change of the sampleis observed as it is heated usually at a constant heatingrate under a controlled atmosphere such as nitrogen air orother gases The record of weight loss with respect to thetime or temperature is termed a thermogravimetric (TG)thermogramWhen the rate of weight loss (the first derivativewith respect to time) is recorded as a function of time or tem-perature it is called a differential thermogravimetric (DTG)thermogram The DTG has been used to study the kineticsof thermal decomposition reactions of a variety of solidsincluding coal Much of this work is based on the assumptionthat thermal decomposition is describable by an overall first-order reaction and follows the Arrhenius-type equation Thekinetics of the thermal behavior of a material can be deter-mined by the application of a kinetic model to the rates ofmass degradationThemain advantages of TGA for the studyof coal pyrolysis are simplicity in implementation and uti-lization and good repeatability [8] A large number of studieshave reported on thermogravimetric and differential thermalanalysis in an attempt to explain kinetics of thermal decom-position of coal and to obtain qualitative information on coalpyrolysis The literature reviews on these subjects regardingthermal analysis are present from Howard [9] Lawson [10]Anthony and Howard [11] Hathi [12] and Khawam [13] Themain differences in the thermobalances used for the studies ofHonda (1915) Guichard (1926) Vallet (1932) Rigollet (1934)Dubois (1935) Longechambon (1936) and Jouin (1947) werementioned by Hathi [12] and Khawam [13] These ther-mobalances recorded mass versus temperature or time VanHeerdan and Huntjens studied the rates of decomposition ofDutch coals on a thermobalance that recorded mass loss datacontinuously over the temperature range 200ndash550∘CAmath-ematical equation in the form of the Arrhenius equation wasconsidered to explain the rate of coal decomposition Theyconcluded that the decomposition process is first order withregard to the fraction of unreacted coal They observed thatinitial devolatilization is fast removal of moisture and oxidesof carbon the middle devolatilization is slow and contains
the removal of the major volatile matter from coal and thefinal devolatilization is a slow process for liberating the gasfrom residuals [12] Scaccia et al investigated the pyrolysisof low-rank Sulcis coal by thermogravimetric techniques(TGDTG) in the temperature range ambient to 1000∘C atthree different heating rates From thermogravimetric resultsit was established that coal pyrolysis involved three mainstages water evaporation devolatilization of thermally labileand more stable volatiles and char formation [14]
The knowledge of kinetic parameters is essential formodeling the reactor and optimization of the process con-ditions There are various methods for evaluating kineticparameters from nonisothermal thermogravimetric analysis(TGA) and the most common of them can be classifiedinto two major types model-fitting and model-free [14ndash17]In the model-fitting method different models are fit to theexperimental data and the model giving the best statisticalfit is selected as the model from which the activation energy(119864119886
) and frequency factor (119860) are evaluated Historicallymodel-fitting methods were broadly used because of theirability to directly calculate the kinetic parameters from thethermogravimetric analysis results However these methodshave several drawbacks the most important one being theirinability to uniquely select the appropriate reaction model[13] Furthermore comparing the results of these models inthe literature can be difficult especially for nonisothermaldata since a wide range of kinetic parameters have been deter-mined for the coal pyrolysis process This led to the declineof these methods in favor of isoconversional (model-free)methods which can estimate the activation energy withoutevaluating the reaction model [13] The greatest advantagesof this model are its simplicity and avoidance of errorsrelated to selecting specific reaction models Isoconversionalmethod is called model-free method because of its ability todetermine the activation energy for different constant extentsof conversion without considering any particular form ofthe reaction model These methods require several kineticcurves to perform the analysis and thus are sometimes calledmulticurve methods [18] These methods can calculate theactivation energy at different heating rates on the same valueof conversion The terms ldquomodel-freerdquo and ldquoisoconversionalrdquoare sometimes used interchangeably however not all model-free methods are isoconversional For example the Kissingermethod is a model-free method but is not isoconversionalbecause it does not calculate activation energy at differentconstant extents of conversion but instead assumes constantactivation energy [13]
Isoconversional methods are helpful tools for the analysisof solid-state kinetics Theoretically they include many ben-efits and applications However practically they have somedisadvantages especially regarding reproducibility when per-forming a series of runs at different heating rates in whichtheir fluctuation may enhance experimental errors Thus fornonisothermal experiments each run must be conductedunder the same experimental conditions (sample weightpurge gas rate and sample size) so the only variable is theheating rate In order to obtain accurate results with high res-olution curves low ranges of heating rates can be consideredfor the experiments
International Journal of Chemical Engineering 3
Numerous recent studies on the TGA pyrolysis of coal[19ndash21] and coal-biomass blends [22ndash24] are available in theliterature and most of them are based on model-fitting tech-niques There are a few reports relating to thermal decompo-sition behavior of coal based on model-free techniques [14]Moreover most of the previous studies have been performedon coal-biomass blends in order to determine the kinetics ofcopyrolysis of coal and biomass mixtures To the best of ourknowledge there is very little information regarding pyrolysisof coal itself based on model-free methods
The aim of the present work is to study the pyrolysiskinetics of Canadian lignite coal by means of thermogravi-metric analysis (TGA) within the temperature range of 298ndash1173 K at different heating rates under nitrogen atmosphereThe effect of the heating rate on decomposition will also bestudied In this study different model-free methods such asthe Kissinger and the isoconversional methods of OzawaKissinger-Akahira-Sunose and Friedman are employed andcompared in order to analyze nonisothermal kinetic data andinvestigate thermal behavior of a Canadian lignite coal Thekinetic parameters of the coal decomposition processwill alsobe determined These results may provide helpful informa-tion for pyrolysis researchers to predict a kineticmodel of coalpyrolysis and optimization of the process conditions
2 Materials and Methods
21 Sample Preparation It is estimated that approximatelyhalf of the coal resources of the world are low-rank coalsuch as lignite and subbituminous coal [25] Lignite coal isabundant in Canada and plays an important role in energyproduction It was thus chosen as the experimental samplein the present study Canadian lignite coal was obtainedfrom Poplar River Mine located in southern SaskatchewanCanadaThe bulk coal sample was crushed by means of a jawcrusher and ground in a ball mill and blended to homogenizethe coal and reduce the particle size between 106 and 150120583mThe coal sample was received wet with 32 moisture contentandwas dried in vacuum oven at 80∘C for 8 h until a moisturecontent of 12 was achieved The sample was submitted toboth proximate analysis according to the ASTM D7582 byMacro Thermogravimetric Analyzer and ultimate analysisaccording to ASTMD3176 in Elemental Vario MICRO CubeThe results of the proximate and ultimate analysis (CHNS) aswell as higher heating value of the sample used are presentedin Table 1 Higher heating value of coal was also calculatedwith Channiwala and Parikh formula [26]
HHV = 03491C + 11783H + 01005S minus 01034O
minus 00151N minus 00211A (MJkg) (1)
where C H S O N and A are the mass fractions of carbonhydrogen sulfur oxygen nitrogen and ash respectively
22 Experimental Method The TGA experiments were per-formed using a thermogravimetric analyzer TGAndashSDTQ600 at the coal research center of University of AlbertaAbout 10mg of fine coal particle size between 106 and
Table 1 Characteristics of the coal sample
Proximate analysis (wt) Ultimate analysis (wt daf)Moisture 1278 C 4463Volatile matter 4124 H 468Ash 1983 N 066Fixed carbon 2615 S 057
Olowast 4946HHV (MJkg) 1602
daf = dry and ash-free basis lowastObtained by difference
150 120583m was placed in a small Alumina crucible for eachrun and heated from 298K to the maximum temperatureof 1173 K at six different heating rates of 1 6 9 12 15 and18 Kminminus1 respectively under nitrogen atmosphere with aflow rate of 100mLmin During the heating variation of theweight loss and its derivative with respect to the time andtemperature was collected automatically by the instrumentand determined through the TA universal analysis softwareThe experiments were repeated under identical conditions tocheck the reproducibility of the results
23 Kinetic Analysis There are a number of approaches formodelling the complex pyrolysis process The simplest is theempirical model which employs global kinetics where theArrhenius expression is used to correlate the rates of massloss with temperature The pyrolysis process of coal can beexpressed by the following reaction
Coal 119896997888rarr Volatiles + Char (2)
The general expression for the decomposition of a solidsample is
119889119909
119889119905= 119896 (119879) 119891 (119909) (3)
where 119909 is the degree of conversion which represents thedecomposed amount of the sample at time 119905 and is definedin terms of the change in mass of the sample
119909 =(119898119894
minus 119898119905
)
(119898119894
minus 119898119891
)
(4)
where 119898119894
is the initial mass 119898119891
is the final mass and 119898119905
isthe mass at time 119905 of the sample analyzed by TGA 119891(119909) is afunction of 119909 depending on the reaction mechanism 119896(119879) isthe rate constant at temperature 119879 which generally obeys theArrhenius equation
119896 (119879) = 119860 exp(minus119864119886
119877119879) (5)
where119860 is the preexponential factor (minminus1) Equation is theactivation energy (kJmolminus1) 119877 is the universal gas constant(J Kminus1molminus1) and 119879 is the absolute temperature (K)
Substitution of (5) into (3) gives the general expression tocalculate the kinetic parameters
119889119909
119889119905= 119891 (119909)119860 exp(
minus119864119886
119877119879) (6)
4 International Journal of Chemical Engineering
There are various possibilities to express the conversionfunction 119891(119909) for the solid-state reactions Most of theprevious authors used the conversion function as follows
119891 (119909) = (1 minus 119909)119899
(7)
where 119899 is the reaction order here it is considered first orderCombining (6) and (7) the kinetic equation of decomposi-tion is obtained as follows
119889119909
119889119905= 119860 exp(
minus119864119886
119877119879) (1 minus 119909)
119899
(8)
Under nonisothermal conditions inwhich samples are heatedat constant heating rates the actual temperature under thiscondition can be expressed as
119879 = 1198790
+ 120573119905 (9)
where 1198790
