Biomass and Bioenergy 143 (2020) 105847

Contents lists available at ScienceDirect

Biomass and Bioenergy

journal homepage: http://www.elsevier.com/locate/biombioe

Process modeling and optimization for biomass steam-gasification

employing response surface methodology
Sk Arafat Zaman , Dibyendu Roy , Sudip Ghosh *
Department of Mechanical Engineering, Indian Institute of Engineering Science and Technology, Shibpur, Howrah, 711103, West Bengal, India

A R T I C L E I N F O A B S T R A C T

Keywords: This paper presents a process modeling and optimization study about steam-gasification of biomass. The gasi­
Biomass fication model is developed using Aspen Plus process simulation tool, and rice husk is considered as biomass fuel.
Gasification Simulation results were validated with reported experimental results. The effects of the critical parameters,
Exergy
namely, steam-to-biomass ratio (S/B) and gasification temperature on the quality of the product gas as well as
Cold gas efficiency
Response surface methodology
the gasifier cold-gas efficiency were analyzed. Response surface methodology (RSM) is employed to understand
the synchronized effects of the critical decision parameters and thus to determine the optimized zone of oper­
ating condition. The study reveals that steam gasification can yield relatively clean, H2-rich (up to 58%, dry
basis) product gas and the RSM analysis suggests that optimum performance is obtained for gasification tem­
perature in the range of 750–900 ◦ C and S/B in the range of 0.70–0.81, when the cold gas efficiency (CGE)
approaches 90% and yields dry gas LHV of 12 MJ/kg and more.

1. Introduction suitable for the gasification of agricultural residues such as rice husk,
straw, stalk etc. utilizing steam as gasification agent.
Keeping scarcity and environmental hazards of fossil fuels in mind, The parameters that affect the performance of gasification are the
researchers are increasingly shifting their attention to renewable energy gasifying agent (air or steam), gasification temperature and the size of
sources. Biomass is considered as one of the most favored forms of the biomass particle [8,9]. Higher gasification temperature and small
renewable sources [1]. Apparent carbon neutrality and worldwide particle size are favourable for the gasification process, as they help to
availability are the most notable characteristic of biomass [2,3]. Major increase in conversion of biomass while reducing the concentration of
types of biomass are primary biomass, collected directly from plantation char.
areas and waste biomass like municipal solid wastes. Energy crops are Advanced simulation and process engineering (ASPEN) can be used
mainly grown for use in energy conversion systems. In both large and to model gasification processes and to estimate the composition of
small scale power generation, biomass can be utilized as a replacement syngas obtained after gasification. Li et al. [10] developed a novel
of fossil fuels [4–6]. tri-generation system taking biomass and solar energy as co-feeds, and
Biomass gasification is a process where biomass undergoes thermal they performed exergy analysis on the system. The effects of
conversion to produce a combustible gas mixture, which contains steam-to-biomass ratio (S/B) and equivalence ratio were investigated,
hydrogen, methane, carbon monoxide, carbon dioxide, and water and the highest destruction of exergy was found to be in the gasifier.
vapour. In air gasification, solid biomass is combusted partially in the Zhang et al. [11] modeled a biomass partial gasification process. The
presence of air at sub-stoichiometric ratio and the product gas contains authors carried out the exergy and energy analyses of the gasification
substantial amount of N2. In steam gasification, steam is the primary model. The performance of gasification model was investigated
gasifying agent and the product gas is rich in H2 while N2 content is considering different parameters of the system and exponential increase
minimal. Further, steam gasification produces minimal amounts of ox­ in the product of exergy destruction and time was observed after carbon
ides of sulphur and nitrogen because of the oxygen-deficient condition conversion ratio value of 0.7. Chen et al. [12] modeled a supercritical
and lower gasification temperature [7]. There are different gasifiers water coal gasification system and O2–H2O coal gasification system, and
available in the market. Of them, fluidized bed gasifiers (CFBs) are reported the comparative performances. It was observed that the coal

* Corresponding author.
E-mail addresses: skarz1995@gmail.com (S.A. Zaman), dibyenduroy8@gmail.com (D. Roy), sudipghosh.becollege@gmail.com (S. Ghosh).

https://doi.org/10.1016/j.biombioe.2020.105847
Received 28 February 2020; Received in revised form 7 September 2020; Accepted 24 October 2020
Available online 17 November 2020
0961-9534/© 2020 Elsevier Ltd. All rights reserved.
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Fig. 1. Biomass steam-gasification model.

