A comparative study on gaseous fuel generation capability of biomass materials by thermo-chemical gasification using stoichiometric quasi-steady-state model

The present work investigates the potential of different biomasses for air–steam gasification using thermodynamic equilibrium model. A stoichiometric thermodynamic equilibrium model considering char conversion and tar formation is developed using MATLAB software to compare the gasification performance of different biomasses in terms of product gas yield, heating value and energy efficiency. Regression analysis is performed and correlations for the yield of different syngas constituents are developed in terms of temperature, steam to biomass ratio and equivalence ratio for all the biomasses. For a temperature of 1000 K, steam to biomass ratio of unity and equivalence ratio of 0.25, maximum mole fraction of hydrogen (14.89 %) was obtained from rubber seed shell with a lower heating value and gas yield of 4.71 MJ/Nm3 and 1.18 Nm3/h, respectively.


Introduction
Thermo-chemical gasification is one of the practical methods to extract hydrogen from renewable energy sources such as biomass. Being a clean fuel, use of hydrogen is a potential solution for problems associated with fossil fuel depletion and global warming. Hydrogen yield from biomass gasification depends on many factors such as type of biomass, moisture content, operating conditions, gasifying agents etc. Effect of different gasifying agents on syngas composition was analysed by Gil et al. [1] and found that compared to steam, tar yield and lower heating value (LHV) of syngas are lower when air is used as the gasifying agent. Another important factor that affects the gasification performance is the type of gasifier used. Warnecke [2] made a comparison of fixed and fluidised bed gasifiers and concluded that in spite of high dust content in the product gas and low ash melting point, fluidised bed gasifiers are attractive for their high heat and mass transfer, favourable solid-gas contact, temperature controllability etc.
Biomass gasification can be investigated using mathematical models when large-scale experimental studies are expensive and difficult to conduct. Extensive reviews on biomass gasification models were presented by Baruah and Baruah [3] and Puig-Arnavat et al. [4]. Among the available gasification models, thermodynamic equilibrium model (TEM) serves as an effective preliminary tool to analyse the effect of different feed stocks and operating parameters on biomass gasification. In general, equilibrium modelling can be achieved through two distinct approaches namely, stoichiometric and non-stoichiometric [5]. Even though both the approaches are conceptually similar, they differ in procedure, as the latter one is comparatively complicated in procedure. Thus, stoichiometric thermodynamic equilibrium models were formulated by many researchers to investigate biomass gasification [6][7][8][9][10][11][12][13]. Zainal et al. [14] successfully demonstrated air gasification of biomass using equilibrium constant expressions of water gas shift reaction and methanation reaction in terms of gasification temperature. TEMs can be modified to improve its prediction accuracy by considering char conversion, tar formation and incorporating suitable coefficients to equilibrium constants. Modification of TEM to augment its prediction accuracy was demonstrated by Jarungthammachote and Dutta [15] and Huang and Ramaswamy [16] for air gasification, Loha et al. [17] for steam gasification and Lim and Lee [18] for air-steam gasification. Barman et al. [19] considered tar as a compound containing carbon, hydrogen and oxygen in air gasification. Azzone et al. [20] considered char conversion as a function of equivalence ratio (ER) in air-steam gasification, whereas Lim and Lee [18] expressed it as a function of equivalence ratio and temperature. Abuadula et al. [13] included tar as benzene and unreacted char as 5 % of biomass carbon content in steam gasification model. Ng et al. [9] considered char as solid carbon and tar as a compound containing carbon, hydrogen and oxygen along with correction factors for equilibrium constants as temperature dependent. Nguyen et al. [21] developed a three-stage quasi-equilibrium model for steam gasification of biomass where the deviation from thermodynamic equilibrium was reduced using empirical relations. Application of engineering equation solver (EES) in the modelling of biomass gasification was demonstrated by Puig-Arnavat et al. [22]. Deviation of this model from pure equilibrium is minimised by considering char and tar, pyrolysis, heat loss in pyrolysis, particles leaving the gasifier and setting the amount of CH 4 produced. Present work deals with the stoichiometric modelling of air-steam gasification considering tar and char and its application to compare the effect of different feed stocks in biomass gasification. The present work deals with the comparison of gasification process with different feed stocks in terms of syngas composition, gas yield and first law efficiency.

Characterisation of biomass sample
Knowledge of chemical and physical properties of biomass, which serves as an essential pre-requisite of any biomass gasification study, can be obtained through characterisation tests. Characterisation tests were conducted on locally available biomasses in Calicut, Kerala, India such as rice husk, coconut shell, saw dust, coir pith and rubber seed shell [23]. The proximate and ultimate analyses results of the biomasses are given in Tables 1 and 2, respectively.