is the initial temperature 120573 is the linear heating rate(∘Cmin) and 119879 is the temperature at time 119905 Nonisothermalmethods are usually common in solid-state kinetics becausethey require less experimental data in comparison to isother-mal methodsThe following expression can be considered fornonisothermal experiments
119889119909
119889119879=119889119909
119889119905sdot119889119905
119889119879 (10)
where 119889119909119889119879 is the nonisothermal reaction rate 119889119909119889119905 is theisothermal reaction rate and 119889119879119889119905 is the heating rate (120573)Substituting (8) into (10) gives
119889119909
119889119879=119860
119861exp(minus119864119886
119877119879) (1 minus 119909)
119899
(11)
Equation (11) represents the differential form of the non-isothermal rate law In this study the data fromnonisothermalexperiments are considered to calculate kinetic parametersbased on model-free methods such as Kissinger and the iso-conversional methods of Ozawa Kissinger-Akahira-Sunoseand Friedman and compared in order to analyze and toinvestigate thermal behavior of a Canadian lignite coal
24 Model-Free Methods The kinetic analysis based onmodel-free methods allows the kinetic parameters to beevaluated for different constant extents of conversion withoutevaluating any particular form of the reaction model Thetemperature sensitivity of the reaction rate depends on theextent of conversion to products This is partly a result ofthe heterogeneous nature of solid-state reactions such as coalpyrolysis it also arises somewhat because many solid-statereactions follow complex mechanisms including multipleseries and parallel stages with different activation energiesModel-fitting methods are applied to extract a single setof Arrhenius parameters for an overall process and are notcapable to show this type of complexity in the solid-statereactions Model-free methods are able of addressing theaforementioned drawbacks of themodel-fittingmethodsTheability of model-free methods to show this type of reactioncomplexity is therefore a critical step toward the ability toexplain mechanistic conclusions from kinetic data
241 Kissinger Method According to Kissinger the maxi-mum reaction rate occurs with an increase in the reactiontemperature [27] The degree of conversion at the peaktemperature of the DTG curve is a constant at differentheating rates Kissinger method is a model-free methodbut it is not isoconversional method because it assumesconstant activation energy with the progress of conversionIn Kissinger equation (12) 119879
119898
representing the peak temper-ature is expressed as
ln( 1198611198792119898
) = ln(119860119877119864119886
) minus119864119886
119877119879119898
(12)
Therefore kinetic parameters including activation energy(119864119886
) and preexponential factor (119860) can be obtained from aplot of ln(1198611198792
119898
) versus 1000119879119898
for a series of experimentsat different heating rates
242 Kissinger-Akahira-Sunose (KAS)MethodTheKissinger-Akahira-Sunose (KAS) method was based on the followingequation
ln( 1198611198792) = ln( 119860119877
119864119886
119892 (119909)) minus119864119886
119877119879 (13)
where 119892(119909) is the integral conversion function (reactionmodel) which is reported in the literature [15] For constantconversion a plot of left side of the above equation against1000119879 at different heating rates is a straight line whoseslope and intercept can evaluate the activation energy andpreexponential factor respectively
243 The Flynn-Wall-Ozawa (FWO)Method TheKissinger-Akahira-Sunose (FWO) method is based on the followingequation
ln (119861) = ln(119860119864119886
119877119892 (119909)) minus 5331 minus 1052
119864119886
119877119879 (14)
Thus for a constant conversion a plot of natural logarithmof heating rates ln(119861) versus 1000119879 obtained from thermalcurves recorded at different heating rateswill be a straight linewhose slope (minus1052(119864
119886
119877119879)) will calculate the activationenergy
244 Friedman Method This method is one of the firstisoconversional methods Using (2) and (4) and taking thenatural logarithm of each side the expression proposed byFriedman can be presented as
ln(119889119909119889119905) = ln [119860119891 (119909)] minus
119864119886
119877119879 (15)
The activation energy (119864119886
) is determined from the slope ofthe plot of ln(119889119909119889119905) versus 1000119879 at a constant conversionvalue
International Journal of Chemical Engineering 5
3 Results and Discussion
31Thermal Decomposition Process TheTG andDTG curvesof the pyrolysis of a Canadian lignite coal under nitrogenatmosphere obtained at six different heating rates of 1 6 9 1215 and 18K minminus1 are shown in Figures 1 and 2 respectivelyTheTG curves show the percentagemass loss of a coal sampleover the range of temperature from 298K to 1173 K The rateof mass loss is temperature dependent the higher the tem-perature the larger the mass loss because pyrolysis processproceeds slowly at low temperatures As shown in Figure 1the devolatilization process launches at temperature about450K and proceeds fast with elevating the temperature up to850K and then themass loss of the sample drops slowly to theultimate temperature The DTG curves of sample at differentheating rates are illustrated in Figure 2 The DTG curveexhibits three zones related to moisture evaporation primarydecomposition and secondary decompositionThe first zonerepresents elimination of moisture which occurs below 450K[28] The second region is related to main decompositionstage in the temperature range 450ndash850K for low heatingrate and 925K for high heating rate Major volatile matterat this stage liberated from coal structure that was formedby thermal decomposition some covalent bond such as etherbonds and methylene group which will form gases such ashydrogen carbon monoxide and lighter hydrocarbons [29]This region is themost significant region to examine since themajor weight loss and complicated chemical reaction such asrelease of tar and gaseous products and semicoke formationtake place in this temperature range [30 31] The third zonethat is the second pyrolysis stage where low decompositionrates are observed can be attributed to the further gasificationof the formed char due to high temperature effects On theother hand the coal sample contains high ash and the phasetransitions of the inorganics found in the mineral matterlosses of the molecular water contents of the clay mineralsand decomposition of carbonate minerals may contributeto weight loss of this step There is only a small drop ofmass observed at this stage The TGA data are normalizedfrom 0 to 1 before analysis The temperature at which thederivative of mass loss starts to increase is selected as thezero conversion point and the temperature at which themassderivative returned to the base line is chosen as end point Itis known that the heating rate affects all TGA curves and themaximum decomposition rate When heating rate increasesthe temperature of the maximum decomposition rate ofthe coal shifted toward higher temperature Figure 3 showsconversion curves versus temperature at different heatingrates The curves showed typical sigmoid shape of kineticcurves With increasing the heating rate conversion valuesreachedhigher temperatures because at the same temperatureand time a high heating rate has a short decomposition timeand the temperature required for the sample to reach thesame conversion will be higher The heat transfer limitation(thermal lag) exists between furnace and sample temperatureIt means that temperature in the particle can be a little lowerthan furnace temperature and gradient of temperature mayexist in the coal sample so in order to reduce the thermallag the coal sample should be ground to the fine particle to
300
40
50
60
70
80
90
100
Wei
ght (
)
400 500 600 700 800 900 1000 1100
Temperature (K)1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 1 Thermal behavior of Poplar coal at different heating ratesunder N
2
atmosphere
300
5
4
3
2
1
0
400 500 600 700 800 900 1000 1100
Temperature (K)
DTG
(m
in)
1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 2 DTG curves of Poplar coal at different heating rates underN2
atmosphere
increase the surface area of particle and consequently increasethe heat transfer effect between the sample surface and thecrucible as large as possible
32 Kinetic Analysis The results of TGDTG experimentaldata of coal pyrolysis obtained under nonisothermal con-dition under nitrogen atmosphere were used for kineticanalysis Different model-free methods such as Kissinger andthe isoconversional methods of Ozawa Kissinger-Akahira-Sunose and Friedman are employed in order to obtainparameters like the activation energy and preexponential
6 International Journal of Chemical Engineering
45000
02
04
06
08
10
550 650 750 850500 600 700 800 900
Temperature (K)
Con
vers
ion
(x)
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 3 Conversion curves for different heating rates for pyrolysisof lignite coal under N
2
atmosphere
factor In theKissingermethod the degree of conversion at thepeak temperature (119879
119898
) is a constant under different heatingrates The kinetic parameters using Kissinger method werefound by linear regression line which is shown in Figure 4The activation energy and preexponential factor extractedfrom the slope and intercept are 281 kJmolminus1 and 261 times1017minminus1 respectively The activation energy and preexpo-nential factor were calculated as a function of conversionby using isoconversional methods of KAS FWO and Fried-man methods The isoconversional plots of these methodsare shown in Figures 5ndash7 respectively Different range ofconversion from 005 to 09 is considered for calculating thekinetic parameters based on isoconversional method Theactivation energies from the slope and preexponential factorsfrom the intercept of three different isoconversional methodswere obtained and listed in Table 2 It can be observed fromTable 2 that the values of activation energies are not similar atdifferent constant extents of conversion because most solid-state reactions are not simple one-stepmechanism and followa complex multistep reaction The thermogravimetric dataanalysis by isoconversional technique may reveal complexityof the solid-state reactions such as coal pyrolysis [14] Itmeansthat in the pyrolysis process of coal the activation energy isa function of conversion Figure 8 shows the dependence ofthe activation energy on extent of conversion The activationenergy rises from about 130 kJmolminus1 at low conversion tonearly 350 kJmolminus1 at 75 conversion and it subsequentlydrops to about 300 kJmolminus1 near the end of reaction Theinitial activation energy valuewas lowdue to cleavage of someweak bonds and elimination of volatile components fromthe coal matrix because at the beginning of the process allthe strong bonds are not cleaved Therefore more activationenergy is required to decompose these stable moleculesWith the progress of pyrolysis process the value of activation
138 140 142 144 146 148
Kissinger
R2 = 09858
y = minus33803x + 36586
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
ln(120573
1000Tm (Kminus1)
T2 m
)
Figure 4 Kissinger plot of lignite coal pyrolysis at different heatingrates
12 14 16 18 20 22
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus95
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
minus140
ln(120573T
2)
1000T (Kminus1)
Figure 5 KAS plots of lignite coal pyrolysis at different values ofconversion