Hasanzadeh et al. [20] investigated heat transfer mechanisms of

Table 1 microcellular polymeric foams through RSM technique. Mishra et al.
Proximate and ultimate analyses of rice husk [24] and almond shells [37].
[21] performed an optimization procedure employing RSM for air-film
Parameters Rice husk Almond shells Unit blade cooled gas turbine cycle. Sarafraz et al. [22] performed the opti­
Proximate analysis mization of a solar collector employing RSM technique. Christwardana
FC 14.99 20.08 % et al. [23] conducted experimentation and optimization of a microbial
VM 55.54 78.66 % fuel cell.
Moisture 9.95 – %
Ash 19.67 1.26 %
Although the concept of response surface methodology (RSM) has
HHV 12.7 19.52 MJ/kg been applied in other engineering areas, it has apparently not been
Ultimate analysis explored in the study of gasification technology. In this study, therefore,
C 38.43 50.65 % response surface methodology (RSM) is employed to understand the
O 36.36 42.06 %
synchronized effects of the critical decision parameters and thus to
H 2.97 6.03 %
S 0.07 – % determine the optimized zone of operating condition. The effects of the
N 0.49 – % critical parameters, namely, S/B and gasification temperature on the
Ash 21.68 1.26 % quality of the product gas as well as the gasifier cold-gas efficiency were
analyzed. Energetic performance study has been complemented with
exergetic analysis as well. The simulation model for steam gasification
gasification in supercritical water is thermodynamically advantageous
was developed in Aspen Plus process simulation software and was
over O2–H2O coal gasification system. Galletti al [13]. developed a
validated with reported experimental results.
biomass furnace for an externally fired gas turbine. The model was
validated through in-flame estimation of the temperature and chemical
2. System modeling and assumptions
species. Adnan et al. [14] thermodynamically modeled a biomass
gasifier and it was found that the controlled utilization of oxygen im­
The Aspen Plus process flow diagram for biomass gasification is
proves system efficiency. Pala et al. [15] modeled a steam gasification
shown in Fig. 1.
system employing Aspen Plus software, considering different biomass
Development of the model comprises of “stream class specification”
feeds, and producer gas compositions were estimated for respective
for conventional and non-conventional components and “property
biomass feed. Doherty et al. [16] modeled a CFB biomass gasifier. Pre­
method selection”. The process flow has been simulated by integrating
heating effect of the air (gasifying agent) was studied in their work, and
relevant blocks or components specified in Aspen Plus inbuilt database.
preheating of air resulted in the better composition of syngas and per­
The input parameters to different streams (mass flow rate, thermody­
formance of the system.
namic condition and chemical composition) and blocks (thermodynamic
Fluidized bed gasification technology has been employed widely for
condition and chemical reactions) of the model have been suitably
steam gasification process as it possesses advantages like positive solid-
specified.
gas contact and better heat transfer. Arteaga et al. [17] modeled biomass
The assumptions for the modeling of steam gasification are as
gasification for the production of methanol and SNG, reporting an en­
follows:
ergy efficiency of 74%. Shayan et al. [18] developed a model for biomass
gasification of “paper” and “wood” using various gasifying agents to
• The simulation process is run considering the steady-state condition
produce hydrogen and for steam gasification calorific value of syngas
and at atmospheric pressure.
exceeding 11 MJ/kg was observed. Abuadala et al. [19] carried out
• Isothermal condition is considered in the gasifier.
exergy analysis of a biomass gasifier, and studied the effects of gasifi­
• Uniform pressure and temperature are considered inside the gasifier.
cation temperature, biomass input rate and steam feed rate on the per­
• Rate of biomass feed is 1 kg/s and steam supplied at 150 ◦ C.
formance of the system, reporting improved H2 production at the
• Gases are considered to be ideal.
optimized operating condition.
• Char comprises ash and carbon.
Effective design of experiment (DOE) minimizes time, the number of
• The formation of tar is not considered.
trials, and the associated cost. Response surface methodology (RSM) is
one of the popular DOE tools which is generally employed to optimize
An equilibrium approach has been considered for the reactions
the objective functions. RSM has been employed in various engineering
concerned, which renders the model somewhat independent of the
areas by numerous researchers for optimization purpose. For example,
gasifier type. However, the assumptions made for the model