Model development
The objective of present work is to develop a thermodynamic equilibrium model to analyse the influence of different locally available feed stocks on gasification process. A stoichiometric thermodynamic equilibrium model for air-steam biomass gasification, incorporating char and tar, is developed based on the following assumptions, • Gasifier is considered as a steady-state system with uniform temperature and pressure throughout. • All the gases behave ideally.
• Gases except H 2 , CO, CO 2 , CH 4 and N 2 are considered dilute. • N 2 is considered as inert in the entire process.
• Biomass is considered to be made up of Carbon, Hydrogen, Oxygen and Nitrogen. • Steam is supplied at superheated condition of 1 bar and 300°C. • Tar and char are assumed as benzene and graphitic carbon, respectively.
By considering chemical formula of feedstock as C a-H b O c N d , global gasification reaction can be written as: where, a, b, c and d are the number of atoms of carbon, hydrogen, oxygen, and nitrogen per mole of dry and ash free biomass. m is the number of moles of moisture per mole of  biomass. n O 2 , n H 2 , n CO , n CO 2 , n CH 4 , n N 2 , n H 2 O and n tar are the stoichiometric coefficients of oxygen, hydrogen, carbon monoxide, carbon dioxide, methane, nitrogen, steam and tar, respectively. s is the number of moles of steam supplied and a is the carbon conversion factor, expressed as a function of temperature and equivalence ratio [18].
Tar yield as a weight percentage of total gasification products is given by [13], Total weight of the gasification product is obtained by applying mass balance to the global reaction between the reactants and the products. So mass of tar yield (m tar ) is given by, where SBR is steam to biomass ratio, defined as the ratio of mass flow rate of steam to mass flow rate of biomass. Five separate equations are required to determine the five unknown constituents of product gas. These equations can be developed from the elemental balance of C, H, O and N and from the equilibrium constant expressions [Eqs. (10) and (12)] of water gas shift reaction [Eq. (9)] and hydrogasification reaction [Eq. (11)], respectively.
Elemental balance Carbon balance, Hydrogen balance, Oxygen balance, Nitrogen balance, Water gas shift reaction Considering equilibrium constant K 1 for water gas shift reaction, Hydrogasification reaction, Considering equilibrium constant K 2 for hydrogasification reaction, For ideal gas, K 1 and K 2 can be expressed as a function of temperature [14], given by Eqs. (13) and (14).
Lower heating value of the dry product gas is estimated from the gas composition and is expressed in volume basis as [24], Gasification efficiency of the process is given by,

Model validation
Prediction capability of the developed model is checked by comparing the predicted gas composition with experimental results [25]. The deviation of model result from experimental values is quantified using the statistical parameter of root mean square (RMS) error, where X e , X p and N are experimental data, predicted value and number of observations, respectively. An average RMS = 4.7205 is obtained when eight sets of experimental results are compared with their corresponding theoretical predictions, for a gasifier pressure of 1 bar, as given in Fig. 1.

Model modification and non-equilibrium factors
It is observed that CO 2 and CH 4 concentrations were underpredicted and H 2 and CO concentrations were over-predicted by the present model from the experimental values. Melgar et al. [26] made similar observation when the syngas composition predicted by their model is compared with the experimental results of Jayah et al. [27]. Same results were obtained when the model predicted gas composition [28][29][30][31] is compared with the experimental results of Rapagna et al. [32] and Hofbauer et al. [33]. The prediction accuracy of the model is improved by multiplying K 1 and K 2 with suitable coefficients C 1 and C 2 , expressed as functions of ER [18].
Average RMS error is checked by varying the values of p 1 , p 2 , p 3 , p 4 , C 1 and C 2 , and the values corresponding to minimum average RMS error are selected for the model. Average RMS error is reduced from 4.7205 to a minimum of 1.3713 by introducing suitable values of C 1 and C 2 in the model (C 1 ¼ 0:12e 3:2 ER and C 2 ¼ 41 À 50ER). Figure 2 shows a fair agreement between experimental results and that obtained from modified model.

Model application
The stoichiometric quasi-equilibrium model developed is used to predict the influence of key process parameters such as gasification temperature, steam to biomass ratio and equivalence ratio (ER) on syngas composition, heating value and energy efficiency. Gasification performance of different biomasses, at a gasifier pressure of 1 bar, is compared by keeping mass flow rate as 1.0 kg/h and varying temperature, SBR and ER in the ranges of 900-1500 K, 0-3.5, 0.15-0.45, respectively.