energy increased up to conversion of 75 with breaking ofsome strong covalent linkages For higher conversion valuesabove 75 the activation energy gradually decreases Thereason arises from the fact that during the decompositionprocess at high temperature with high conversion whenmostof the stable bonds are broken less stablemolecules which areeasier to break are present so less energy barrier is requiredfor decomposition at this step and the value of activationenergy decreases with progress of conversionThe arithmeticmeans of the activation energy calculated by KAS FWO andFriedmanmethod are 282 275 and 283 kJmolminus1 respectively
International Journal of Chemical Engineering 7
12 14 16 18 20 22 24
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus05
00
05
10
15
20
25
30
35
1000T (Kminus1)
ln 120573
Figure 6 FWO plots of lignite coal pyrolysis at different values ofconversion
12 14 16 18 20 22
minus20
minus25
minus30
minus35
minus40
minus45
minus50
minus55
minus60
minus65
minus70
minus75
ln(dxdt)
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
1000T (Kminus1)
Figure 7 Friedman plots of lignite coal pyrolysis at different valuesof conversion
which are close to average activation energy obtained fromthe Kissinger method (28103 kJmolminus1) The results obtainedwith KAS and Friedman methods are very close and in goodagreement [32]The kinetic data obtained for pyrolysis of coalare found to agree closely with some of the literature dataHowever the differences observed in the literature data canbe attributed to the fact that the pyrolysis characteristics ofcoal highly depend on the properties of the coal which in turndiffers based on origin of the coal [28 30 31]
The KAS and FWOmethods were originally derived withconstant activation energies so the errors associated withkinetic measurements from methods should be dependent
Activ
atio
n en
ergy
Conversion (x)
KissingerKAS
FWOFriedman
00100
150
200
250
300
350
400
02 04 06 08 10
Figure 8 The activation energy as a function of conversion usingmodel-free isoconversional technique
on the variation of the activation energy with respect toconversion This error does not appear in the Friedmanmethod [33 34] Another advantage that can be attributed toFriedman method is that the activation energies obtained bythe Friedmanmethod are independent of the range of heatingrates which can decrease the systematic error in evaluatingthe activation energy values Thus Friedman method canbe considered to be the best among the four model-freemethods in order to evaluate kinetic parameters for solid-state reactions [33 34] The kinetic parameters obtainedin this study can be useful for pyrolysis and gasificationresearchers to predict kinetic model of coal pyrolysis andoptimization of the process conditions
4 Conclusion
In this study the pyrolysis kinetics of a Canadian lignite coalwas carried out bymeans of thermogravimetric analysis (TG)in the temperature range of 298ndash1173K at six different heatingrates of 1 6 9 12 15 and 18 Kminminus1 under nitrogen atmo-sphere It was found that the main pyrolysis process occurredin the temperature range 450ndash850K In this work kineticstudy and thermal behavior of lignite coal were presentedwhere Arrhenius parameters were determined and comparedthrough four different methods of Kissinger Ozawa KASand Friedman The activation energy is calculated as afunction of conversion by using these methods and is foundto be similar Among these methods Friedman methodwas considered to be the best in order to evaluate kineticparameters for solid-state reactions such as coal pyrolysisMethods such as FWO and KAS are restricted to the use of alinear variation of the temperature and positive heating rateMoreover they are generated based onmathematical approx-imation which can enhance systematic error The advantage
8 International Journal of Chemical Engineering
Table 2 Calculated kinetic parameters for a Canadian lignite coal by three different isoconversional methods
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
Numerous recent studies on the TGA pyrolysis of coal[19ndash21] and coal-biomass blends [22ndash24] are available in theliterature and most of them are based on model-fitting tech-niques There are a few reports relating to thermal decompo-sition behavior of coal based on model-free techniques [14]Moreover most of the previous studies have been performedon coal-biomass blends in order to determine the kinetics ofcopyrolysis of coal and biomass mixtures To the best of ourknowledge there is very little information regarding pyrolysisof coal itself based on model-free methods
The aim of the present work is to study the pyrolysiskinetics of Canadian lignite coal by means of thermogravi-metric analysis (TGA) within the temperature range of 298ndash1173 K at different heating rates under nitrogen atmosphereThe effect of the heating rate on decomposition will also bestudied In this study different model-free methods such asthe Kissinger and the isoconversional methods of OzawaKissinger-Akahira-Sunose and Friedman are employed andcompared in order to analyze nonisothermal kinetic data andinvestigate thermal behavior of a Canadian lignite coal Thekinetic parameters of the coal decomposition processwill alsobe determined These results may provide helpful informa-tion for pyrolysis researchers to predict a kineticmodel of coalpyrolysis and optimization of the process conditions
2 Materials and Methods
21 Sample Preparation It is estimated that approximatelyhalf of the coal resources of the world are low-rank coalsuch as lignite and subbituminous coal [25] Lignite coal isabundant in Canada and plays an important role in energyproduction It was thus chosen as the experimental samplein the present study Canadian lignite coal was obtainedfrom Poplar River Mine located in southern SaskatchewanCanadaThe bulk coal sample was crushed by means of a jawcrusher and ground in a ball mill and blended to homogenizethe coal and reduce the particle size between 106 and 150120583mThe coal sample was received wet with 32 moisture contentandwas dried in vacuum oven at 80∘C for 8 h until a moisturecontent of 12 was achieved The sample was submitted toboth proximate analysis according to the ASTM D7582 byMacro Thermogravimetric Analyzer and ultimate analysisaccording to ASTMD3176 in Elemental Vario MICRO CubeThe results of the proximate and ultimate analysis (CHNS) aswell as higher heating value of the sample used are presentedin Table 1 Higher heating value of coal was also calculatedwith Channiwala and Parikh formula [26]
HHV = 03491C + 11783H + 01005S minus 01034O
minus 00151N minus 00211A (MJkg) (1)
where C H S O N and A are the mass fractions of carbonhydrogen sulfur oxygen nitrogen and ash respectively
22 Experimental Method The TGA experiments were per-formed using a thermogravimetric analyzer TGAndashSDTQ600 at the coal research center of University of AlbertaAbout 10mg of fine coal particle size between 106 and
Table 1 Characteristics of the coal sample
Proximate analysis (wt) Ultimate analysis (wt daf)Moisture 1278 C 4463Volatile matter 4124 H 468Ash 1983 N 066Fixed carbon 2615 S 057
Olowast 4946HHV (MJkg) 1602
daf = dry and ash-free basis lowastObtained by difference
150 120583m was placed in a small Alumina crucible for eachrun and heated from 298K to the maximum temperatureof 1173 K at six different heating rates of 1 6 9 12 15 and18 Kminminus1 respectively under nitrogen atmosphere with aflow rate of 100mLmin During the heating variation of theweight loss and its derivative with respect to the time andtemperature was collected automatically by the instrumentand determined through the TA universal analysis softwareThe experiments were repeated under identical conditions tocheck the reproducibility of the results
23 Kinetic Analysis There are a number of approaches formodelling the complex pyrolysis process The simplest is theempirical model which employs global kinetics where theArrhenius expression is used to correlate the rates of massloss with temperature The pyrolysis process of coal can beexpressed by the following reaction
Coal 119896997888rarr Volatiles + Char (2)
The general expression for the decomposition of a solidsample is
119889119909
119889119905= 119896 (119879) 119891 (119909) (3)
where 119909 is the degree of conversion which represents thedecomposed amount of the sample at time 119905 and is definedin terms of the change in mass of the sample
119909 =(119898119894
minus 119898119905
)
(119898119894
minus 119898119891
)
(4)
where 119898119894
is the initial mass 119898119891
is the final mass and 119898119905
isthe mass at time 119905 of the sample analyzed by TGA 119891(119909) is afunction of 119909 depending on the reaction mechanism 119896(119879) isthe rate constant at temperature 119879 which generally obeys theArrhenius equation
119896 (119879) = 119860 exp(minus119864119886
119877119879) (5)
where119860 is the preexponential factor (minminus1) Equation is theactivation energy (kJmolminus1) 119877 is the universal gas constant(J Kminus1molminus1) and 119879 is the absolute temperature (K)
Substitution of (5) into (3) gives the general expression tocalculate the kinetic parameters
119889119909
119889119905= 119891 (119909)119860 exp(
minus119864119886
119877119879) (6)
4 International Journal of Chemical Engineering
There are various possibilities to express the conversionfunction 119891(119909) for the solid-state reactions Most of theprevious authors used the conversion function as follows
119891 (119909) = (1 minus 119909)119899
(7)
where 119899 is the reaction order here it is considered first orderCombining (6) and (7) the kinetic equation of decomposi-tion is obtained as follows
119889119909
119889119905= 119860 exp(
minus119864119886
119877119879) (1 minus 119909)
119899
(8)
Under nonisothermal conditions inwhich samples are heatedat constant heating rates the actual temperature under thiscondition can be expressed as
119879 = 1198790
+ 120573119905 (9)
where 1198790
is the initial temperature 120573 is the linear heating rate(∘Cmin) and 119879 is the temperature at time 119905 Nonisothermalmethods are usually common in solid-state kinetics becausethey require less experimental data in comparison to isother-mal methodsThe following expression can be considered fornonisothermal experiments
119889119909
119889119879=119889119909
119889119905sdot119889119905
119889119879 (10)
where 119889119909119889119879 is the nonisothermal reaction rate 119889119909119889119905 is theisothermal reaction rate and 119889119879119889119905 is the heating rate (120573)Substituting (8) into (10) gives
119889119909
119889119879=119860
119861exp(minus119864119886
119877119879) (1 minus 119909)
119899
(11)
Equation (11) represents the differential form of the non-isothermal rate law In this study the data fromnonisothermalexperiments are considered to calculate kinetic parametersbased on model-free methods such as Kissinger and the iso-conversional methods of Ozawa Kissinger-Akahira-Sunoseand Friedman and compared in order to analyze and toinvestigate thermal behavior of a Canadian lignite coal