2
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Table 2 free energy is extensively utilized to execute performance analysis in the

Input parameters for the gasifier model. gasification process [27]. The temperature of gasification is varied be­
Stream/Block Parameter Value Unit tween 650 ◦ C and 900 ◦ C. A CYCLONE separator (SOLID SEP) is used to
o separate the ash, and H2S is separated from the produced gas using a
Biomass Temperature 25 C
Pressure 1 bar separator (H2S SEP). As H2S is harmful to any downstream applications
Mass flow rate 1 kg/s like in fuel cells and gas turbine, it needs to be separated before using the
Steam Temperature 150 o
C syngas in any such system.
Pressure 1 bar Of all the material streams, “BIOMASS” and “STEAM” are user-
o
Gasifier (reactor) Temperature 650–900 C
Pressure 1 bar
defined streams. Non-consideration of tar, which is a mix of a large
number of hydrocarbon species, simplified the model (char was however
considered and was assumed to account for all unconverted carbon). As
far as the optimization study is concerned, non-consideration of tar is
Table 3
not expected to affect the trends and solution sets considerably. The
Reactions involved in the gasification model [15].
description of different blocks used in the Aspen flowsheet is shown in
Reaction Name Reaction Reaction Heat of reaction Table 2.
No. (kJ/mol)

Partial oxidation of C + 0.5O2 →CO R1 − 112 3. Thermal modeling

carbon
Combustion of carbon C + O2 →CO2 R2 − 393
The reactions involved in the biomass gasification process using
Boudouard C + CO2 →2CO R3 +172
steam as gasifying agent are listed in Table 3. Reactions R1, R2 and R6 are
Water gas shift CO + H2 O→CO2 + R4 − 41
H2 combustion reactions, and the rest of the reactions are reduction
Water gas C + H2 O→CO + H2 R5 +131 reactions.
Partial combustion of H2 + 0.5O2 →H2 O R6 − 242 Reactions like Boudouard, water gas and methane steam reforming
H2 are endothermic. Whereas, other reactions like combustion reactions,
Carbon methanation C + 2H2 →CH4 R7 − 74 water gas shift, carbon methanation and formation of H2S are
Methane steam CH4 + H2 O→CO + R8 +206 exothermic reactions.
reforming 3H2 The performance of the steam gasification is judged based upon the
Formation of H2S H2 + S→H2 S R9 − 20.2 cold gas efficiency of the system (CGE) that takes into account the
composition of the product gas (syngas) and the corresponding heat
value. The heating values (HHV and LHV) of syngas can be obtained
development suit a circulating fluidized bed (CFB) gasifier the best.
using a property set in the Aspen Plus. The heating values of syngas
Moreover, steam gasification is commonly carried out in a fluidized bed
depend on various factors like biomass composition used, S/B and the
gasifier. In validating the model, therefore, experimental data from CFB
gasification temperature, as shown in the sensitivity analysis.
systems were used.
The cold gas efficiency (CGE) is the ratio of energy obtained from
Biomass was defined as a non-conventional fuel stream and its
syngas to the energy content available in the biomass, steam and heat
properties were assigned accordingly. Ultimate and proximate analyses
supplied, as shown in equation 1, [24]. The amount of heat supplied can
of selected biomass were entered under component attributes of
be estimated from the difference of enthalpy of product gas and enthalpy
“BIOMASS” stream. The compositions of rice husk and almond shells
input in the form of biomass and steam. It is evident from the expression
were considered as biomass fuel in the proposed model are given in
of cold gas efficiency that for a biomass feed, the cold gas efficiency is
Table 1.
dependent on the HHV of the syngas obtained from the biomass and the
Peng Robinson equation of state (PENG-ROB) is used as a property
amount of steam supplied in that process. So, CGE is also a function of
method for modeling the steam gasification system as this property
S/B and gasification temperature.
method provides excellent precision in gasification process simulation
[12,14]. ṁsyngas *HHVsyngas
CGE = ( ) ( ) (1)
The input parameters used in the model are given below in Table 2.
ṁb *HHVb + ṁsteam *Hsteam + Q̇in
Decomposition of the biomass into conventional components and ash
is done using yield reactor, RYIELD (DECOMP) and by specifying yield
distribution according to the ultimate analysis of biomass. Char where Hsteam and ṁsteam are enthalpy and mass flow rates of steam.
(unconvertible carbon) is separated from the decomposed stream based The higher heating value of biomass in MJ/kg is calculated using
on specified flows or split fractions using a separator. The formation of equation 2 [17]
tar (mix of heavy hydrocarbon molecules) is not considered in the
HHVb = (0.3491 × C) + (1.1783 × H) − (0.1034 × O) − (0.0151 × N)
decomposer [25]. Steam gasification in circulating fluidized bed gas­
ifiers produce H2 rich syngas having a very low amount of tar. Other − (0.0211 × A) (2)
experimental study has also reported low tar formation at similar
operating conditions [26]. Ash is also specified as a non-conventional where C, H, O, N and A are mass fraction of carbon, hydrogen, oxygen,
component. The sulphur content in the decomposed product is made nitrogen and ash respectively taken from the proximate analysis.
to react with H2 in a stoichiometric reactor. Full conversion of “S” takes LHV of biomass can be calculated using equation (3) as shown below
place in the stoichiometric reactor block. Steam at temperature 150 ◦ C [28].
and 1 atm pressure is mixed with the gas stream from the decomposer LHVb = HHVb − 21.97[H] (3)
using a mixture and is fed to the RGIBBS reactor, in which gasification
process is carried out. The mass flow rate of steam is varied so that S/B 4. Exergy analysis
varies between 0.6 and 1.5 as shown in the sensitivity analysis. The
stoichiometric reactor and the RGIBBS reactor are together considered Exergy is the utmost extractable work from a certain amount of en­
as “GASIFIER”. RGIBBS model works on chemical equilibrium and ergy. The exergy balance for the proposed model can be expressed from
minimization of Gibbs free energy technique which can be useful in the second law of thermodynamics as follows
multiphase equilibrium designs. The approach of minimization of Gibbs

