- Email: [email protected]
Kinetic parameters determination using optimization approach in integrated catalytic adsorption steam gasification for hydrogen production
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Kinetic parameters determination using optimization approach in integrated catalytic adsorption steam gasification for hydrogen production Zakir Khan a,*, Abrar Inayat b, Suzana Yusup b, Murni M. Ahmad b,1 a
Biomass Conversion Research Centre, Department of Chemical Engineering, COMSATS Institute of Information Technology, Lahore, 54000, Pakistan b Biomass Processing Laboratory, Center for Biofuel and Biochemical, Mission Oriented Research of Green Technology, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, Tronoh, 31750, Perak, Malaysia
article info
abstract
Article history:
Integrated catalytic adsorption (ICA) gasification provides an efficient mean to produce
Received 29 March 2015
hydrogen rich gas. This article presents the prospect of ICA steam gasification of palm
Received in revised form
kernel shell. The effect of temperature, steam to biomass ratio and adsorbent to biomass
11 May 2015
are investigated for H2, CO, CO2 and CH4 composition to determine kinetic parameters by
Accepted 12 May 2015
minimizing the error between experimental and modelling results. Based on the evaluated
Available online 6 June 2015
kinetic parameters, the model predicts the product gas composition for the effect of temperature, steam to biomass ratio and adsorbent to biomass ratio. A significant fitting of
Keywords:
model predicted values to the experimental results is achieved. Furthermore, it is also
Kinetic parameter
found that the water gas shift reaction is non-spontaneous and far away from the equi-
Optimization
librium at a temperature range of 600
Hydrogen production
adsorption reaction.
CO2 adsorption
Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Ce675
C which may be due to strong CO2
reserved.
Palm kernel shell
Introduction Fossil fuel energy dependency causes numerous environmental problems such as greenhouse effect, ozone layer depletion and acid rain. Due to associated problems with fossil fuel, the search for alternative clean, sustainable and environmental friendly energy sources should be intensified. Hydrogen as an energy carrier comprises of numerous
advantages over other conventional energy carriers. Hydrogen combustion provides more energy (lower heating value based on mass basis) than that of methane, gasoline and coal [1]. In addition, it is a clean fuel as the combustion of hydrogen produces only water as by-product. Biomass is a promising source among the renewable sources to produce clean and renewable hydrogen due to its net zero carbon, and low NOx and SOx emissions in the product gas. Among thermal conversion processes, gasification is
* Corresponding author. Tel.: þ92 42 111 001 007x150; fax: þ92 42 9203100. E-mail address: [email protected] (Z. Khan). 1 Formerly affiliated. http://dx.doi.org/10.1016/j.ijhydene.2015.05.069 0360-3199/Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
considered as most potential process for hydrogen production. Biomass steam gasification is recognized as an efficient process to produce hydrogen rich gas. Use of catalyst in biomass steam gasification is attained prime interest due to enhancement of H2 content up to more than 60 vol% in product gas [2]. Catalyst reduces tar content significantly in product gas and improves the gas quality [3]. Besides, in situ CO2 adsorption in biomass gasification increases hydrogen content up to 75 vol% (dry basis) which has been reported 40 vol% (dry basis) in conventional biomass gasification [4]. CO2 adsorption process is an exothermic reaction thus it provides heat for endothermic gasification reactions and reduces overall energy requirement for the process in the gasifier [5]. Addition of CO2 adsorbent allows gasification process to take place at temperature less than 800 C [5e7]. Biomass gasification is a mixture of complex reactions. Numerous models are reported to simulate biomass gasification reactions. These models are based on different aspects of the process such as kinetic, equilibrium and hydrodynamics of different types of reactors [8]. The modeling approaches for biomass gasification can be divided into kinetic modeling and equilibrium modeling [9]. A kinetic model predicts the product gas composition and gas yield based on the kinetics of main reactions involved in the process. At given operating conditions, kinetic model is capable to predict product gas profiles and overall gasification efficiency of the process. Limited studies have been conducted on the modeling and simulation of hydrogen production via biomass steam gasification with in-situ CO2 capture. Florin and Harris [10] developed a thermodynamic equilibrium model to investigate the effect of fundamental process parameters such as temperature, steam to biomass ratio, adsorbent to biomass ratio and pressure on the hydrogen production from methyl cellulose using concept of gasification and combustion steps in separate reactors. The model prediction was also compared and validated with € ll and experimental work taken from the literature [11]. Pro Hofbauer [12] presented thermodynamic equilibrium model for hydrogen rich gas production by selective CO2 transport in dual fluidized bed system. The CaO/CaCO3 system was used as bed material for selective CO2 transport from gasification to the combustion reactor by carbonation and calcination reactions. Mahishi et al. [5] developed an equilibrium model for biomass steam gasification using CaO as an adsorbent. Ethanol was taken as the model compound for the steam gasification using Gibbs free energy minimization approach. Very few kinetic models have been reported for biomass steam gasification with in-situ CO2 adsorbent. Inayat et al. [13] developed a reaction kinetic model for oil palm empty fruit bunch (EFB) to produce hydrogen using sum of squared technique in MATLAB. The CO2 adsorbent reaction along with water gas shift, steam methane reforming, char gasification, methanation and Boudouard reactions were considered to simulate the process. The reaction kinetics data was taken from the literature [14e19]. They predicted the increment of hydrogen composition with temperature and steam to biomass ratio. For similar reactions in the subject study [13], Yunus et al. [20] simulated kinetic model in iCON software and predicted the effect of temperature, steam to biomass ratio
8825
and adsorbent to biomass ratio. Sreejit et al. [21] carried out the process simulation of biomass air-steam gasification with CO2 sorption through kinetic approach using kinetics data from the literature [13,14,22]. Biomass gasification processes include drying; pyrolysis, char gasification and homogeneous gas phase reactions were considered. They predicted the increase of hydrogen composition and heating values with CO2 sorption in the process. In further kinetic study by Sreejit et al. [23], tar cracking and reforming reactions were also considered along with other reactions in the previous study [21]. For kinetic parameters determination, Inayat et al. [24] carried out optimization approach in MATLAB to produce hydrogen from oil palm wastes steam gasification with in-situ CO2 sorption. They evaluated kinetic parameters by fitting experimental data to kinetic model, and minimized least squared error between the experimental data and model predictions. The experimental data was taken from the previous work [24]. To date, the kinetic studies of biomass steam gasification are carried out by using in-situ CO2 sorption in the process. The present work extends the kinetic study to the effect of catalyst and CO2 adsorbent together in a pilot scale fluidized bed reactor for biomass steam gasification. The present study investigates the kinetic parameter evaluation using optimization approach for hydrogen production in integrated catalytic adsorption (ICA) steam gasification of palm kernel shell. The kinetic parameters are evaluated by minimizing the residual error between model predictions and experimental values. The model prediction of product gas composition is then generated via evaluated kinetic parameters and compared with experimental data. In addition, thermodynamic parameters i.e. Gibbs free energy and equilibrium constant for water gas shift reactions are evaluated and compared with the previously reported literature.