24 Model-Free Methods The kinetic analysis based onmodel-free methods allows the kinetic parameters to beevaluated for different constant extents of conversion withoutevaluating any particular form of the reaction model Thetemperature sensitivity of the reaction rate depends on theextent of conversion to products This is partly a result ofthe heterogeneous nature of solid-state reactions such as coalpyrolysis it also arises somewhat because many solid-statereactions follow complex mechanisms including multipleseries and parallel stages with different activation energiesModel-fitting methods are applied to extract a single setof Arrhenius parameters for an overall process and are notcapable to show this type of complexity in the solid-statereactions Model-free methods are able of addressing theaforementioned drawbacks of themodel-fittingmethodsTheability of model-free methods to show this type of reactioncomplexity is therefore a critical step toward the ability toexplain mechanistic conclusions from kinetic data
241 Kissinger Method According to Kissinger the maxi-mum reaction rate occurs with an increase in the reactiontemperature [27] The degree of conversion at the peaktemperature of the DTG curve is a constant at differentheating rates Kissinger method is a model-free methodbut it is not isoconversional method because it assumesconstant activation energy with the progress of conversionIn Kissinger equation (12) 119879
119898
representing the peak temper-ature is expressed as
ln( 1198611198792119898
) = ln(119860119877119864119886
) minus119864119886
119877119879119898
(12)
Therefore kinetic parameters including activation energy(119864119886
) and preexponential factor (119860) can be obtained from aplot of ln(1198611198792
119898
) versus 1000119879119898
for a series of experimentsat different heating rates
242 Kissinger-Akahira-Sunose (KAS)MethodTheKissinger-Akahira-Sunose (KAS) method was based on the followingequation
ln( 1198611198792) = ln( 119860119877
119864119886
119892 (119909)) minus119864119886
119877119879 (13)
where 119892(119909) is the integral conversion function (reactionmodel) which is reported in the literature [15] For constantconversion a plot of left side of the above equation against1000119879 at different heating rates is a straight line whoseslope and intercept can evaluate the activation energy andpreexponential factor respectively
243 The Flynn-Wall-Ozawa (FWO)Method TheKissinger-Akahira-Sunose (FWO) method is based on the followingequation
ln (119861) = ln(119860119864119886
119877119892 (119909)) minus 5331 minus 1052
119864119886
119877119879 (14)
Thus for a constant conversion a plot of natural logarithmof heating rates ln(119861) versus 1000119879 obtained from thermalcurves recorded at different heating rateswill be a straight linewhose slope (minus1052(119864
119886
119877119879)) will calculate the activationenergy
244 Friedman Method This method is one of the firstisoconversional methods Using (2) and (4) and taking thenatural logarithm of each side the expression proposed byFriedman can be presented as
ln(119889119909119889119905) = ln [119860119891 (119909)] minus
119864119886
119877119879 (15)
The activation energy (119864119886
) is determined from the slope ofthe plot of ln(119889119909119889119905) versus 1000119879 at a constant conversionvalue
International Journal of Chemical Engineering 5
3 Results and Discussion
31Thermal Decomposition Process TheTG andDTG curvesof the pyrolysis of a Canadian lignite coal under nitrogenatmosphere obtained at six different heating rates of 1 6 9 1215 and 18K minminus1 are shown in Figures 1 and 2 respectivelyTheTG curves show the percentagemass loss of a coal sampleover the range of temperature from 298K to 1173 K The rateof mass loss is temperature dependent the higher the tem-perature the larger the mass loss because pyrolysis processproceeds slowly at low temperatures As shown in Figure 1the devolatilization process launches at temperature about450K and proceeds fast with elevating the temperature up to850K and then themass loss of the sample drops slowly to theultimate temperature The DTG curves of sample at differentheating rates are illustrated in Figure 2 The DTG curveexhibits three zones related to moisture evaporation primarydecomposition and secondary decompositionThe first zonerepresents elimination of moisture which occurs below 450K[28] The second region is related to main decompositionstage in the temperature range 450ndash850K for low heatingrate and 925K for high heating rate Major volatile matterat this stage liberated from coal structure that was formedby thermal decomposition some covalent bond such as etherbonds and methylene group which will form gases such ashydrogen carbon monoxide and lighter hydrocarbons [29]This region is themost significant region to examine since themajor weight loss and complicated chemical reaction such asrelease of tar and gaseous products and semicoke formationtake place in this temperature range [30 31] The third zonethat is the second pyrolysis stage where low decompositionrates are observed can be attributed to the further gasificationof the formed char due to high temperature effects On theother hand the coal sample contains high ash and the phasetransitions of the inorganics found in the mineral matterlosses of the molecular water contents of the clay mineralsand decomposition of carbonate minerals may contributeto weight loss of this step There is only a small drop ofmass observed at this stage The TGA data are normalizedfrom 0 to 1 before analysis The temperature at which thederivative of mass loss starts to increase is selected as thezero conversion point and the temperature at which themassderivative returned to the base line is chosen as end point Itis known that the heating rate affects all TGA curves and themaximum decomposition rate When heating rate increasesthe temperature of the maximum decomposition rate ofthe coal shifted toward higher temperature Figure 3 showsconversion curves versus temperature at different heatingrates The curves showed typical sigmoid shape of kineticcurves With increasing the heating rate conversion valuesreachedhigher temperatures because at the same temperatureand time a high heating rate has a short decomposition timeand the temperature required for the sample to reach thesame conversion will be higher The heat transfer limitation(thermal lag) exists between furnace and sample temperatureIt means that temperature in the particle can be a little lowerthan furnace temperature and gradient of temperature mayexist in the coal sample so in order to reduce the thermallag the coal sample should be ground to the fine particle to
300
40
50
60
70
80
90
100
Wei
ght (
)
400 500 600 700 800 900 1000 1100
Temperature (K)1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 1 Thermal behavior of Poplar coal at different heating ratesunder N
2
atmosphere
300
5
4
3
2
1
0
400 500 600 700 800 900 1000 1100
Temperature (K)
DTG
(m
in)
1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 2 DTG curves of Poplar coal at different heating rates underN2
atmosphere
increase the surface area of particle and consequently increasethe heat transfer effect between the sample surface and thecrucible as large as possible
32 Kinetic Analysis The results of TGDTG experimentaldata of coal pyrolysis obtained under nonisothermal con-dition under nitrogen atmosphere were used for kineticanalysis Different model-free methods such as Kissinger andthe isoconversional methods of Ozawa Kissinger-Akahira-Sunose and Friedman are employed in order to obtainparameters like the activation energy and preexponential
6 International Journal of Chemical Engineering
45000
02
04
06
08
10
550 650 750 850500 600 700 800 900
Temperature (K)
Con
vers
ion
(x)
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 3 Conversion curves for different heating rates for pyrolysisof lignite coal under N
2
atmosphere
factor In theKissingermethod the degree of conversion at thepeak temperature (119879
119898
) is a constant under different heatingrates The kinetic parameters using Kissinger method werefound by linear regression line which is shown in Figure 4The activation energy and preexponential factor extractedfrom the slope and intercept are 281 kJmolminus1 and 261 times1017minminus1 respectively The activation energy and preexpo-nential factor were calculated as a function of conversionby using isoconversional methods of KAS FWO and Fried-man methods The isoconversional plots of these methodsare shown in Figures 5ndash7 respectively Different range ofconversion from 005 to 09 is considered for calculating thekinetic parameters based on isoconversional method Theactivation energies from the slope and preexponential factorsfrom the intercept of three different isoconversional methodswere obtained and listed in Table 2 It can be observed fromTable 2 that the values of activation energies are not similar atdifferent constant extents of conversion because most solid-state reactions are not simple one-stepmechanism and followa complex multistep reaction The thermogravimetric dataanalysis by isoconversional technique may reveal complexityof the solid-state reactions such as coal pyrolysis [14] Itmeansthat in the pyrolysis process of coal the activation energy isa function of conversion Figure 8 shows the dependence ofthe activation energy on extent of conversion The activationenergy rises from about 130 kJmolminus1 at low conversion tonearly 350 kJmolminus1 at 75 conversion and it subsequentlydrops to about 300 kJmolminus1 near the end of reaction Theinitial activation energy valuewas lowdue to cleavage of someweak bonds and elimination of volatile components fromthe coal matrix because at the beginning of the process allthe strong bonds are not cleaved Therefore more activationenergy is required to decompose these stable moleculesWith the progress of pyrolysis process the value of activation
138 140 142 144 146 148
Kissinger
R2 = 09858
y = minus33803x + 36586
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
ln(120573
1000Tm (Kminus1)
T2 m
)
Figure 4 Kissinger plot of lignite coal pyrolysis at different heatingrates
12 14 16 18 20 22
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus95
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
minus140
ln(120573T
2)
1000T (Kminus1)
Figure 5 KAS plots of lignite coal pyrolysis at different values ofconversion
energy increased up to conversion of 75 with breaking ofsome strong covalent linkages For higher conversion valuesabove 75 the activation energy gradually decreases Thereason arises from the fact that during the decompositionprocess at high temperature with high conversion whenmostof the stable bonds are broken less stablemolecules which areeasier to break are present so less energy barrier is requiredfor decomposition at this step and the value of activationenergy decreases with progress of conversionThe arithmeticmeans of the activation energy calculated by KAS FWO andFriedmanmethod are 282 275 and 283 kJmolminus1 respectively
International Journal of Chemical Engineering 7
12 14 16 18 20 22 24
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus05
00
05
10
15
20
25
30
35
1000T (Kminus1)
ln 120573
Figure 6 FWO plots of lignite coal pyrolysis at different values ofconversion
12 14 16 18 20 22
minus20
minus25
minus30
minus35
minus40
minus45
minus50
minus55
minus60
minus65
minus70
minus75
ln(dxdt)
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
1000T (Kminus1)
Figure 7 Friedman plots of lignite coal pyrolysis at different valuesof conversion
which are close to average activation energy obtained fromthe Kissinger method (28103 kJmolminus1) The results obtainedwith KAS and Friedman methods are very close and in goodagreement [32]The kinetic data obtained for pyrolysis of coalare found to agree closely with some of the literature dataHowever the differences observed in the literature data canbe attributed to the fact that the pyrolysis characteristics ofcoal highly depend on the properties of the coal which in turndiffers based on origin of the coal [28 30 31]