3
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Fig. 2. Flowchart for model analysis and optimization.

Table 4
Comparison of model results with reported experimental data for rice husk [36].
Parameters H2 CO CO2 CH4 RMSE
0
T( C) S/B Model (%) Exp (%) Model (%) Exp (%) Model (%) Exp (%) Model (%) Exp (%)

690 1.32 57.56 50.5 15.42 14.3 26.56 26.6 0.212 8.6 5.51
730 1.32 57.18 52.2 17.01 16.4 25.49 23.5 0.071 7.9 4.75
750 1 55.3 49.5 22.11 23.7 22.25 21.2 0.089 5.6 4.11
750 0.6 51.54 48.8 31.55 27.5 16.33 19.5 0.3 4.2 3.5
750 1.32 56.96 52.3 17.76 17.75 25 22.25 0.04 7.4 4.56

Table 5
Comparison of model results with reported experimental data for almond shells
[37].
Model Exp RMSE
(%) (%)

H2 58.58 55.5 3.79

CO 27.11 24
CO2 14.12 14.1
CH4 0.19 6.4
LHV (MJ/kg) 11.25 10.94

∑ ∑
Ėin + Ėheat = Ėout + Ėd (4)

∑ ∑
where Ėin , Ėout ,Ėheat and Ėd are exergy rates of the total input, total
output, in the form of rate of heat flow to the gasifier (Q̇in ) and
destruction in the rate of exergy. Ėheat can be calculated from the rate of
heat flow to the gasifier as follows
[ ]
Tatm
Ėheat = 1 − *Q̇in (5) Fig. 3. Effect of gasification temperature on syngas composition(dry basis).
T
Total exergy rate input to the model is in the form of biomass and
steam as shown below

4
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

∑
Ėin = Ėb + Ėsteam (6)

The exergy rate of biomass can be calculated as a function of lower

heating value as shown below [29].

Ėb = ṁb ∗ζ∗LHVb (7)

where ζ is a coefficient and can be expressed as follows [29,30].