Methodology Reaction kinetics model The kinetic modeling approach was carried out by considering six reactions occurring in the gasification process. The product gas from these reactions was mainly consisted of H2, CO, CO2 and CH4. These reactions are presented in Table 1. Among these reactions, char gasification, methanation and boudouard reactions were modified by replacing C (carbon) in the chemical formula of palm kernel shell. This approach was adopted due to its applicability in biomass steam gasification with in-situ CO2 adsorbent [25]. The chemical formula of palm kernel shell is C4.15H5.68O2.71. The following assumptions were made for the kinetic model: The fluidized bed was under isothermal conditions, temperature distribution was homogeneous throughout the bed and operation was at atmospheric pressure [14,26]. The simple approach of first order kinetics with respect to reacting species was selected, as the reactions are assumed to take place under isothermal and constant reactor
8826
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Table 1 e Reaction scheme for kinetic parameters determination. No
Name
Reaction
1 2 3 4 5 6
Char gasification Methanation Boudouard Methane reforming Water gas shift Carbonation
C4:15 H6:13 O2:73 þ 1:44H2 O#4:25H2 þ 4:15CO C4:15 H6:13 O2:73 þ 8:2H2 #4:15CH4 þ 2:71H2 O C4:15 H6:13 O2:73 þ CO2 #2:25H2 þ 4:15CO þ 0:56H2 O CH4 þ H2 O# CO þ 3H2 CO þ H2 O#CO2 þ H2 CO2 þ CaO#CaCO3
a
DH (kJ mol1) 131.5a 74.8a 172a 206 41 170.5
Reaction enthalpy based on the reacting carbon in the form of biomass char.
volume conditions. The rate of reaction i of reactant A and B is represented as: ri ¼ ki CA CB
(1)
Here ki (1/s) is Arrhenius constant and represented by following equation: Ei
ki ¼ Ai expRT
(2)
where Ai is the frequency factor or pre-exponential factor (s1), Ei is activation energy (J mol1), R is universal gas constant (J mol1 K1) and T is temperature (K). Biomass devolatilization was an instantaneous process [27]. Tar formation in the product gas was negligible. This assumption was taken due to the double effect of CaO (adsorbent) [21] and Ni catalyst for active tar removal in the product gas [28]. Bed hydrodynamics was insensitive to the reactor performance. This was considered due to the assumption of perfect mixing and uniform temperature distribution in the fluidized bed gasifier [29]. Gaseous product mainly consisted of H2, CO, CO2 and CH4. The volumetric flow rate of individual gas component was defined by Ref. [14]: RH2 ¼ 4:25r1 8:2r2 þ 2:25r3 þ 3r4 þ r5 þ r05
(3)
RCO ¼ 4:15r1 þ 4:15r2 þ r4 r5 þ r05
(4)
RCO2 ¼ r5 r05 r6 r3
(5)
RCH4 ¼ 4:15r2 r4
(6)
the experimental values (yexp). The residual error was described by: N X yexp ymod residual error ¼ yexp i¼1
!2
Fig. 1 shows the kinetic modeling approach used in the present study. The kinetic parameters evaluated were then used as input variables to calculate the volumetric rate of individual components in the kinetic model. The model results were then evaluated and compared with experimental data. The deviation between ymod and yexp was carried out using sum squared method [24]: RSS ¼
N X yexp ymod yexp i¼1
MRSS ¼
!2
RSS N
Mean error ¼
(8)
(9) pffiffiffiffiffiffiffiffiffiffiffiffiffi MRSS
Mean errorrelative ¼
Mean error yexp
(10)
where RSS is residual sum of squared, MRSS is mean of RSS, N is the total number of points, ӯexp is mean experimental value of an individual gas component over three data points, Mean error and Mean errorrealtive represent absolute and relative mean errors. The MATLAB fmincon function was used to carry out the nonlinear programming (NLP). Nonlinear programming is the
where r5 and r5 are rate of forward and reverse water gas shift reactions. The numerical values multiplied with r1 to r6 in volumetric rate of the individual gas component are the stoichiometric coefficients that appeared in reactions 1 e 6 (Table 1).
Kinetic parameters evaluation The kinetic parameter evaluation was carried out by minimizing the residual error between the model values (ymod) and
(7)
Fig. 1 e Flow chart of kinetic model using error minimization approach.
8827
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
technique used in the mathematics to solve a set of unknown variables based on the objective function to be minimized or maximized, where some of the nonlinear functions are present [30]. Product gas compositions from fluidized bed gasification system were considered under the effect of temperature, steam to biomass and adsorbent to biomass ratios. Three experimental points were taken for each process variables. Twelve parameters were estimated through nonlinear regression by considering E and A for all six reactions (Table 1). RSS was considered as an objective function of the optimization process (Equation (8)). Mean error was then evaluated using RSS values (Equation (10)). The present model can be advantageous for extensive range of process variables. Moreover, the model can also be applied for scaling up the biomass gasification processes.
Thermodynamic parameters such as equilibrium constant and Gibbs free energy are calculated for water gas shift reaction in the present study. Several studies [9,11] found that the water gas shift seems to reach the equilibrium, and control the gas phase kinetics in biomass steam gasification in fluidized bed reactor. Equilibrium constants and Gibbs free energy are evaluated based on the concentration of product gas at three different temperature of 600 C, 675 C and 750 C at steam to biomass ratio of 2.0 (wt/wt), adsorbent to biomass ratio of 1.0 (wt/wt) and catalyst to biomass ratio of 0.1 (wt/wt). Equilibrium constant for the water gas reaction (reaction 5) can be calculated as [31]: ½CO2 ½H2 ½CO½H2 O
(11)
The quantity in the bracket shows the concentration (mol/ m3) of product (CO2, H2) and reactant (CO, H2O). Once Ke is determined, change in Gibbs free energy is calculated [32] to check spontaneous and nonspontaneous nature of the reaction at given operating conditions: DG+ ¼ RT ln Ke
Moisture (%) Proximate analysis (wt. % dry basis) Volatile matter Fixed carbon Ash content Ultimate analysis (wt. % dry basis) C H N S O (by difference) Higher heating value (HHV)
9.61 80.92 14.67 4.31 49.74 5.68 1.02 0.27 43.36 18.46 MJ/kg
Table 3 e Physical properties of bed material.
Thermodynamic parameters evaluation
Ke ¼
Table 2 e Proximate and ultimate analysis of palm kernel shell.
(12)
R and T represent universal gas constant (J/mol.K) and temperature (K), respectively.
Experimental Palm kernel shell was utilized as the feedstock. The local palm kernel shell was obtained from My 4-Seasons International Sdn. Bhd, Selangor, Malaysia. The palm kernel shell received had initial diameter in the range of 0.1 mme4.0 mm. Then, the drying process was performed under sunlight for 3e4 h to remove excess moisture content. The dried palm kernel shell was further sieved into required size of 1.0 mme2.0 mm. Subsequently, the samples were stored in air tight bags. The proximate and ultimate analysis of biomass is shown in Table 2. In addition, quicklime (CaO), supplied by Universal Lime Sdn. Bhd, Malaysia was used as a bed material and as an adsorbents to adsorb CO2 in the product gas. The sample was grinded and sieved to particle size of 0.15e0.25 mm. The physical properties of the bed material are given in Table 3.
Particle density (kg/m3) Bulk density (kg/m3) Chemical composition (wt. %) CaO MgO SiO2 Fe2O3 Other metal oxide (MnO, CuO, SrO, ZnO)
3053 1047 93.32 4.24 0.95 0.23 1.0
The catalyst was pure nickel (Ni) powder (99.5% purity) with particle size of 10 mm. Ni catalyst was purchased from Merck kGaA, Germany. For the entire gasification process, Ni catalyst was mixed with the palm kernel shell and was introduced to the gasifier through a biomass feeding system. Fig. 2 shows a schematic diagram of pilot-scale fluidized bed gasification system used for the experimental runs that was operated under atmospheric pressure. The unit mainly consists of fluidized bed reactor with external heaters, biomass feeder and steam generator with superheater, cyclone solid separator, wet scrubber, water separator and gas analyzing system. The diameter and height of the fluidized bed reactor are 0.15 m and 2.5 m, respectively. A perforated type distributor plate was used. The gasifier was operated at superficial velocity of 0.21 m/s (4 times the minimum fluidization velocity). The fluidized bed gasifier was continuously fed with biomass at 1.0e1.8 kg/h from the biomass feeding system. The cooling water jacket was provided to avoid biomass decomposition in the feeding line. Nitrogen (N2) was used to transfer the biomass into the gasifier in order to avoid any back flow. Saturated steam was provided by the steam generator and was further heated from 250 C to 300 C in the superheater prior its injection into the gasifier. Before the start of each experiment, N2 gas was purged into the system to remove any entrapped gases. The product gas from the outlet of the gasifier was passed through the cyclone to separate entrained solid particles. The product gas was then passed through the scrubber prior to a separator unit to remove traces of water in the product gas. Gas sampling point was located at the exit of the water separator unit. At the same point, the volumetric flow rate was measured. All experiments were run for 60 min. The product gases such as CO2, CO and CH4 were analyzed by using a Gas Chromatography (Teledyne 7500,
8828
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 2 e Schematic of pilot scale fluidized bed gasification system.