Results and discussion
Effect of gasification temperature on product gas composition Effect of gasification temperature on product gas composition is depicted through Fig. 3. Among all the feed stocks, rubber seed has the maximum potential for hydrogen generation under same operating conditions. For rubber seed shell, coconut shell and saw dust, H 2 mole fraction is found to be increasing with temperature up to 1300 K, and remains unchanged thereafter. However, for rice husk and coir pith H 2 concentration decreases with increase in temperature beyond 1300 K. For all feed stocks, the increase in H 2 concentration is more pronounced at lower temperature ranges (900-1300 K). Similar variation on H 2 concentration with temperature was observed by Lv et al. [34]. This trend of H 2 is mainly due to the reversal of exothermic water gas shift reaction at higher temperature ranges. For unit SBR and ER = 0.25, maximum hydrogen concentration values for rice husk and coir pith are found to be 17.67 and 16.92 %, respectively, at 1300 K. Figure 3b shows that coir pith, rice husk and saw dust have almost   Fig. 3b, c, it is clear that CO and CO 2 concentrations, respectively, increase and decrease with temperature.
This variation indicates the reversal of water gas shift reaction in gasification process. Decrease in CH 4 concentration with temperature, as shown in Fig. 3d, is due to the effect of shifting of exothermic hydrogasification reaction towards the reactant side.
Effect of ER on product gas composition Figure 4 shows the effect of ER on different gas constituents in the product. All the gas constituents except CO 2 decrease with ER. This is due to shifting of the process more towards combustion at higher ER values.
Even though the number of moles of CO 2 increases, there is a decrease in CO 2 mole fraction with ER. This is attributed to the higher rate of increase of total number of moles of gas constituents, due to the addition of N 2 , compared to the increase in number of moles of CO 2 . The increase in number of moles of CO 2 with ER is shown in Fig. 5.

Effect of SBR on product gas composition
The influence of steam addition on product gas composition is depicted through Fig. 6. Increase in H 2 concentration with SBR is due to the effect of water gas shift reaction. The increase in H 2 mole fraction with SBR is found to be insignificant beyond a SBR value of 2.5 for all the feed stocks except coir pith, for which it decreases slightly beyond SBR = 2.5. The decrease and increase in CO and CO 2 concentrations with SBR is depicted through Fig. 6b, c, respectively. This is due to the effect of water gas shift reaction. Influence of methanation reaction on gasification process is reflected in the decrease of CH 4 mole fraction with SBR as shown in Fig. 6d. Similar effects of decrease in CO and increase in H 2 and CO 2 concentrations with SBR are observed in the literature [35] at a temperature of 988 K and ER of 0.12.  Figure 7 shows the decrease of LHV of syngas produced from all the feed stocks with ER. This is due to the shifting of partial oxidation to combustion at higher ER values. Lower heating value of syngas decreases as the feed stock is changed in the order saw dust, coconut shell, rubber seed shell, coir pith and rice husk.

Effect of temperature, SBR and ER on efficiency
Influence of temperature, SBR and ER on gasification efficiency is illustrated through Fig. 8. The increase in gasification efficiency with temperature is due to the hike in values of lower heating values of syngas at higher temperatures.
Efficiency is found to decrease with both SBR and ER. This decrease with ER is due to the reduced lower heating value of the product gas, whereas the reason for efficiency degradation with SBR is due to the increased energy input in the form of steam. Comparison of gas composition, lower heating value and gas yield for different biomasses is shown in Table 3.
For T = 1000 K, SBR = 1 and ER = 0.25, rubber seed shell and coir pith have the maximum and minimum potential for hydrogen generation, respectively. It is also observed that rice husk, biomass with maximum oxygen content, yields lowest calorific value syngas and biomass with maximum carbon content, saw dust, contributes to highest gas yield under the same operating conditions. The biomass with largest hydrogen content is found to produce syngas with higher hydrogen concentration.

Regression analysis
A regression analysis is performed by considering 392 values for each gas constituents, using statistical software Minitab 16, and correlations for the yield of different gas species in terms of gasification temperature, SBR and ER are formulated. The regression equations for the product yield of different biomasses are given in Table 4.

Conclusions
A stoichiometric thermodynamic equilibrium model for air-steam gasification of biomass considering char conversion and tar formation was developed to compare the gasification performance of different locally available feed stocks. After comparing the model results with experimental results for product gas composition, its prediction accuracy was improved by incorporating suitable coefficients with equilibrium constants. The resulting quasiequilibrium model was used to compare the gasification performance of different feed stocks such as rice husk, coconut shell, saw dust, coir pith and rubber seed shell in terms of product gas composition, lower heating value, gas yield and gasification efficiency.  Operating conditions: T = 1000 K, SBR = 1 and ER = 0.25