1.0414 + 0.0177(H/C) − 0.3328(O/C)(1 + 0.0537(H/C))
ζ= (8)
1 − 0.4021(O/C)
Total exergy rate output from the model in the form of dry syngas,
char and separated steam from the gas product is given below
∑
Ėout = Ėsyngas + Ėchar (9)

Exergy efficiency signifies the efficacy of the model relative to its

performance in reversible conditions. Exergy efficiency, considering
only the product exergy of the syngas is defined as
Fig. 4. Effect of gasification temperature on syngas mass flow rate. Ėsyngas
ηsyngas,ex = (10)
Ėb + Ėsteam + Ėheat
Since char separated from the gasifier is useful, the overall exergy
efficiency can also be defined as the ratio of the sum of exergy rate of the
syngas and char to the total exergy input to the system. The overall
exergy efficiency of the model is defined by

Ėsyngas + Ėchar
ηmodel,ex = (11)
Ėb + Ėsteam + Ėheat

5. Methodology

5.1. Response surface methodology

RSM is widely used to design, develop and optimize new systems or

improve existing ones [31,32]. Minitab is a popular software for car­
rying out statistical analysis employing RSM and the same was used in
the present work. RSM incorporates the regression model, separate
model coefficients and deficiency of fit [33]. The significance of the
model coefficients can be checked by analysis of variance or ANOVA
[34]. F- value is associated with the fraction between the variance across
and the variance within the group. Greater F-values suggests that there
Fig. 5. Effect of steam to biomass ratio on gas composition (dry basis).
is more variability accounted for the equivalent model [35]. If null hy­
pothesis is true, F-value will be near about 1. Whereas p-value is the
likelihood of attainment of closer results to the accurate data. Minor
p-value (less than or equal to 0.05) implies that the corresponding co­
efficient is noteworthy and the null hypothesis is rejected. Greater
F-value and smaller p-value enumerate the impact of variables on
response. Regression coefficient (R2) and the adjusted regression coef­
ficient (R2adj ) are used to evaluate the precision of the model with respect
to the actual data. The range R2 is from 0 to 100%. Nearer value of R2 to
100% implies an excellent forecast of the model of the output response.
Unlike R2 value which rises with the process variables count, R2adj , on the
other hand, offers more suitable results as it decreases when immaterial
variables to the model fit are added.
A 2nd order quadratic equation as shown in equation (12) [38] is
used to establish the relation between response and variables; where “r”
is response; “v” is variable; “n” is the number of variables; β0, βi, βii, βij
are constant term, linear, square and interaction terms respectively.
∑
n ∑
n n− 1 ∑
∑ n
r = β0 + βi vi + βii vi 2 + βij vij (12)
Fig. 6. Effect of steam to biomass ratio on syngas mass flow rate. i=1 i=1 i=1 j=i− 1

The regression coefficient (R2) is calculated using the following

equation

5
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Table 6
ANOVA for cold gas efficiency.
Source DF Seq SS Contribution Adj SS Adj MS F-Value P-Value

Model 5 984.57 98.20% 984.569 196.914 76.20 0.000

Linear 2 979.00 97.64% 979.000 489.500 189.41 0.000
S/B 1 963.62 96.11% 963.620 963.620 372.88 0.000
T 1 15.38 1.53% 15.379 15.379 5.95 0.045
Square 2 2.91 0.29% 2.906 1.453 0.56 0.594
S/B*S/B 1 2.45 0.24% 2.690 2.690 1.04 0.342
T*T 1 0.46 0.05% 0.455 0.455 0.18 0.687
2-Way Interaction 1 2.66 0.27% 2.664 2.664 1.03 0.344
S/B*T 1 2.66 0.27% 2.664 2.664 1.03 0.344
Error 7 18.09 1.80% 18.090 2.584

Total 12 1002.66 100.00%

Fig. 7. Pareto chart for variable influence on CGE.