Teledyne Analytical Instrument) with Infrared (IR) type detector. H2 and N2 were detected by Gas Chromatograph utilizing a Molecular Sieve 5A column (Teledyne 4060, Teledyne Analytical Instrument) with a Thermal Conductive Detector (TCD). The product gas was measured for every 6 min interval. Three process variables such as temperature (T), steam to biomass ratio (S/B) and adsorbent to biomass ratio (A/B) were tested in the present study. Fluidized bed gasification temperature was tested from 600 C to 675 C. Steam to biomass ratio was varied from 1.5 to 2.5 (wt/wt). Adsorbent (bed material) to biomass ratio was varied from 0.5 to 1.5 (wt/wt). For all experiments, catalyst to biomass ratio of 0.1 (wt/wt) was considered.
Results and discussions Kinetic parameters The kinetic parameter evaluation was carried out using product gas composition. The kinetic parameters were evaluated by minimizing the residual error between the experimental data and model prediction. Table 4 summarizes all the kinetic parameters for the main reactions occurring in ICA steam gasification process. Among all the reactions, char steam gasification has high activation energy. This is due to the endothermic nature of the reaction and requires
Table 4 e Evaluated kinetic parameters. No
1 2 3 4 5 6
Reaction
Char gasification Methanation Boudouard Methane steam reforming Water gas shift Carbonation
Frequency factor, A (1/s)
Activation energy, E (kJ/mol)
104 104 103 104
999.95 0.98 16.84 21.57
0.4 101 2.81 102
22.38 17.57
3.32 3.19 1.7 3.19
temperature >800 C for effective conversion rates in fluidized bed reactors [33].
Effect of gasification temperature Fig. 3 shows the effect of temperature at 600 C, 675 C and 750 C on product gas composition. The concentration profiles of H2, CO, CO2 and CH4 are plotted. As observed, the model prediction is in good agreement with the experimental results. The coefficient of determination (R2) is found to be higher than 0.91. Overall, H2 content increases when temperature is elevated from 600 C to 675 C while concentration of CO2 and CO decreases. This is due to active CO2 adsorption which drives the water gas shift reaction towards forward direction and enhances hydrogen production. This is in line with the previous work reported by Marquard et al. [7]. At 750 C, H2 content reduces in the product gas due to the reverse carbonation reaction that promotes high concentration of CO2 in the product gas. This observation is further verified by a high CO content in the product gas at 750 C. These results are in general agreement with the other studies [5]. However, the opposite trends of CO composition observed at high temperature 675e750 C is may be due to the consideration of forward and backward water gas shift reactions in the present model. Meanwhile, CH4 content gradually decreases from 600 to 750 C which can be elucidated by the endothermic nature of steam methane reforming. Table 5 provides the relative mean error calculated for individual product gas. Low mean error infers good fit of the modeling results towards the experimental values.
Effect of steam to biomass ratio Fig. 4 illustrates the effect of S/B ratio on product gas composition for both experiments and modeling results. The modeling results are in good agreement with experimental values at 675 C, A/B ratio of 1.0 and catalyst to biomass ratio of 0.1 at varying S/B ratio from 1.5 to 2.5. The R2 is found to be higher than 0.90. The addition of steam promotes methane reforming, water gas shift and char gasification reactions in
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 3 e Effect of temperature on product gas composition; modeling (
Table 5 e Relative mean error of product gas composition. Gas component
H2 CO2 CO CH4
Mean error relative Temperature
Steam to biomass ratio
Adsorbent to biomass ratio
0.00005 0.09980 0.00023 0.00250
0.00054 0.06010 0.01877 0.02624
0.00017 0.11612 0.00024 0.00208
8829
) and experiment ( ).
the forward direction to produce more H2 in product gas. Likewise, the increase in H2 content can be inferred by decrease in concentration of CO and CH4 and increase in concentration of CO2 content in product gas. This is consistent with the results reported by Mahishi et al. [5] for steam gasification with in-situ CO2 adsorbent in fluidized bed reactor. However, varying S/B ratio from 2.0 to 2.5 is not found to increase concentration of H2 significantly. This is due to excessive amount of steam present in the system. This observation is supported by the results of Han et al. [34] for
Fig. 4 e Effect of steam to biomass ratio on product gas composition; modelling (
) and experiment ( ).
8830
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 5 e Effect of adsorbent to biomass ratio on product gas composition; modeling (
saw dust steam gasification with CO2 adsorbent in fluidized bed reactor. It can be concluded that the S/B ratio of 2.0 is the optimum to produce hydrogen rich gas in ICA steam gasification process. A good agreement can be seen between experimental and model values based on the mean error values (Table 4).
Effect of adsorbent to biomass ratio The A/B ratio is tested for product gas distribution as shown in Fig. 5. The model results fits well to the experimental data conducted at 675 C, S/B ratio of 2.0, and catalyst to biomass ratio of 0.1 by varying A/B ratio from 0.5 to 1.5. The R2 is found be higher than 0.91. Furthermore, the low mean errors between modeling and experimental values are observed for all the product gases as given in Table 4. The product gases i.e. H2, CO, CO2 and CH4 show similar trends to the experimental data. The addition of adsorbent increases the H2 content in the product gas. The excess amount of adsorbent (CaO) captured CO2 in the process which subsequently enhances the activity of water gas shift and promotes all the gasification and reforming reactions towards enhanced H2 generation. Beside, CO2 content slightly increases at A/B ratio of 1.5 which clearly indicates high activity of water gas shift reaction. This can also be explained by the increase of CO content at A/B ratio of 1.5. Han et al. [34] reported similar increase of CO2 with respect to A/B ratio carrying sawdust steam gasification with CO2 adsorption in a fluidized bed reactor.
) and experiment ( ).
equilibrium constant increases with increasing temperature. These results infer that the concentration of product gas i.e. H2 and CO2 is increasing with increasing temperature. This can be justified under the effect of temperature on individual gas composition as shown in Fig. 3. In the present study, H2 composition increases at temperature range of 600 Ce675 C while no CO2 composition is depicted at this temperature. At 750 C, H2 composition decreases but CO2 increases due to reverse carbonation reaction. The positive Gibbs free energy shows that the water gas shift reaction is non-spontaneous at given temperature range (Table 6). To assess the results further, equilibrium constant observed in the present study are plotted along with theoretical equilibrium constant [35] and experimental equilibrium constant observed by Herguido et al. [11] at 600 C, 700 C and 750 C for steam gasification in fluidized bed reactor as shown in Fig. 6. The experimental equilibrium constant from the present study increases with increasing temperature which is in a good agreement with that of Herguido et al. [11]. Similar finding is also observed by other researchers [36]. Comparative study shows that lower experimental equilibrium constant values in the present study may be due to the presence of CO2 adsorbent which captures most of the CO2 in the product gas thus reduces overall equilibrium constant values for water gas shift reaction.
Table 6 e Effect of temperature on equilibrium constant and Gibbs free energy. Temperature ( C)
Thermodynamic parameters
Parameter
Equilibrium constants and Gibbs free energy for water gas shift reaction are listed in Table 6. The result shows that the
Equilibrium constant () Gibbs free energy (KJ/mol)
600
675
750
0.04 23.50
0.14 15.77
0.40 8.81
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 6 e Effect of reactor temperature on the equilibrium constant of water gas shift reaction.