The KAS and FWOmethods were originally derived withconstant activation energies so the errors associated withkinetic measurements from methods should be dependent
Activ
atio
n en
ergy
Conversion (x)
KissingerKAS
FWOFriedman
00100
150
200
250
300
350
400
02 04 06 08 10
Figure 8 The activation energy as a function of conversion usingmodel-free isoconversional technique
on the variation of the activation energy with respect toconversion This error does not appear in the Friedmanmethod [33 34] Another advantage that can be attributed toFriedman method is that the activation energies obtained bythe Friedmanmethod are independent of the range of heatingrates which can decrease the systematic error in evaluatingthe activation energy values Thus Friedman method canbe considered to be the best among the four model-freemethods in order to evaluate kinetic parameters for solid-state reactions [33 34] The kinetic parameters obtainedin this study can be useful for pyrolysis and gasificationresearchers to predict kinetic model of coal pyrolysis andoptimization of the process conditions
4 Conclusion
In this study the pyrolysis kinetics of a Canadian lignite coalwas carried out bymeans of thermogravimetric analysis (TG)in the temperature range of 298ndash1173K at six different heatingrates of 1 6 9 12 15 and 18 Kminminus1 under nitrogen atmo-sphere It was found that the main pyrolysis process occurredin the temperature range 450ndash850K In this work kineticstudy and thermal behavior of lignite coal were presentedwhere Arrhenius parameters were determined and comparedthrough four different methods of Kissinger Ozawa KASand Friedman The activation energy is calculated as afunction of conversion by using these methods and is foundto be similar Among these methods Friedman methodwas considered to be the best in order to evaluate kineticparameters for solid-state reactions such as coal pyrolysisMethods such as FWO and KAS are restricted to the use of alinear variation of the temperature and positive heating rateMoreover they are generated based onmathematical approx-imation which can enhance systematic error The advantage
8 International Journal of Chemical Engineering
Table 2 Calculated kinetic parameters for a Canadian lignite coal by three different isoconversional methods
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
There are various possibilities to express the conversionfunction 119891(119909) for the solid-state reactions Most of theprevious authors used the conversion function as follows
119891 (119909) = (1 minus 119909)119899
(7)
where 119899 is the reaction order here it is considered first orderCombining (6) and (7) the kinetic equation of decomposi-tion is obtained as follows
119889119909
119889119905= 119860 exp(
minus119864119886
119877119879) (1 minus 119909)
119899
(8)
Under nonisothermal conditions inwhich samples are heatedat constant heating rates the actual temperature under thiscondition can be expressed as
119879 = 1198790
+ 120573119905 (9)
where 1198790
is the initial temperature 120573 is the linear heating rate(∘Cmin) and 119879 is the temperature at time 119905 Nonisothermalmethods are usually common in solid-state kinetics becausethey require less experimental data in comparison to isother-mal methodsThe following expression can be considered fornonisothermal experiments
119889119909
119889119879=119889119909
119889119905sdot119889119905
119889119879 (10)
where 119889119909119889119879 is the nonisothermal reaction rate 119889119909119889119905 is theisothermal reaction rate and 119889119879119889119905 is the heating rate (120573)Substituting (8) into (10) gives
119889119909
119889119879=119860
119861exp(minus119864119886
119877119879) (1 minus 119909)
119899
(11)
Equation (11) represents the differential form of the non-isothermal rate law In this study the data fromnonisothermalexperiments are considered to calculate kinetic parametersbased on model-free methods such as Kissinger and the iso-conversional methods of Ozawa Kissinger-Akahira-Sunoseand Friedman and compared in order to analyze and toinvestigate thermal behavior of a Canadian lignite coal
24 Model-Free Methods The kinetic analysis based onmodel-free methods allows the kinetic parameters to beevaluated for different constant extents of conversion withoutevaluating any particular form of the reaction model Thetemperature sensitivity of the reaction rate depends on theextent of conversion to products This is partly a result ofthe heterogeneous nature of solid-state reactions such as coalpyrolysis it also arises somewhat because many solid-statereactions follow complex mechanisms including multipleseries and parallel stages with different activation energiesModel-fitting methods are applied to extract a single setof Arrhenius parameters for an overall process and are notcapable to show this type of complexity in the solid-statereactions Model-free methods are able of addressing theaforementioned drawbacks of themodel-fittingmethodsTheability of model-free methods to show this type of reactioncomplexity is therefore a critical step toward the ability toexplain mechanistic conclusions from kinetic data
241 Kissinger Method According to Kissinger the maxi-mum reaction rate occurs with an increase in the reactiontemperature [27] The degree of conversion at the peaktemperature of the DTG curve is a constant at differentheating rates Kissinger method is a model-free methodbut it is not isoconversional method because it assumesconstant activation energy with the progress of conversionIn Kissinger equation (12) 119879
119898
representing the peak temper-ature is expressed as
ln( 1198611198792119898
) = ln(119860119877119864119886
) minus119864119886
119877119879119898
(12)
Therefore kinetic parameters including activation energy(119864119886
) and preexponential factor (119860) can be obtained from aplot of ln(1198611198792
119898
) versus 1000119879119898
for a series of experimentsat different heating rates
242 Kissinger-Akahira-Sunose (KAS)MethodTheKissinger-Akahira-Sunose (KAS) method was based on the followingequation
ln( 1198611198792) = ln( 119860119877
119864119886
119892 (119909)) minus119864119886
119877119879 (13)
where 119892(119909) is the integral conversion function (reactionmodel) which is reported in the literature [15] For constantconversion a plot of left side of the above equation against1000119879 at different heating rates is a straight line whoseslope and intercept can evaluate the activation energy andpreexponential factor respectively
243 The Flynn-Wall-Ozawa (FWO)Method TheKissinger-Akahira-Sunose (FWO) method is based on the followingequation
ln (119861) = ln(119860119864119886
119877119892 (119909)) minus 5331 minus 1052
119864119886
119877119879 (14)
Thus for a constant conversion a plot of natural logarithmof heating rates ln(119861) versus 1000119879 obtained from thermalcurves recorded at different heating rateswill be a straight linewhose slope (minus1052(119864
119886
119877119879)) will calculate the activationenergy
244 Friedman Method This method is one of the firstisoconversional methods Using (2) and (4) and taking thenatural logarithm of each side the expression proposed byFriedman can be presented as
ln(119889119909119889119905) = ln [119860119891 (119909)] minus
119864119886
119877119879 (15)
The activation energy (119864119886
) is determined from the slope ofthe plot of ln(119889119909119889119905) versus 1000119879 at a constant conversionvalue
International Journal of Chemical Engineering 5
3 Results and Discussion
31Thermal Decomposition Process TheTG andDTG curvesof the pyrolysis of a Canadian lignite coal under nitrogenatmosphere obtained at six different heating rates of 1 6 9 1215 and 18K minminus1 are shown in Figures 1 and 2 respectivelyTheTG curves show the percentagemass loss of a coal sampleover the range of temperature from 298K to 1173 K The rateof mass loss is temperature dependent the higher the tem-perature the larger the mass loss because pyrolysis processproceeds slowly at low temperatures As shown in Figure 1the devolatilization process launches at temperature about450K and proceeds fast with elevating the temperature up to850K and then themass loss of the sample drops slowly to theultimate temperature The DTG curves of sample at differentheating rates are illustrated in Figure 2 The DTG curveexhibits three zones related to moisture evaporation primarydecomposition and secondary decompositionThe first zonerepresents elimination of moisture which occurs below 450K[28] The second region is related to main decompositionstage in the temperature range 450ndash850K for low heatingrate and 925K for high heating rate Major volatile matterat this stage liberated from coal structure that was formedby thermal decomposition some covalent bond such as etherbonds and methylene group which will form gases such ashydrogen carbon monoxide and lighter hydrocarbons [29]This region is themost significant region to examine since themajor weight loss and complicated chemical reaction such asrelease of tar and gaseous products and semicoke formationtake place in this temperature range [30 31] The third zonethat is the second pyrolysis stage where low decompositionrates are observed can be attributed to the further gasificationof the formed char due to high temperature effects On theother hand the coal sample contains high ash and the phasetransitions of the inorganics found in the mineral matterlosses of the molecular water contents of the clay mineralsand decomposition of carbonate minerals may contributeto weight loss of this step There is only a small drop ofmass observed at this stage The TGA data are normalizedfrom 0 to 1 before analysis The temperature at which thederivative of mass loss starts to increase is selected as thezero conversion point and the temperature at which themassderivative returned to the base line is chosen as end point Itis known that the heating rate affects all TGA curves and themaximum decomposition rate When heating rate increasesthe temperature of the maximum decomposition rate ofthe coal shifted toward higher temperature Figure 3 showsconversion curves versus temperature at different heatingrates The curves showed typical sigmoid shape of kineticcurves With increasing the heating rate conversion valuesreachedhigher temperatures because at the same temperatureand time a high heating rate has a short decomposition timeand the temperature required for the sample to reach thesame conversion will be higher The heat transfer limitation(thermal lag) exists between furnace and sample temperatureIt means that temperature in the particle can be a little lowerthan furnace temperature and gradient of temperature mayexist in the coal sample so in order to reduce the thermallag the coal sample should be ground to the fine particle to
300
40
50
60
70
80
90
100
Wei
ght (
)
400 500 600 700 800 900 1000 1100
Temperature (K)1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 1 Thermal behavior of Poplar coal at different heating ratesunder N
2
atmosphere
300
5
4
3
2
1
0
400 500 600 700 800 900 1000 1100
Temperature (K)
DTG
(m
in)
1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 2 DTG curves of Poplar coal at different heating rates underN2
atmosphere
increase the surface area of particle and consequently increasethe heat transfer effect between the sample surface and thecrucible as large as possible