Table 7
ANOVA for lower heating value.
Source DF Seq SS Contribution Adj SS Adj MS F-Value P-Value

Model 5 4.60748 99.80% 4.60748 0.92150 715.93 0.000

Linear 2 4.37219 94.71% 4.37219 2.18610 1698.43 0.000
S/B 1 3.34795 72.52% 3.34795 3.34795 2601.10 0.000
T 1 1.02424 22.19% 1.02424 1.02424 795.75 0.000
Square 2 0.23279 5.04% 0.23279 0.11640 90.43 0.000
S/B*S/B 1 0.22653 4.91% 0.21304 0.21304 165.52 0.000
T*T 1 0.00626 0.14% 0.00626 0.00626 4.86 0.063
2-Way Interaction 1 0.00250 0.05% 0.00250 0.00250 1.94 0.206
S/B*T 1 0.00250 0.05% 0.00250 0.00250 1.94 0.206
Error 7 0.00901 0.20% 0.00901 0.00129

Total 12 4.61649 100.00%

k (
∑ )2
ri,p − rm k (
∑ )2
ri,p − rm
R 2
= i=1k (13) k− 1
∑ R2adj =1− i=1
* (14)
(ri − rm )2 ∑k
2 k− n− 1
i=1 (ri − rm )
i=1
The equation used to estimate the adjusted regression coefficient
Here, k denotes the number of experiments; ri, ri,p and rm are
(R2adj ) is as follows
experimental, predicted and mean values, respectively.
A flowchart for analysis and optimizing strategy is shown in Fig. 2.

6
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Fig. 8. Pareto chart for variable influence on LHV.

inside which the gasifier was placed. The second comparison has been
Table 8 made with experimental work carried out by Rapagna et al. [37] as
Comparison of the full model and simple model in statistical analysis. shown in Table 5. In their work, a bubbling fluidized bed gasifier was
Simple model Full model used having inner diameter 60 mm and steam was used as fluidizing
R2 (%) R2adj (%) R2(%) R2adj (%)
medium. Almond shells were used as the biomass feed. The comparison
is done for five different sets of gasification temperature and S/B for rice
CGE 97.64 97.17 98.2 96.91 husk, and one set gasification of temperature and S/B for almond shells,
LHV 94.71 93.65 99.8 99.67
as obtained from the literature referred above.
The root mean square error can be estimated using equation (15), as
5.2. Model validation follows
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
√∑
A validation step has been performed in order to test the simulation √n
√ [Model − Exp]2
√ i
model. Aspen Plus simulation results have been compared with two RMSE = (15)
different sets of published experimental results obtained from two types n
of biomass feeds having different compositions, as shown in Table 1. The
Where, n is the count of data point sets.
first comparison is made with results reported by Loha et al. [36] as
Nitrogen gas percentage in the syngas is negligible and nitrogen-free
shown in Table 4. In their work, a laboratory-scale fluidized bed gasifier
syngas (dry basis) is considered for validation. Little discrepancies as
was used and superheated steam was supplied, which worked both as a
witnessed in the comparison approve, model accuracy. The small devi­
gasifying agent and fluidizing agent. Rice husk was used as biomass and
ation in the results is due to the thermodynamic equilibrium model
heat required for gasification was obtained from an electric furnace,

Fig. 9. Effect of decision variable on CGE.

7
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Fig. 10. Effect of decision variable on LHV.

Fig. 11. Interaction consequence of decision parameters on CGE.

which does not take into account time, specific material and equipment gas flow rate have been investigated. Subsequently, RSM is employed to
data. The estimated error is presented in the form of RMS (root – mean – understand the synchronized effects of the critical decision parameters
square) for each set of data as shown in Tables 4 and 5. in order to determine the optimized zone of operating conditions.
Further, an equilibrium model results in almost 100% conversion of Further, the results of the present study are compared with previously
CH4 but it is not feasible for actual gasifiers to reach thermodynamic reported studies.
equilibrium because of the short residence period. Therefore, under-
forecasting of CH4 is quite common in case of equilibrium modeling of
6.1. Sensitivity analysis for the decision variables
fluidized bed gasifiers [36].
Here, the average root mean square errors are found to be 4.486 and
Fig. 3 shows the effect of gasification temperature on the product gas
3.79 respectively.
composition. It is observed that the concentration of CO ascends as the
temperature rises from 650 ◦ C to 900 ◦ C, whereas the percentage of CO2
6. Results and discussion
and CH4 decline with the rise in temperature. The percentage of H2
almost remains unaltered with the variation of temperature. Here the
Two most critical parameters that influence biomass steam gasifi­
gasification temperature is varied from 650 ◦ C to 900 ◦ C for S/B = 1 as
cation process and the product gas quality derived from the same are
shown in Fig. 3.
gasification temperature and S/B. Therefore, the effects of these two
Endothermic reactions like Boudouard reaction, water gas reaction
decision parameters on the producer gas composition and the product
and steam methane reaction favour forward reaction with the rise in

8
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Fig. 12. Interaction consequence of decision parameters on LHV.