Conclusions The kinetic parameters determination for six major reactions involved in ICA steam gasification carried out at a pilot scale fluidized bed gasifier was presented. The kinetic parameters determination was carried out through optimization approach by minimizing error between experimental and modelling values for H2, CO, CO2 and CH4. It was predicted that the H2 content gradually increased by increasing steam to biomass and adsorbent to biomass ratios. However, H2 content increased from 600 C 675 C while later on decreased at 750 C. The concentration decreased due to presence of reverse carbonation reaction in the process. Furthermore, the study of thermodynamic parameters inferred that the water gas shift was a nonspontaneous reaction at a temperature range between 600 C and 750 C. Furthermore, it was observed that low equilibrium constant of 0.014 and 0.14 at 600 C and 675 C, respectively, provided the evidence of strong CO2 adsorption reaction in ICA steam gasification process.
references
[1] Khan Z, Yusup S, Ahmad MM, Rashidi NA. Integrated catalytic adsorption (ICA) steam gasification system for enhanced hydrogen production using palm kernel shell. Int J Hydrogen Energy 2014;39:3286e93. S, Jand N, Foscolo PU. Catalytic gasification of [2] Rapagna biomass to produce hydrogen rich gas. Int J Hydrogen Energy 1998;23:551e7. [3] Hu G, Xu S, Li S, Xiao C, Liu S. Steam gasification of apricot stones with olivine and dolomite as downstream catalysts. Fuel Process Technol 2006;87:375e82. [4] Pfeifer C, Bernhard P, Hofbauer Hermann. In-situ CO2absorption in a dual fluidized bed biomass steam gasifier to produce a hydrogen rich syngas. Int J Chem React Eng 2007;5(A9):1e13. [5] Mahishi MR, Sadrameli MS, Vijayaraghavan S, Goswami DY. A novel approach to enhance the hydrogen yield of biomass gasification using CO2 sorbent. J Eng Gas Turbines Power 2008;130:1e8.
8831
[6] Xu G, Murakami T, Suda T, Kusama S, Fujimori T. Distinctive effects of cao additive on atmospheric gasification of biomass at different temperatures. Ind Eng Chem Res 2005;44:5864e8. € llenstedt T, Sichler P, Specht M, Michel M, [7] Marquard-Mo Berger R, Hein KRG, et al. New approch for biomass gasification to hydrogen 2nd world conference and technology exhibition on biomass for Energy. Rome, Italy: Industry and Climate Protection; 2004. [8] Basu P, Kaushal P. Modeling of pyrolysis and gasification of biomass in fluidized beds: a review. Chem Prod Process Model 2009;4(A21):1e45. [9] Gordillo ED, Belghit A. A two phase model of high temperature steam-only gasification of biomass char in bubbling fluidized bed reactors using nuclear heat. Int J Hydrogen Energy 2011;36:374e81. [10] Florin NH, Harris AT. Hydrogen production from biomass coupled with carbon dioxide capture: the implications of thermodynamic equilibrium. Int J Hydrogen Energy 2007;32:4119e34. [11] Herguido J, Corella J, Gonzalez-Saiz J. Steam gasification of lignocellulosic residues in a fluidized bed at a small pilot scale. Effect of the type of feedstock. Ind Eng Chem Res 1992;31:1274e82. € ll T, Hofbauer H. H2 rich syngas by selective CO2 removal [12] Pro from biomass gasification in a dual fluidized bed system e process modelling approach. Fuel Process Technol 2008;89:1207e17. [13] Inayat A, Ahmad MM, Yusup S, Mutalib MIA. Biomass steam gasification with in-situ CO2 capture for enriched hydrogen gas production: a reaction kinetics modelling approach. Energies 2010;3:1472e84. [14] Corella J, Sanz A. Modeling circulating fluidized bed biomass gasifiers. A pseudo-rigorous model for stationary state. Fuel Process Technol 2005;86:1021e53. [15] Gonzalez-Saiz J. Advances in biomass gasfication in fluidized bed. Spain: University of Saragossa; 1988. [16] de Souza-Santos ML. Comprehensive modelling and simulation of fluidized bed boilers and gasifiers. Fuel 1989;68:1507e21. [17] Liu H, Gibbs BM. Modeling NH3 and HCN emissions from biomass circulating fluidized bed gasifiers. Fuel 2003;82:1591e604. [18] Simell PA, Hirvensalo EK, Smolander VT, Krause AOI. Steam reforming of gasification gas tar over dolomite with Benzene as a model compound. Ind Eng Chem Res 1999;38:1250e7. [19] Sun P, Grace JR, Lim CJ, Anthony EJ. Determination of intrinsic rate constants of the CaO-CO2 reaction. Chem Eng Sci 2008;63:47e56. [20] Yunus MK, Ahmad MM, Inayat A, Yusup S. Simulation of enhanced biomass gasification for hydrogen production using iCON. In: Proceedings of world academy of science, engineering and technology62; 2010. p. 753e60. [21] Sreejith CC, Haridasan N, Muraleedharan C, Arun P. Allothermal airesteam gasification of biomass with CO2 (carbon dioxide) sorption: performance prediction based on a chemical kinetic model. Energy 2014;69:399e408. [22] Sommariva S, Grana R, Maffei T, Pierucci S, Ranzi E. A kinetic approach to the mathematical model of fixed bed gasifiers. Comput Chem Eng 2011;35:928e35. [23] Sreejith CC, Muraleedharan C, Arun P. Airesteam gasification of biomass in fluidized bed with CO2 absorption: a kinetic model for performance prediction. Fuel Process Technol 2015;130:197e207. [24] Inayat A, Ahmad MM, Mutalib MIA, Yusup S. Optimization approach for kinetics parameters determination for oil palm waste steam gasification with in-situ CO2 capture for hydrogen production. Natl Postgrad Conf NPC 2011:1e6.
8832
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
[25] Inayat A, Ahmad MM, Mutalib MIA, Yusup S. Process modeling for parametric study on oil palm empty fruit bunch steam gasification for hydrogen production. Fuel Process Technol 2012;93:26e34. [26] Shen L, Xiao J, Niklasson F, Johnsson F. Biomass mixing in a fluidized bed biomass gasifier for hydrogen production. Chem Eng Sci 2007;62:636e43. [27] Kaushal P, Abedi J, Mahinpey N. A comprehensive mathematical model for biomass gasification in a bubbling fluidized bed reactor. Fuel 2010;89:3650e61. [28] Kimura T, Miyazawa T, Nishikawa J, Kado S, Okumura K, Miyao T, et al. Development of Ni catalysts for tar removal by steam gasification of biomass. Appl Catal B 2006;68:160e70. [29] Puig-Arnavat M, Bruno JC, Coronas A. Review and analysis of biomass gasification models. Renew Sustain Energy Rev 2010;14:2841e51. [30] Avriel M. Nonlinear programming analysis and methods. Dover Publishing; 2003.