32 Kinetic Analysis The results of TGDTG experimentaldata of coal pyrolysis obtained under nonisothermal con-dition under nitrogen atmosphere were used for kineticanalysis Different model-free methods such as Kissinger andthe isoconversional methods of Ozawa Kissinger-Akahira-Sunose and Friedman are employed in order to obtainparameters like the activation energy and preexponential
6 International Journal of Chemical Engineering
45000
02
04
06
08
10
550 650 750 850500 600 700 800 900
Temperature (K)
Con
vers
ion
(x)
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 3 Conversion curves for different heating rates for pyrolysisof lignite coal under N
2
atmosphere
factor In theKissingermethod the degree of conversion at thepeak temperature (119879
119898
) is a constant under different heatingrates The kinetic parameters using Kissinger method werefound by linear regression line which is shown in Figure 4The activation energy and preexponential factor extractedfrom the slope and intercept are 281 kJmolminus1 and 261 times1017minminus1 respectively The activation energy and preexpo-nential factor were calculated as a function of conversionby using isoconversional methods of KAS FWO and Fried-man methods The isoconversional plots of these methodsare shown in Figures 5ndash7 respectively Different range ofconversion from 005 to 09 is considered for calculating thekinetic parameters based on isoconversional method Theactivation energies from the slope and preexponential factorsfrom the intercept of three different isoconversional methodswere obtained and listed in Table 2 It can be observed fromTable 2 that the values of activation energies are not similar atdifferent constant extents of conversion because most solid-state reactions are not simple one-stepmechanism and followa complex multistep reaction The thermogravimetric dataanalysis by isoconversional technique may reveal complexityof the solid-state reactions such as coal pyrolysis [14] Itmeansthat in the pyrolysis process of coal the activation energy isa function of conversion Figure 8 shows the dependence ofthe activation energy on extent of conversion The activationenergy rises from about 130 kJmolminus1 at low conversion tonearly 350 kJmolminus1 at 75 conversion and it subsequentlydrops to about 300 kJmolminus1 near the end of reaction Theinitial activation energy valuewas lowdue to cleavage of someweak bonds and elimination of volatile components fromthe coal matrix because at the beginning of the process allthe strong bonds are not cleaved Therefore more activationenergy is required to decompose these stable moleculesWith the progress of pyrolysis process the value of activation
138 140 142 144 146 148
Kissinger
R2 = 09858
y = minus33803x + 36586
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
ln(120573
1000Tm (Kminus1)
T2 m
)
Figure 4 Kissinger plot of lignite coal pyrolysis at different heatingrates
12 14 16 18 20 22
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus95
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
minus140
ln(120573T
2)
1000T (Kminus1)
Figure 5 KAS plots of lignite coal pyrolysis at different values ofconversion
energy increased up to conversion of 75 with breaking ofsome strong covalent linkages For higher conversion valuesabove 75 the activation energy gradually decreases Thereason arises from the fact that during the decompositionprocess at high temperature with high conversion whenmostof the stable bonds are broken less stablemolecules which areeasier to break are present so less energy barrier is requiredfor decomposition at this step and the value of activationenergy decreases with progress of conversionThe arithmeticmeans of the activation energy calculated by KAS FWO andFriedmanmethod are 282 275 and 283 kJmolminus1 respectively
International Journal of Chemical Engineering 7
12 14 16 18 20 22 24
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus05
00
05
10
15
20
25
30
35
1000T (Kminus1)
ln 120573
Figure 6 FWO plots of lignite coal pyrolysis at different values ofconversion
12 14 16 18 20 22
minus20
minus25
minus30
minus35
minus40
minus45
minus50
minus55
minus60
minus65
minus70
minus75
ln(dxdt)
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
1000T (Kminus1)
Figure 7 Friedman plots of lignite coal pyrolysis at different valuesof conversion
which are close to average activation energy obtained fromthe Kissinger method (28103 kJmolminus1) The results obtainedwith KAS and Friedman methods are very close and in goodagreement [32]The kinetic data obtained for pyrolysis of coalare found to agree closely with some of the literature dataHowever the differences observed in the literature data canbe attributed to the fact that the pyrolysis characteristics ofcoal highly depend on the properties of the coal which in turndiffers based on origin of the coal [28 30 31]
The KAS and FWOmethods were originally derived withconstant activation energies so the errors associated withkinetic measurements from methods should be dependent
Activ
atio
n en
ergy
Conversion (x)
KissingerKAS
FWOFriedman
00100
150
200
250
300
350
400
02 04 06 08 10
Figure 8 The activation energy as a function of conversion usingmodel-free isoconversional technique
on the variation of the activation energy with respect toconversion This error does not appear in the Friedmanmethod [33 34] Another advantage that can be attributed toFriedman method is that the activation energies obtained bythe Friedmanmethod are independent of the range of heatingrates which can decrease the systematic error in evaluatingthe activation energy values Thus Friedman method canbe considered to be the best among the four model-freemethods in order to evaluate kinetic parameters for solid-state reactions [33 34] The kinetic parameters obtainedin this study can be useful for pyrolysis and gasificationresearchers to predict kinetic model of coal pyrolysis andoptimization of the process conditions
4 Conclusion
In this study the pyrolysis kinetics of a Canadian lignite coalwas carried out bymeans of thermogravimetric analysis (TG)in the temperature range of 298ndash1173K at six different heatingrates of 1 6 9 12 15 and 18 Kminminus1 under nitrogen atmo-sphere It was found that the main pyrolysis process occurredin the temperature range 450ndash850K In this work kineticstudy and thermal behavior of lignite coal were presentedwhere Arrhenius parameters were determined and comparedthrough four different methods of Kissinger Ozawa KASand Friedman The activation energy is calculated as afunction of conversion by using these methods and is foundto be similar Among these methods Friedman methodwas considered to be the best in order to evaluate kineticparameters for solid-state reactions such as coal pyrolysisMethods such as FWO and KAS are restricted to the use of alinear variation of the temperature and positive heating rateMoreover they are generated based onmathematical approx-imation which can enhance systematic error The advantage
8 International Journal of Chemical Engineering
Table 2 Calculated kinetic parameters for a Canadian lignite coal by three different isoconversional methods
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
31Thermal Decomposition Process TheTG andDTG curvesof the pyrolysis of a Canadian lignite coal under nitrogenatmosphere obtained at six different heating rates of 1 6 9 1215 and 18K minminus1 are shown in Figures 1 and 2 respectivelyTheTG curves show the percentagemass loss of a coal sampleover the range of temperature from 298K to 1173 K The rateof mass loss is temperature dependent the higher the tem-perature the larger the mass loss because pyrolysis processproceeds slowly at low temperatures As shown in Figure 1the devolatilization process launches at temperature about450K and proceeds fast with elevating the temperature up to850K and then themass loss of the sample drops slowly to theultimate temperature The DTG curves of sample at differentheating rates are illustrated in Figure 2 The DTG curveexhibits three zones related to moisture evaporation primarydecomposition and secondary decompositionThe first zonerepresents elimination of moisture which occurs below 450K[28] The second region is related to main decompositionstage in the temperature range 450ndash850K for low heatingrate and 925K for high heating rate Major volatile matterat this stage liberated from coal structure that was formedby thermal decomposition some covalent bond such as etherbonds and methylene group which will form gases such ashydrogen carbon monoxide and lighter hydrocarbons [29]This region is themost significant region to examine since themajor weight loss and complicated chemical reaction such asrelease of tar and gaseous products and semicoke formationtake place in this temperature range [30 31] The third zonethat is the second pyrolysis stage where low decompositionrates are observed can be attributed to the further gasificationof the formed char due to high temperature effects On theother hand the coal sample contains high ash and the phasetransitions of the inorganics found in the mineral matterlosses of the molecular water contents of the clay mineralsand decomposition of carbonate minerals may contributeto weight loss of this step There is only a small drop ofmass observed at this stage The TGA data are normalizedfrom 0 to 1 before analysis The temperature at which thederivative of mass loss starts to increase is selected as thezero conversion point and the temperature at which themassderivative returned to the base line is chosen as end point Itis known that the heating rate affects all TGA curves and themaximum decomposition rate When heating rate increasesthe temperature of the maximum decomposition rate ofthe coal shifted toward higher temperature Figure 3 showsconversion curves versus temperature at different heatingrates The curves showed typical sigmoid shape of kineticcurves With increasing the heating rate conversion valuesreachedhigher temperatures because at the same temperatureand time a high heating rate has a short decomposition timeand the temperature required for the sample to reach thesame conversion will be higher The heat transfer limitation(thermal lag) exists between furnace and sample temperatureIt means that temperature in the particle can be a little lowerthan furnace temperature and gradient of temperature mayexist in the coal sample so in order to reduce the thermallag the coal sample should be ground to the fine particle to
300
40
50
60
70
80
90
100
Wei
ght (
)
400 500 600 700 800 900 1000 1100
Temperature (K)1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 1 Thermal behavior of Poplar coal at different heating ratesunder N
2
atmosphere
300
5
4
3
2
1
0
400 500 600 700 800 900 1000 1100
Temperature (K)
DTG
(m
in)
1200
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 2 DTG curves of Poplar coal at different heating rates underN2
atmosphere
increase the surface area of particle and consequently increasethe heat transfer effect between the sample surface and thecrucible as large as possible
32 Kinetic Analysis The results of TGDTG experimentaldata of coal pyrolysis obtained under nonisothermal con-dition under nitrogen atmosphere were used for kineticanalysis Different model-free methods such as Kissinger andthe isoconversional methods of Ozawa Kissinger-Akahira-Sunose and Friedman are employed in order to obtainparameters like the activation energy and preexponential
6 International Journal of Chemical Engineering
45000
02
04
06
08
10
550 650 750 850500 600 700 800 900
Temperature (K)
Con
vers
ion
(x)
1Kmin6Kmin9Kmin
12Kmin15Kmin18Kmin
Figure 3 Conversion curves for different heating rates for pyrolysisof lignite coal under N
2
atmosphere