[31] Mountouris A, Voutsas E, Tassios D. Solid waste plasma gasification: equilibrium model development and exergy analysis. Energy Convers Manage 2006;47:1723e37. [32] Loha C, Chatterjee PK, Chattopadhyay H. Performance of fluidized bed steam gasification of biomass e modeling and experiment. Energy Convers Manage 2011;52:1583e8. [33] Basu P. Combustion and gasification in fluidized beds. NY: Taylor & Francis; 2006. [34] Han L, Wang Q, Yang Y, Yu C, Fang M, Luo Z. Hydrogen production via cao sorption enhanced anaerobic gasification of sawdust in a bubbling fluidized bed. Int J Hydrogen Energy 2011;36:4820e9. [35] Twigg MV. Catalyst handbook. 2nd ed. London: Manson Publisher; 1996. [36] Wei L, Xu S, Zhang L, Liu C, Zhu H, Liu S. Steam gasification of biomass for hydrogen-rich gas in a free-fall reactor. Int J Hydrogen Energy 2007;32:24e31.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Kinetic parameters determination using optimization approach in integrated catalytic adsorption steam gasification for hydrogen production Zakir Khan a,*, Abrar Inayat b, Suzana Yusup b, Murni M. Ahmad b,1 a
Biomass Conversion Research Centre, Department of Chemical Engineering, COMSATS Institute of Information Technology, Lahore, 54000, Pakistan b Biomass Processing Laboratory, Center for Biofuel and Biochemical, Mission Oriented Research of Green Technology, Universiti Teknologi PETRONAS, Bandar Seri Iskandar, Tronoh, 31750, Perak, Malaysia
article info
abstract
Article history:
Integrated catalytic adsorption (ICA) gasification provides an efficient mean to produce
Received 29 March 2015
hydrogen rich gas. This article presents the prospect of ICA steam gasification of palm
Received in revised form
kernel shell. The effect of temperature, steam to biomass ratio and adsorbent to biomass
11 May 2015
are investigated for H2, CO, CO2 and CH4 composition to determine kinetic parameters by
Accepted 12 May 2015
minimizing the error between experimental and modelling results. Based on the evaluated
Available online 6 June 2015
kinetic parameters, the model predicts the product gas composition for the effect of temperature, steam to biomass ratio and adsorbent to biomass ratio. A significant fitting of
Keywords:
model predicted values to the experimental results is achieved. Furthermore, it is also
Kinetic parameter
found that the water gas shift reaction is non-spontaneous and far away from the equi-
Optimization
librium at a temperature range of 600
Hydrogen production
adsorption reaction.
CO2 adsorption
Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Ce675
C which may be due to strong CO2
reserved.
Palm kernel shell
Introduction Fossil fuel energy dependency causes numerous environmental problems such as greenhouse effect, ozone layer depletion and acid rain. Due to associated problems with fossil fuel, the search for alternative clean, sustainable and environmental friendly energy sources should be intensified. Hydrogen as an energy carrier comprises of numerous
advantages over other conventional energy carriers. Hydrogen combustion provides more energy (lower heating value based on mass basis) than that of methane, gasoline and coal [1]. In addition, it is a clean fuel as the combustion of hydrogen produces only water as by-product. Biomass is a promising source among the renewable sources to produce clean and renewable hydrogen due to its net zero carbon, and low NOx and SOx emissions in the product gas. Among thermal conversion processes, gasification is
* Corresponding author. Tel.: þ92 42 111 001 007x150; fax: þ92 42 9203100. E-mail address: [email protected] (Z. Khan). 1 Formerly affiliated. http://dx.doi.org/10.1016/j.ijhydene.2015.05.069 0360-3199/Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
considered as most potential process for hydrogen production. Biomass steam gasification is recognized as an efficient process to produce hydrogen rich gas. Use of catalyst in biomass steam gasification is attained prime interest due to enhancement of H2 content up to more than 60 vol% in product gas [2]. Catalyst reduces tar content significantly in product gas and improves the gas quality [3]. Besides, in situ CO2 adsorption in biomass gasification increases hydrogen content up to 75 vol% (dry basis) which has been reported 40 vol% (dry basis) in conventional biomass gasification [4]. CO2 adsorption process is an exothermic reaction thus it provides heat for endothermic gasification reactions and reduces overall energy requirement for the process in the gasifier [5]. Addition of CO2 adsorbent allows gasification process to take place at temperature less than 800 C [5e7]. Biomass gasification is a mixture of complex reactions. Numerous models are reported to simulate biomass gasification reactions. These models are based on different aspects of the process such as kinetic, equilibrium and hydrodynamics of different types of reactors [8]. The modeling approaches for biomass gasification can be divided into kinetic modeling and equilibrium modeling [9]. A kinetic model predicts the product gas composition and gas yield based on the kinetics of main reactions involved in the process. At given operating conditions, kinetic model is capable to predict product gas profiles and overall gasification efficiency of the process. Limited studies have been conducted on the modeling and simulation of hydrogen production via biomass steam gasification with in-situ CO2 capture. Florin and Harris [10] developed a thermodynamic equilibrium model to investigate the effect of fundamental process parameters such as temperature, steam to biomass ratio, adsorbent to biomass ratio and pressure on the hydrogen production from methyl cellulose using concept of gasification and combustion steps in separate reactors. The model prediction was also compared and validated with € ll and experimental work taken from the literature [11]. Pro Hofbauer [12] presented thermodynamic equilibrium model for hydrogen rich gas production by selective CO2 transport in dual fluidized bed system. The CaO/CaCO3 system was used as bed material for selective CO2 transport from gasification to the combustion reactor by carbonation and calcination reactions. Mahishi et al. [5] developed an equilibrium model for biomass steam gasification using CaO as an adsorbent. Ethanol was taken as the model compound for the steam gasification using Gibbs free energy minimization approach. Very few kinetic models have been reported for biomass steam gasification with in-situ CO2 adsorbent. Inayat et al. [13] developed a reaction kinetic model for oil palm empty fruit bunch (EFB) to produce hydrogen using sum of squared technique in MATLAB. The CO2 adsorbent reaction along with water gas shift, steam methane reforming, char gasification, methanation and Boudouard reactions were considered to simulate the process. The reaction kinetics data was taken from the literature [14e19]. They predicted the increment of hydrogen composition with temperature and steam to biomass ratio. For similar reactions in the subject study [13], Yunus et al. [20] simulated kinetic model in iCON software and predicted the effect of temperature, steam to biomass ratio
8825
and adsorbent to biomass ratio. Sreejit et al. [21] carried out the process simulation of biomass air-steam gasification with CO2 sorption through kinetic approach using kinetics data from the literature [13,14,22]. Biomass gasification processes include drying; pyrolysis, char gasification and homogeneous gas phase reactions were considered. They predicted the increase of hydrogen composition and heating values with CO2 sorption in the process. In further kinetic study by Sreejit et al. [23], tar cracking and reforming reactions were also considered along with other reactions in the previous study [21]. For kinetic parameters determination, Inayat et al. [24] carried out optimization approach in MATLAB to produce hydrogen from oil palm wastes steam gasification with in-situ CO2 sorption. They evaluated kinetic parameters by fitting experimental data to kinetic model, and minimized least squared error between the experimental data and model predictions. The experimental data was taken from the previous work [24]. To date, the kinetic studies of biomass steam gasification are carried out by using in-situ CO2 sorption in the process. The present work extends the kinetic study to the effect of catalyst and CO2 adsorbent together in a pilot scale fluidized bed reactor for biomass steam gasification. The present study investigates the kinetic parameter evaluation using optimization approach for hydrogen production in integrated catalytic adsorption (ICA) steam gasification of palm kernel shell. The kinetic parameters are evaluated by minimizing the residual error between model predictions and experimental values. The model prediction of product gas composition is then generated via evaluated kinetic parameters and compared with experimental data. In addition, thermodynamic parameters i.e. Gibbs free energy and equilibrium constant for water gas shift reactions are evaluated and compared with the previously reported literature.
Methodology Reaction kinetics model The kinetic modeling approach was carried out by considering six reactions occurring in the gasification process. The product gas from these reactions was mainly consisted of H2, CO, CO2 and CH4. These reactions are presented in Table 1. Among these reactions, char gasification, methanation and boudouard reactions were modified by replacing C (carbon) in the chemical formula of palm kernel shell. This approach was adopted due to its applicability in biomass steam gasification with in-situ CO2 adsorbent [25]. The chemical formula of palm kernel shell is C4.15H5.68O2.71. The following assumptions were made for the kinetic model: The fluidized bed was under isothermal conditions, temperature distribution was homogeneous throughout the bed and operation was at atmospheric pressure [14,26]. The simple approach of first order kinetics with respect to reacting species was selected, as the reactions are assumed to take place under isothermal and constant reactor
8826
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Table 1 e Reaction scheme for kinetic parameters determination. No
Name
Reaction
1 2 3 4 5 6
Char gasification Methanation Boudouard Methane reforming Water gas shift Carbonation
C4:15 H6:13 O2:73 þ 1:44H2 O#4:25H2 þ 4:15CO C4:15 H6:13 O2:73 þ 8:2H2 #4:15CH4 þ 2:71H2 O C4:15 H6:13 O2:73 þ CO2 #2:25H2 þ 4:15CO þ 0:56H2 O CH4 þ H2 O# CO þ 3H2 CO þ H2 O#CO2 þ H2 CO2 þ CaO#CaCO3
a
DH (kJ mol1) 131.5a 74.8a 172a 206 41 170.5
Reaction enthalpy based on the reacting carbon in the form of biomass char.
volume conditions. The rate of reaction i of reactant A and B is represented as: ri ¼ ki CA CB
(1)
Here ki (1/s) is Arrhenius constant and represented by following equation: Ei
ki ¼ Ai expRT
(2)
where Ai is the frequency factor or pre-exponential factor (s1), Ei is activation energy (J mol1), R is universal gas constant (J mol1 K1) and T is temperature (K). Biomass devolatilization was an instantaneous process [27]. Tar formation in the product gas was negligible. This assumption was taken due to the double effect of CaO (adsorbent) [21] and Ni catalyst for active tar removal in the product gas [28]. Bed hydrodynamics was insensitive to the reactor performance. This was considered due to the assumption of perfect mixing and uniform temperature distribution in the fluidized bed gasifier [29]. Gaseous product mainly consisted of H2, CO, CO2 and CH4. The volumetric flow rate of individual gas component was defined by Ref. [14]: RH2 ¼ 4:25r1 8:2r2 þ 2:25r3 þ 3r4 þ r5 þ r05
(3)
RCO ¼ 4:15r1 þ 4:15r2 þ r4 r5 þ r05
(4)
RCO2 ¼ r5 r05 r6 r3
(5)
RCH4 ¼ 4:15r2 r4
(6)
the experimental values (yexp). The residual error was described by: N X yexp ymod residual error ¼ yexp i¼1
!2
Fig. 1 shows the kinetic modeling approach used in the present study. The kinetic parameters evaluated were then used as input variables to calculate the volumetric rate of individual components in the kinetic model. The model results were then evaluated and compared with experimental data. The deviation between ymod and yexp was carried out using sum squared method [24]: RSS ¼
N X yexp ymod yexp i¼1
MRSS ¼
!2
RSS N
Mean error ¼
(8)
(9) pffiffiffiffiffiffiffiffiffiffiffiffiffi MRSS
Mean errorrelative ¼
Mean error yexp
(10)
where RSS is residual sum of squared, MRSS is mean of RSS, N is the total number of points, ӯexp is mean experimental value of an individual gas component over three data points, Mean error and Mean errorrealtive represent absolute and relative mean errors. The MATLAB fmincon function was used to carry out the nonlinear programming (NLP). Nonlinear programming is the
where r5 and r5 are rate of forward and reverse water gas shift reactions. The numerical values multiplied with r1 to r6 in volumetric rate of the individual gas component are the stoichiometric coefficients that appeared in reactions 1 e 6 (Table 1).
Kinetic parameters evaluation The kinetic parameter evaluation was carried out by minimizing the residual error between the model values (ymod) and
(7)
Fig. 1 e Flow chart of kinetic model using error minimization approach.
8827
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
technique used in the mathematics to solve a set of unknown variables based on the objective function to be minimized or maximized, where some of the nonlinear functions are present [30]. Product gas compositions from fluidized bed gasification system were considered under the effect of temperature, steam to biomass and adsorbent to biomass ratios. Three experimental points were taken for each process variables. Twelve parameters were estimated through nonlinear regression by considering E and A for all six reactions (Table 1). RSS was considered as an objective function of the optimization process (Equation (8)). Mean error was then evaluated using RSS values (Equation (10)). The present model can be advantageous for extensive range of process variables. Moreover, the model can also be applied for scaling up the biomass gasification processes.
Thermodynamic parameters such as equilibrium constant and Gibbs free energy are calculated for water gas shift reaction in the present study. Several studies [9,11] found that the water gas shift seems to reach the equilibrium, and control the gas phase kinetics in biomass steam gasification in fluidized bed reactor. Equilibrium constants and Gibbs free energy are evaluated based on the concentration of product gas at three different temperature of 600 C, 675 C and 750 C at steam to biomass ratio of 2.0 (wt/wt), adsorbent to biomass ratio of 1.0 (wt/wt) and catalyst to biomass ratio of 0.1 (wt/wt). Equilibrium constant for the water gas reaction (reaction 5) can be calculated as [31]: ½CO2 ½H2 ½CO½H2 O
(11)
The quantity in the bracket shows the concentration (mol/ m3) of product (CO2, H2) and reactant (CO, H2O). Once Ke is determined, change in Gibbs free energy is calculated [32] to check spontaneous and nonspontaneous nature of the reaction at given operating conditions: DG+ ¼ RT ln Ke
Moisture (%) Proximate analysis (wt. % dry basis) Volatile matter Fixed carbon Ash content Ultimate analysis (wt. % dry basis) C H N S O (by difference) Higher heating value (HHV)
9.61 80.92 14.67 4.31 49.74 5.68 1.02 0.27 43.36 18.46 MJ/kg
Table 3 e Physical properties of bed material.
Thermodynamic parameters evaluation
Ke ¼
Table 2 e Proximate and ultimate analysis of palm kernel shell.
(12)
R and T represent universal gas constant (J/mol.K) and temperature (K), respectively.
Experimental Palm kernel shell was utilized as the feedstock. The local palm kernel shell was obtained from My 4-Seasons International Sdn. Bhd, Selangor, Malaysia. The palm kernel shell received had initial diameter in the range of 0.1 mme4.0 mm. Then, the drying process was performed under sunlight for 3e4 h to remove excess moisture content. The dried palm kernel shell was further sieved into required size of 1.0 mme2.0 mm. Subsequently, the samples were stored in air tight bags. The proximate and ultimate analysis of biomass is shown in Table 2. In addition, quicklime (CaO), supplied by Universal Lime Sdn. Bhd, Malaysia was used as a bed material and as an adsorbents to adsorb CO2 in the product gas. The sample was grinded and sieved to particle size of 0.15e0.25 mm. The physical properties of the bed material are given in Table 3.
Particle density (kg/m3) Bulk density (kg/m3) Chemical composition (wt. %) CaO MgO SiO2 Fe2O3 Other metal oxide (MnO, CuO, SrO, ZnO)
3053 1047 93.32 4.24 0.95 0.23 1.0
The catalyst was pure nickel (Ni) powder (99.5% purity) with particle size of 10 mm. Ni catalyst was purchased from Merck kGaA, Germany. For the entire gasification process, Ni catalyst was mixed with the palm kernel shell and was introduced to the gasifier through a biomass feeding system. Fig. 2 shows a schematic diagram of pilot-scale fluidized bed gasification system used for the experimental runs that was operated under atmospheric pressure. The unit mainly consists of fluidized bed reactor with external heaters, biomass feeder and steam generator with superheater, cyclone solid separator, wet scrubber, water separator and gas analyzing system. The diameter and height of the fluidized bed reactor are 0.15 m and 2.5 m, respectively. A perforated type distributor plate was used. The gasifier was operated at superficial velocity of 0.21 m/s (4 times the minimum fluidization velocity). The fluidized bed gasifier was continuously fed with biomass at 1.0e1.8 kg/h from the biomass feeding system. The cooling water jacket was provided to avoid biomass decomposition in the feeding line. Nitrogen (N2) was used to transfer the biomass into the gasifier in order to avoid any back flow. Saturated steam was provided by the steam generator and was further heated from 250 C to 300 C in the superheater prior its injection into the gasifier. Before the start of each experiment, N2 gas was purged into the system to remove any entrapped gases. The product gas from the outlet of the gasifier was passed through the cyclone to separate entrained solid particles. The product gas was then passed through the scrubber prior to a separator unit to remove traces of water in the product gas. Gas sampling point was located at the exit of the water separator unit. At the same point, the volumetric flow rate was measured. All experiments were run for 60 min. The product gases such as CO2, CO and CH4 were analyzed by using a Gas Chromatography (Teledyne 7500,
8828
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 2 e Schematic of pilot scale fluidized bed gasification system.