factor In theKissingermethod the degree of conversion at thepeak temperature (119879
119898
) is a constant under different heatingrates The kinetic parameters using Kissinger method werefound by linear regression line which is shown in Figure 4The activation energy and preexponential factor extractedfrom the slope and intercept are 281 kJmolminus1 and 261 times1017minminus1 respectively The activation energy and preexpo-nential factor were calculated as a function of conversionby using isoconversional methods of KAS FWO and Fried-man methods The isoconversional plots of these methodsare shown in Figures 5ndash7 respectively Different range ofconversion from 005 to 09 is considered for calculating thekinetic parameters based on isoconversional method Theactivation energies from the slope and preexponential factorsfrom the intercept of three different isoconversional methodswere obtained and listed in Table 2 It can be observed fromTable 2 that the values of activation energies are not similar atdifferent constant extents of conversion because most solid-state reactions are not simple one-stepmechanism and followa complex multistep reaction The thermogravimetric dataanalysis by isoconversional technique may reveal complexityof the solid-state reactions such as coal pyrolysis [14] Itmeansthat in the pyrolysis process of coal the activation energy isa function of conversion Figure 8 shows the dependence ofthe activation energy on extent of conversion The activationenergy rises from about 130 kJmolminus1 at low conversion tonearly 350 kJmolminus1 at 75 conversion and it subsequentlydrops to about 300 kJmolminus1 near the end of reaction Theinitial activation energy valuewas lowdue to cleavage of someweak bonds and elimination of volatile components fromthe coal matrix because at the beginning of the process allthe strong bonds are not cleaved Therefore more activationenergy is required to decompose these stable moleculesWith the progress of pyrolysis process the value of activation
138 140 142 144 146 148
Kissinger
R2 = 09858
y = minus33803x + 36586
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
ln(120573
1000Tm (Kminus1)
T2 m
)
Figure 4 Kissinger plot of lignite coal pyrolysis at different heatingrates
12 14 16 18 20 22
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus95
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
minus140
ln(120573T
2)
1000T (Kminus1)
Figure 5 KAS plots of lignite coal pyrolysis at different values ofconversion
energy increased up to conversion of 75 with breaking ofsome strong covalent linkages For higher conversion valuesabove 75 the activation energy gradually decreases Thereason arises from the fact that during the decompositionprocess at high temperature with high conversion whenmostof the stable bonds are broken less stablemolecules which areeasier to break are present so less energy barrier is requiredfor decomposition at this step and the value of activationenergy decreases with progress of conversionThe arithmeticmeans of the activation energy calculated by KAS FWO andFriedmanmethod are 282 275 and 283 kJmolminus1 respectively
International Journal of Chemical Engineering 7
12 14 16 18 20 22 24
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus05
00
05
10
15
20
25
30
35
1000T (Kminus1)
ln 120573
Figure 6 FWO plots of lignite coal pyrolysis at different values ofconversion
12 14 16 18 20 22
minus20
minus25
minus30
minus35
minus40
minus45
minus50
minus55
minus60
minus65
minus70
minus75
ln(dxdt)
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
1000T (Kminus1)
Figure 7 Friedman plots of lignite coal pyrolysis at different valuesof conversion
which are close to average activation energy obtained fromthe Kissinger method (28103 kJmolminus1) The results obtainedwith KAS and Friedman methods are very close and in goodagreement [32]The kinetic data obtained for pyrolysis of coalare found to agree closely with some of the literature dataHowever the differences observed in the literature data canbe attributed to the fact that the pyrolysis characteristics ofcoal highly depend on the properties of the coal which in turndiffers based on origin of the coal [28 30 31]
The KAS and FWOmethods were originally derived withconstant activation energies so the errors associated withkinetic measurements from methods should be dependent
Activ
atio
n en
ergy
Conversion (x)
KissingerKAS
FWOFriedman
00100
150
200
250
300
350
400
02 04 06 08 10
Figure 8 The activation energy as a function of conversion usingmodel-free isoconversional technique
on the variation of the activation energy with respect toconversion This error does not appear in the Friedmanmethod [33 34] Another advantage that can be attributed toFriedman method is that the activation energies obtained bythe Friedmanmethod are independent of the range of heatingrates which can decrease the systematic error in evaluatingthe activation energy values Thus Friedman method canbe considered to be the best among the four model-freemethods in order to evaluate kinetic parameters for solid-state reactions [33 34] The kinetic parameters obtainedin this study can be useful for pyrolysis and gasificationresearchers to predict kinetic model of coal pyrolysis andoptimization of the process conditions
4 Conclusion
In this study the pyrolysis kinetics of a Canadian lignite coalwas carried out bymeans of thermogravimetric analysis (TG)in the temperature range of 298ndash1173K at six different heatingrates of 1 6 9 12 15 and 18 Kminminus1 under nitrogen atmo-sphere It was found that the main pyrolysis process occurredin the temperature range 450ndash850K In this work kineticstudy and thermal behavior of lignite coal were presentedwhere Arrhenius parameters were determined and comparedthrough four different methods of Kissinger Ozawa KASand Friedman The activation energy is calculated as afunction of conversion by using these methods and is foundto be similar Among these methods Friedman methodwas considered to be the best in order to evaluate kineticparameters for solid-state reactions such as coal pyrolysisMethods such as FWO and KAS are restricted to the use of alinear variation of the temperature and positive heating rateMoreover they are generated based onmathematical approx-imation which can enhance systematic error The advantage
8 International Journal of Chemical Engineering
Table 2 Calculated kinetic parameters for a Canadian lignite coal by three different isoconversional methods
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
Figure 3 Conversion curves for different heating rates for pyrolysisof lignite coal under N
2
atmosphere
factor In theKissingermethod the degree of conversion at thepeak temperature (119879
119898
) is a constant under different heatingrates The kinetic parameters using Kissinger method werefound by linear regression line which is shown in Figure 4The activation energy and preexponential factor extractedfrom the slope and intercept are 281 kJmolminus1 and 261 times1017minminus1 respectively The activation energy and preexpo-nential factor were calculated as a function of conversionby using isoconversional methods of KAS FWO and Fried-man methods The isoconversional plots of these methodsare shown in Figures 5ndash7 respectively Different range ofconversion from 005 to 09 is considered for calculating thekinetic parameters based on isoconversional method Theactivation energies from the slope and preexponential factorsfrom the intercept of three different isoconversional methodswere obtained and listed in Table 2 It can be observed fromTable 2 that the values of activation energies are not similar atdifferent constant extents of conversion because most solid-state reactions are not simple one-stepmechanism and followa complex multistep reaction The thermogravimetric dataanalysis by isoconversional technique may reveal complexityof the solid-state reactions such as coal pyrolysis [14] Itmeansthat in the pyrolysis process of coal the activation energy isa function of conversion Figure 8 shows the dependence ofthe activation energy on extent of conversion The activationenergy rises from about 130 kJmolminus1 at low conversion tonearly 350 kJmolminus1 at 75 conversion and it subsequentlydrops to about 300 kJmolminus1 near the end of reaction Theinitial activation energy valuewas lowdue to cleavage of someweak bonds and elimination of volatile components fromthe coal matrix because at the beginning of the process allthe strong bonds are not cleaved Therefore more activationenergy is required to decompose these stable moleculesWith the progress of pyrolysis process the value of activation
138 140 142 144 146 148
Kissinger
R2 = 09858
y = minus33803x + 36586
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
ln(120573
1000Tm (Kminus1)
T2 m
)
Figure 4 Kissinger plot of lignite coal pyrolysis at different heatingrates
12 14 16 18 20 22
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus95
minus100
minus105
minus110
minus115
minus120
minus125
minus130
minus135
minus140
ln(120573T
2)
1000T (Kminus1)
Figure 5 KAS plots of lignite coal pyrolysis at different values ofconversion
energy increased up to conversion of 75 with breaking ofsome strong covalent linkages For higher conversion valuesabove 75 the activation energy gradually decreases Thereason arises from the fact that during the decompositionprocess at high temperature with high conversion whenmostof the stable bonds are broken less stablemolecules which areeasier to break are present so less energy barrier is requiredfor decomposition at this step and the value of activationenergy decreases with progress of conversionThe arithmeticmeans of the activation energy calculated by KAS FWO andFriedmanmethod are 282 275 and 283 kJmolminus1 respectively
International Journal of Chemical Engineering 7
12 14 16 18 20 22 24
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
minus05
00
05
10
15
20
25
30
35
1000T (Kminus1)
ln 120573
Figure 6 FWO plots of lignite coal pyrolysis at different values ofconversion
12 14 16 18 20 22
minus20
minus25
minus30
minus35
minus40
minus45
minus50
minus55
minus60
minus65
minus70
minus75
ln(dxdt)
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
1000T (Kminus1)
Figure 7 Friedman plots of lignite coal pyrolysis at different valuesof conversion
which are close to average activation energy obtained fromthe Kissinger method (28103 kJmolminus1) The results obtainedwith KAS and Friedman methods are very close and in goodagreement [32]The kinetic data obtained for pyrolysis of coalare found to agree closely with some of the literature dataHowever the differences observed in the literature data canbe attributed to the fact that the pyrolysis characteristics ofcoal highly depend on the properties of the coal which in turndiffers based on origin of the coal [28 30 31]
The KAS and FWOmethods were originally derived withconstant activation energies so the errors associated withkinetic measurements from methods should be dependent
Activ
atio
n en
ergy
Conversion (x)
KissingerKAS
FWOFriedman
00100
150
200
250
300
350
400
02 04 06 08 10
Figure 8 The activation energy as a function of conversion usingmodel-free isoconversional technique
on the variation of the activation energy with respect toconversion This error does not appear in the Friedmanmethod [33 34] Another advantage that can be attributed toFriedman method is that the activation energies obtained bythe Friedmanmethod are independent of the range of heatingrates which can decrease the systematic error in evaluatingthe activation energy values Thus Friedman method canbe considered to be the best among the four model-freemethods in order to evaluate kinetic parameters for solid-state reactions [33 34] The kinetic parameters obtainedin this study can be useful for pyrolysis and gasificationresearchers to predict kinetic model of coal pyrolysis andoptimization of the process conditions