Teledyne Analytical Instrument) with Infrared (IR) type detector. H2 and N2 were detected by Gas Chromatograph utilizing a Molecular Sieve 5A column (Teledyne 4060, Teledyne Analytical Instrument) with a Thermal Conductive Detector (TCD). The product gas was measured for every 6 min interval. Three process variables such as temperature (T), steam to biomass ratio (S/B) and adsorbent to biomass ratio (A/B) were tested in the present study. Fluidized bed gasification temperature was tested from 600 C to 675 C. Steam to biomass ratio was varied from 1.5 to 2.5 (wt/wt). Adsorbent (bed material) to biomass ratio was varied from 0.5 to 1.5 (wt/wt). For all experiments, catalyst to biomass ratio of 0.1 (wt/wt) was considered.
Results and discussions Kinetic parameters The kinetic parameter evaluation was carried out using product gas composition. The kinetic parameters were evaluated by minimizing the residual error between the experimental data and model prediction. Table 4 summarizes all the kinetic parameters for the main reactions occurring in ICA steam gasification process. Among all the reactions, char steam gasification has high activation energy. This is due to the endothermic nature of the reaction and requires
Table 4 e Evaluated kinetic parameters. No
1 2 3 4 5 6
Reaction
Char gasification Methanation Boudouard Methane steam reforming Water gas shift Carbonation
Frequency factor, A (1/s)
Activation energy, E (kJ/mol)
104 104 103 104
999.95 0.98 16.84 21.57
0.4 101 2.81 102
22.38 17.57
3.32 3.19 1.7 3.19
temperature >800 C for effective conversion rates in fluidized bed reactors [33].
Effect of gasification temperature Fig. 3 shows the effect of temperature at 600 C, 675 C and 750 C on product gas composition. The concentration profiles of H2, CO, CO2 and CH4 are plotted. As observed, the model prediction is in good agreement with the experimental results. The coefficient of determination (R2) is found to be higher than 0.91. Overall, H2 content increases when temperature is elevated from 600 C to 675 C while concentration of CO2 and CO decreases. This is due to active CO2 adsorption which drives the water gas shift reaction towards forward direction and enhances hydrogen production. This is in line with the previous work reported by Marquard et al. [7]. At 750 C, H2 content reduces in the product gas due to the reverse carbonation reaction that promotes high concentration of CO2 in the product gas. This observation is further verified by a high CO content in the product gas at 750 C. These results are in general agreement with the other studies [5]. However, the opposite trends of CO composition observed at high temperature 675e750 C is may be due to the consideration of forward and backward water gas shift reactions in the present model. Meanwhile, CH4 content gradually decreases from 600 to 750 C which can be elucidated by the endothermic nature of steam methane reforming. Table 5 provides the relative mean error calculated for individual product gas. Low mean error infers good fit of the modeling results towards the experimental values.
Effect of steam to biomass ratio Fig. 4 illustrates the effect of S/B ratio on product gas composition for both experiments and modeling results. The modeling results are in good agreement with experimental values at 675 C, A/B ratio of 1.0 and catalyst to biomass ratio of 0.1 at varying S/B ratio from 1.5 to 2.5. The R2 is found to be higher than 0.90. The addition of steam promotes methane reforming, water gas shift and char gasification reactions in
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 3 e Effect of temperature on product gas composition; modeling (
Table 5 e Relative mean error of product gas composition. Gas component
H2 CO2 CO CH4
Mean error relative Temperature
Steam to biomass ratio
Adsorbent to biomass ratio
0.00005 0.09980 0.00023 0.00250
0.00054 0.06010 0.01877 0.02624
0.00017 0.11612 0.00024 0.00208
8829
) and experiment ( ).
the forward direction to produce more H2 in product gas. Likewise, the increase in H2 content can be inferred by decrease in concentration of CO and CH4 and increase in concentration of CO2 content in product gas. This is consistent with the results reported by Mahishi et al. [5] for steam gasification with in-situ CO2 adsorbent in fluidized bed reactor. However, varying S/B ratio from 2.0 to 2.5 is not found to increase concentration of H2 significantly. This is due to excessive amount of steam present in the system. This observation is supported by the results of Han et al. [34] for
Fig. 4 e Effect of steam to biomass ratio on product gas composition; modelling (
) and experiment ( ).
8830
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 5 e Effect of adsorbent to biomass ratio on product gas composition; modeling (
saw dust steam gasification with CO2 adsorbent in fluidized bed reactor. It can be concluded that the S/B ratio of 2.0 is the optimum to produce hydrogen rich gas in ICA steam gasification process. A good agreement can be seen between experimental and model values based on the mean error values (Table 4).
Effect of adsorbent to biomass ratio The A/B ratio is tested for product gas distribution as shown in Fig. 5. The model results fits well to the experimental data conducted at 675 C, S/B ratio of 2.0, and catalyst to biomass ratio of 0.1 by varying A/B ratio from 0.5 to 1.5. The R2 is found be higher than 0.91. Furthermore, the low mean errors between modeling and experimental values are observed for all the product gases as given in Table 4. The product gases i.e. H2, CO, CO2 and CH4 show similar trends to the experimental data. The addition of adsorbent increases the H2 content in the product gas. The excess amount of adsorbent (CaO) captured CO2 in the process which subsequently enhances the activity of water gas shift and promotes all the gasification and reforming reactions towards enhanced H2 generation. Beside, CO2 content slightly increases at A/B ratio of 1.5 which clearly indicates high activity of water gas shift reaction. This can also be explained by the increase of CO content at A/B ratio of 1.5. Han et al. [34] reported similar increase of CO2 with respect to A/B ratio carrying sawdust steam gasification with CO2 adsorption in a fluidized bed reactor.
) and experiment ( ).
equilibrium constant increases with increasing temperature. These results infer that the concentration of product gas i.e. H2 and CO2 is increasing with increasing temperature. This can be justified under the effect of temperature on individual gas composition as shown in Fig. 3. In the present study, H2 composition increases at temperature range of 600 Ce675 C while no CO2 composition is depicted at this temperature. At 750 C, H2 composition decreases but CO2 increases due to reverse carbonation reaction. The positive Gibbs free energy shows that the water gas shift reaction is non-spontaneous at given temperature range (Table 6). To assess the results further, equilibrium constant observed in the present study are plotted along with theoretical equilibrium constant [35] and experimental equilibrium constant observed by Herguido et al. [11] at 600 C, 700 C and 750 C for steam gasification in fluidized bed reactor as shown in Fig. 6. The experimental equilibrium constant from the present study increases with increasing temperature which is in a good agreement with that of Herguido et al. [11]. Similar finding is also observed by other researchers [36]. Comparative study shows that lower experimental equilibrium constant values in the present study may be due to the presence of CO2 adsorbent which captures most of the CO2 in the product gas thus reduces overall equilibrium constant values for water gas shift reaction.
Table 6 e Effect of temperature on equilibrium constant and Gibbs free energy. Temperature ( C)
Thermodynamic parameters
Parameter
Equilibrium constants and Gibbs free energy for water gas shift reaction are listed in Table 6. The result shows that the
Equilibrium constant () Gibbs free energy (KJ/mol)
600
675
750
0.04 23.50
0.14 15.77
0.40 8.81
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
Fig. 6 e Effect of reactor temperature on the equilibrium constant of water gas shift reaction.
Conclusions The kinetic parameters determination for six major reactions involved in ICA steam gasification carried out at a pilot scale fluidized bed gasifier was presented. The kinetic parameters determination was carried out through optimization approach by minimizing error between experimental and modelling values for H2, CO, CO2 and CH4. It was predicted that the H2 content gradually increased by increasing steam to biomass and adsorbent to biomass ratios. However, H2 content increased from 600 C 675 C while later on decreased at 750 C. The concentration decreased due to presence of reverse carbonation reaction in the process. Furthermore, the study of thermodynamic parameters inferred that the water gas shift was a nonspontaneous reaction at a temperature range between 600 C and 750 C. Furthermore, it was observed that low equilibrium constant of 0.014 and 0.14 at 600 C and 675 C, respectively, provided the evidence of strong CO2 adsorption reaction in ICA steam gasification process.
references
[1] Khan Z, Yusup S, Ahmad MM, Rashidi NA. Integrated catalytic adsorption (ICA) steam gasification system for enhanced hydrogen production using palm kernel shell. Int J Hydrogen Energy 2014;39:3286e93. S, Jand N, Foscolo PU. Catalytic gasification of [2] Rapagna biomass to produce hydrogen rich gas. Int J Hydrogen Energy 1998;23:551e7. [3] Hu G, Xu S, Li S, Xiao C, Liu S. Steam gasification of apricot stones with olivine and dolomite as downstream catalysts. Fuel Process Technol 2006;87:375e82. [4] Pfeifer C, Bernhard P, Hofbauer Hermann. In-situ CO2absorption in a dual fluidized bed biomass steam gasifier to produce a hydrogen rich syngas. Int J Chem React Eng 2007;5(A9):1e13. [5] Mahishi MR, Sadrameli MS, Vijayaraghavan S, Goswami DY. A novel approach to enhance the hydrogen yield of biomass gasification using CO2 sorbent. J Eng Gas Turbines Power 2008;130:1e8.
8831
[6] Xu G, Murakami T, Suda T, Kusama S, Fujimori T. Distinctive effects of cao additive on atmospheric gasification of biomass at different temperatures. Ind Eng Chem Res 2005;44:5864e8. € llenstedt T, Sichler P, Specht M, Michel M, [7] Marquard-Mo Berger R, Hein KRG, et al. New approch for biomass gasification to hydrogen 2nd world conference and technology exhibition on biomass for Energy. Rome, Italy: Industry and Climate Protection; 2004. [8] Basu P, Kaushal P. Modeling of pyrolysis and gasification of biomass in fluidized beds: a review. Chem Prod Process Model 2009;4(A21):1e45. [9] Gordillo ED, Belghit A. A two phase model of high temperature steam-only gasification of biomass char in bubbling fluidized bed reactors using nuclear heat. Int J Hydrogen Energy 2011;36:374e81. [10] Florin NH, Harris AT. Hydrogen production from biomass coupled with carbon dioxide capture: the implications of thermodynamic equilibrium. Int J Hydrogen Energy 2007;32:4119e34. [11] Herguido J, Corella J, Gonzalez-Saiz J. Steam gasification of lignocellulosic residues in a fluidized bed at a small pilot scale. Effect of the type of feedstock. Ind Eng Chem Res 1992;31:1274e82. € ll T, Hofbauer H. H2 rich syngas by selective CO2 removal [12] Pro from biomass gasification in a dual fluidized bed system e process modelling approach. Fuel Process Technol 2008;89:1207e17. [13] Inayat A, Ahmad MM, Yusup S, Mutalib MIA. Biomass steam gasification with in-situ CO2 capture for enriched hydrogen gas production: a reaction kinetics modelling approach. Energies 2010;3:1472e84. [14] Corella J, Sanz A. Modeling circulating fluidized bed biomass gasifiers. A pseudo-rigorous model for stationary state. Fuel Process Technol 2005;86:1021e53. [15] Gonzalez-Saiz J. Advances in biomass gasfication in fluidized bed. Spain: University of Saragossa; 1988. [16] de Souza-Santos ML. Comprehensive modelling and simulation of fluidized bed boilers and gasifiers. Fuel 1989;68:1507e21. [17] Liu H, Gibbs BM. Modeling NH3 and HCN emissions from biomass circulating fluidized bed gasifiers. Fuel 2003;82:1591e604. [18] Simell PA, Hirvensalo EK, Smolander VT, Krause AOI. Steam reforming of gasification gas tar over dolomite with Benzene as a model compound. Ind Eng Chem Res 1999;38:1250e7. [19] Sun P, Grace JR, Lim CJ, Anthony EJ. Determination of intrinsic rate constants of the CaO-CO2 reaction. Chem Eng Sci 2008;63:47e56. [20] Yunus MK, Ahmad MM, Inayat A, Yusup S. Simulation of enhanced biomass gasification for hydrogen production using iCON. In: Proceedings of world academy of science, engineering and technology62; 2010. p. 753e60. [21] Sreejith CC, Haridasan N, Muraleedharan C, Arun P. Allothermal airesteam gasification of biomass with CO2 (carbon dioxide) sorption: performance prediction based on a chemical kinetic model. Energy 2014;69:399e408. [22] Sommariva S, Grana R, Maffei T, Pierucci S, Ranzi E. A kinetic approach to the mathematical model of fixed bed gasifiers. Comput Chem Eng 2011;35:928e35. [23] Sreejith CC, Muraleedharan C, Arun P. Airesteam gasification of biomass in fluidized bed with CO2 absorption: a kinetic model for performance prediction. Fuel Process Technol 2015;130:197e207. [24] Inayat A, Ahmad MM, Mutalib MIA, Yusup S. Optimization approach for kinetics parameters determination for oil palm waste steam gasification with in-situ CO2 capture for hydrogen production. Natl Postgrad Conf NPC 2011:1e6.
8832
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 8 8 2 4 e8 8 3 2
[25] Inayat A, Ahmad MM, Mutalib MIA, Yusup S. Process modeling for parametric study on oil palm empty fruit bunch steam gasification for hydrogen production. Fuel Process Technol 2012;93:26e34. [26] Shen L, Xiao J, Niklasson F, Johnsson F. Biomass mixing in a fluidized bed biomass gasifier for hydrogen production. Chem Eng Sci 2007;62:636e43. [27] Kaushal P, Abedi J, Mahinpey N. A comprehensive mathematical model for biomass gasification in a bubbling fluidized bed reactor. Fuel 2010;89:3650e61. [28] Kimura T, Miyazawa T, Nishikawa J, Kado S, Okumura K, Miyao T, et al. Development of Ni catalysts for tar removal by steam gasification of biomass. Appl Catal B 2006;68:160e70. [29] Puig-Arnavat M, Bruno JC, Coronas A. Review and analysis of biomass gasification models. Renew Sustain Energy Rev 2010;14:2841e51. [30] Avriel M. Nonlinear programming analysis and methods. Dover Publishing; 2003.
[31] Mountouris A, Voutsas E, Tassios D. Solid waste plasma gasification: equilibrium model development and exergy analysis. Energy Convers Manage 2006;47:1723e37. [32] Loha C, Chatterjee PK, Chattopadhyay H. Performance of fluidized bed steam gasification of biomass e modeling and experiment. Energy Convers Manage 2011;52:1583e8. [33] Basu P. Combustion and gasification in fluidized beds. NY: Taylor & Francis; 2006. [34] Han L, Wang Q, Yang Y, Yu C, Fang M, Luo Z. Hydrogen production via cao sorption enhanced anaerobic gasification of sawdust in a bubbling fluidized bed. Int J Hydrogen Energy 2011;36:4820e9. [35] Twigg MV. Catalyst handbook. 2nd ed. London: Manson Publisher; 1996. [36] Wei L, Xu S, Zhang L, Liu C, Zhu H, Liu S. Steam gasification of biomass for hydrogen-rich gas in a free-fall reactor. Int J Hydrogen Energy 2007;32:24e31.