4 Conclusion
In this study the pyrolysis kinetics of a Canadian lignite coalwas carried out bymeans of thermogravimetric analysis (TG)in the temperature range of 298ndash1173K at six different heatingrates of 1 6 9 12 15 and 18 Kminminus1 under nitrogen atmo-sphere It was found that the main pyrolysis process occurredin the temperature range 450ndash850K In this work kineticstudy and thermal behavior of lignite coal were presentedwhere Arrhenius parameters were determined and comparedthrough four different methods of Kissinger Ozawa KASand Friedman The activation energy is calculated as afunction of conversion by using these methods and is foundto be similar Among these methods Friedman methodwas considered to be the best in order to evaluate kineticparameters for solid-state reactions such as coal pyrolysisMethods such as FWO and KAS are restricted to the use of alinear variation of the temperature and positive heating rateMoreover they are generated based onmathematical approx-imation which can enhance systematic error The advantage
8 International Journal of Chemical Engineering
Table 2 Calculated kinetic parameters for a Canadian lignite coal by three different isoconversional methods
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
Figure 6 FWO plots of lignite coal pyrolysis at different values ofconversion
12 14 16 18 20 22
minus20
minus25
minus30
minus35
minus40
minus45
minus50
minus55
minus60
minus65
minus70
minus75
ln(dxdt)
055 00506 01
015
02
025
03
035
04
045
05
065
07
075
08
085
09
1000T (Kminus1)
Figure 7 Friedman plots of lignite coal pyrolysis at different valuesof conversion
which are close to average activation energy obtained fromthe Kissinger method (28103 kJmolminus1) The results obtainedwith KAS and Friedman methods are very close and in goodagreement [32]The kinetic data obtained for pyrolysis of coalare found to agree closely with some of the literature dataHowever the differences observed in the literature data canbe attributed to the fact that the pyrolysis characteristics ofcoal highly depend on the properties of the coal which in turndiffers based on origin of the coal [28 30 31]
The KAS and FWOmethods were originally derived withconstant activation energies so the errors associated withkinetic measurements from methods should be dependent
Activ
atio
n en
ergy
Conversion (x)
KissingerKAS
FWOFriedman
00100
150
200
250
300
350
400
02 04 06 08 10
Figure 8 The activation energy as a function of conversion usingmodel-free isoconversional technique
on the variation of the activation energy with respect toconversion This error does not appear in the Friedmanmethod [33 34] Another advantage that can be attributed toFriedman method is that the activation energies obtained bythe Friedmanmethod are independent of the range of heatingrates which can decrease the systematic error in evaluatingthe activation energy values Thus Friedman method canbe considered to be the best among the four model-freemethods in order to evaluate kinetic parameters for solid-state reactions [33 34] The kinetic parameters obtainedin this study can be useful for pyrolysis and gasificationresearchers to predict kinetic model of coal pyrolysis andoptimization of the process conditions
4 Conclusion
In this study the pyrolysis kinetics of a Canadian lignite coalwas carried out bymeans of thermogravimetric analysis (TG)in the temperature range of 298ndash1173K at six different heatingrates of 1 6 9 12 15 and 18 Kminminus1 under nitrogen atmo-sphere It was found that the main pyrolysis process occurredin the temperature range 450ndash850K In this work kineticstudy and thermal behavior of lignite coal were presentedwhere Arrhenius parameters were determined and comparedthrough four different methods of Kissinger Ozawa KASand Friedman The activation energy is calculated as afunction of conversion by using these methods and is foundto be similar Among these methods Friedman methodwas considered to be the best in order to evaluate kineticparameters for solid-state reactions such as coal pyrolysisMethods such as FWO and KAS are restricted to the use of alinear variation of the temperature and positive heating rateMoreover they are generated based onmathematical approx-imation which can enhance systematic error The advantage
8 International Journal of Chemical Engineering
Table 2 Calculated kinetic parameters for a Canadian lignite coal by three different isoconversional methods
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
of the Friedman method is that it is free of mathematicalapproximations and is not restricted to the use of a linearvariation of the heating rate Experimental results showedthat values of kinetic parameters were almost the sameand in good agreement The isoconversional technique givescomparably reliable predictions of reaction rates comparedto the more traditional model-fitting There is very littleinformation regarding pyrolysis of coal itself based onmodel-free methods The results can provide useful information forpyrolysis researchers in order to predict kinetic model of coalpyrolysis and optimization of the process conditions
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] M Balat ldquoCoal in the global energy scenerdquo Energy Sources PartB Economics Planning and Policy vol 5 no 1 pp 50ndash62 2009
[2] T Kabe A Ishihara W Qian I P Sutrisna and Y KabeCoal and Coal-Related Compounds Structures Reactivity andCatalytic Reactions Elsevier New York NY USA 2004
[3] S Vasireddy B Morreale A Cugini C Song and J J SpiveyldquoClean liquid fuels from direct coal liquefaction chemistrycatalysis technological status and challengesrdquo Energy and Envi-ronmental Science vol 4 no 2 pp 311ndash345 2011
[4] M R Khan Advances in Clean Hydrocarbon Fuel ProcessingScience and Technology Woodhead Pubishing Series in Energy2011
[5] M Brown Introduction to Thermal Analysis Techniques andApplications Springer 2001
[6] M Hook and K Aleklett ldquoA review on coal-to-liquid fuels andits coal consumptionrdquo International Journal of Energy Researchvol 34 no 10 pp 848ndash864 2010
[7] N Punsuwan and C Tangsathitkulchai ldquoProduct characteri-zation and kinetics of biomass pyrolysis in a three-zone free-fall reactorrdquo International Journal of Chemical Engineering vol2014 Article ID 986719 10 pages 2014
[8] M Varol A T Atimtay B Bay and H Olgun ldquoInvestigationof co-combustion characteristics of low quality lignite coalsand biomass with thermogravimetric analysisrdquoThermochimicaActa vol 510 no 1-2 pp 195ndash201 2010
[9] HC Howard Chemistry of Coal Utilization Second Supplemen-tary Volume John Wiley New York NY USA 1963
[10] G J Lawson Differential Thermal Analysis FundamentalAspects Academic Press New York NY USA 1970
[11] D B Anthony and J B Howard ldquoCoal devolatilization andhydrogastificationrdquo AIChE Journal vol 22 no 4 pp 625ndash6561976
[12] V V HathiThermal and kinetic analysis of the pyrolysis of coals[PhD thesis] University of Oklahoma Norman Okla USA1978
[13] A Khawam Application of solid-state kinetics to desolvationreactions [PhD thesis] University of Iowa Iowa City IowaUSA 2007
[14] S Scaccia ldquoTG-FTIR and kinetics of devolatilization of Sulciscoalrdquo Journal of Analytical and Applied Pyrolysis vol 104 pp95ndash102 2013
[15] S S Idris NA RahmanK Ismail A B Alias Z A Rashid andM J Aris ldquoInvestigation on thermochemical behaviour of lowrank Malaysian coal oil palm biomass and their blends duringpyrolysis via thermogravimetric analysis (TGA)rdquo BioresourceTechnology vol 101 no 12 pp 4584ndash4592 2010
International Journal of Chemical Engineering 9
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
[16] G Jiang D J Nowakowski and A V Bridgwater ldquoA systematicstudy of the kinetics of lignin pyrolysisrdquo Thermochimica Actavol 498 no 1-2 pp 61ndash66 2010
[17] N Sbirrazzuoli L Vincent A Mija and N Guigo ldquoIntegraldifferential and advanced isoconversional methods complexmechanisms and isothermal predicted conversion-time curvesrdquoChemometrics and Intelligent Laboratory Systems vol 96 no 2pp 219ndash226 2009
[18] J Zsako ldquoKinetic analysis of thermogravimetric datardquo TheJournal of Physical Chemistry vol 72 no 7 pp 2406ndash2411 1968
[19] K Miura ldquoNew and simple method to estimate f(E) and k0(E)in the distributed activation energy model from three sets ofexperimental datardquo Energy and Fuels vol 9 no 2 pp 302ndash3071995
[20] A O Aboyade M Carrier E L Meyer J H Knoetze and JF Gorgens ldquoModel fitting kinetic analysis and characterisationof the devolatilization of coal blends with corn and sugarcaneresiduesrdquoThermochimica Acta vol 530 pp 95ndash106 2012
[21] S-L Niu K-H Han and C-M Lu ldquoCharacteristic of coalcombustion in oxygencarbon dioxide atmosphere and nitricoxide release during this processrdquo Energy Conversion andManagement vol 52 no 1 pp 532ndash537 2011
[22] H B Vuthaluru ldquoInvestigations into the pyrolytic behaviour ofcoalbiomass blends using thermogravimetric analysisrdquo Biore-source Technology vol 92 no 2 pp 187ndash195 2004
[23] C A Ulloa A L Gordon and X A Garcıa ldquoThermogravimet-ric study of interactions in the pyrolysis of blends of coal withradiata pine sawdustrdquo Fuel Processing Technology vol 90 no 4pp 583ndash590 2009
[24] H Haykiri-Acma and S Yaman ldquoSynergy in devolatilizationcharacteristics of lignite and hazelnut shell during co-pyrolysisrdquoFuel vol 86 no 3 pp 373ndash380 2007
[25] M Karthikeyan W Zhonghua and A S Mujumdar ldquoLow-rank coal drying technologiesmdashcurrent status and new devel-opmentsrdquo Drying Technology vol 27 no 3 pp 403ndash415 2009
[26] S A Channiwala and P P Parikh ldquoA unified correlation forestimating HHV of solid liquid and gaseous fuelsrdquo Fuel vol81 no 8 pp 1051ndash1063 2002
[27] H E Kissinger ldquoReaction kinetics in differential thermalanalysisrdquo Analytical Chemistry vol 29 no 11 pp 1702ndash17061957
[28] A O Aboyade J F Gorgens M Carrier E L Meyer and JH Knoetze ldquoThermogravimetric study of the pyrolysis char-acteristics and kinetics of coal blends with corn and sugarcaneresiduesrdquo Fuel Processing Technology vol 106 pp 310ndash320 2013
[29] Y Guldogan T Durusoy and T Bozdemir ldquoEffects of heatingrate and particle size on pyrolysis kinetics of gediz ligniterdquoEnergy Sources vol 24 no 8 pp 753ndash760 2002
[30] S Sharma and A K Ghoshal ldquoStudy of kinetics of co-pyrolysisof coal and waste LDPE blends under argon atmosphererdquo Fuelvol 89 no 12 pp 3943ndash3951 2010
[31] M Gunes and S K Gunes ldquoDistributed activation energymodel parameters of some Turkish coalsrdquo Energy Sources PartA Recovery Utilization and Environmental Effects vol 30 no16 pp 1460ndash1472 2008
[32] L Gasparovic Z Korenova and Lrsquo Jelemensky ldquoKinetic studyof wood chips decomposition by TGArdquo Chemical Papers vol64 no 2 pp 174ndash181 2010
[33] S Vyazovkin and N Sbirrazzuoli ldquoIsoconversional kineticanalysis of thermally stimulated processes in polymersrdquoMacro-molecular Rapid Communications vol 27 no 18 pp 1515ndash15322006
[34] W Wu J Cai and R Liu ldquoIsoconversional kinetic analysis ofdistributed activation energy model processes for pyrolysis ofsolid fuelsrdquo Industrial and Engineering Chemistry Research vol52 no 40 pp 14376ndash14383 2013