Table 9
Design matrix with Aspen Plus results and RSM predicted values.
Std Order Run Order Pt Type Blocks S/B T (0C) LHV (MJ/kg) CGE (%)

Aspen Plus RSM Prediction Aspen Plus RSM Prediction

3 1 1 1 0.89038 863.388 12.14 12.18 88.36 87.16

11 2 0 1 1.35 775 11.01 11.01 74.74 74.74
1 3 1 1 0.89038 686.612 11.38 11.42 88.57 86.02
10 4 0 1 1.35 775 11.01 11.01 74.74 74.74
5 5 − 1 1 0.7 775 12.33 12.28 89.17 91.50
2 6 1 1 1.80962 686.612 10.2 10.18 62.82 62.44
7 7 − 1 1 1.35 650 10.45 10.44 71.54 73.28
12 8 0 1 1.35 775 11.01 11.01 74.74 74.74
13 9 0 1 1.35 775 11.01 11.01 74.74 74.74
9 10 0 1 1.35 775 11.01 11.01 74.74 74.74
8 11 − 1 1 1.35 900 11.47 10.46 77.37 77.21
6 12 − 1 1 2 775 10.41 10.44 61.19 60.46
4 13 1 1 1.80962 863.388 10.86 10.84 65.88 66.84

gasification temperature, which results in increase in H2 and CO con­ However, the concentration of CO falls with the rise in S/B as it favours
centrations and decrease in CO2 concentration. Exothermic reactions the exothermic water gas shift reaction in the forward direction.
like methanation favour backward reaction with the rise in gasification Fig. 6 shows the variation of syngas mass flow rate with S/B. As the
temperature, which results in the decrease of CH4 formation. The other biomass mass flow rate is fixed, an increase in S/B results in a rise in the
probable reason for falling CO2 concentration with the rise in gasifica­ mass flow rate of syngas. From Fig. 5 it is evident that with the increase
tion temperature is the combined impact of Boudouard reaction and in S/B, concentrations of CO2 and H2 increases whereas the concentra­
reversible water gas shift reaction. The formation of H2 is primarily tion of CO falls and the concentration of CH4 remains almost same,
determined by water gas and water gas shift reactions. Low concentra­ resulting in an increase in molar mass of syngas produced and hence
tion of CH4 in the gas mixture results in almost unaltered concentration mass flow rate.
of H2 with the rise in gasification temperature.
Fig. 4 shows the variation of syngas mass flow rate with gasification 6.2. Analysis of variance
temperature. As gasification temperature is increased, the mass flow rate
of syngas obtained reduces as Fig. 3 shows that with the increase in Table 6 depicts the results from ANOVA for cold gas efficiency as the
gasification temperature the concentration of CO ascends, whereas the response variable.
percentage of CO2 and CH4 reduces, H2 remaining the same, results in a The p-value for the entire model is noticed to be 0 with the corre­
decrease in molar mass of the syngas and hence mass flow rate. sponding F-value of 76.20 which demonstrates the significance of the
The effect of S/B from 0.6 to 1.5 on the product gas compositions at a model. Investigation of the importance of individual input process var­
gasification temperature of 800 ◦ C is shown in Fig. 5. iables and their exchanges are also done. The p-value of S/B is found to
It is seen that the concentrations of CO2 and H2 increases as S/B be 0 with associated F-value of 372.88 is the most influential variable
enhances whereas concentration CO falls, the concentration of CH4 that influences CGE.
remaining almost the same with the rise in S/B. Increasing S/B aids Fig. 7 represents the Pareto chart for variable influence on CGE. It is
forward reactions such as endothermic water gas and methane steam observed that all the bars do not cross the reference line 2.36, which
reforming reactions. Thus concentrations of H2 and CO increase. implies that all the factors are not statistically significant. The effect of

9
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Fig. 13. The synchronized effect of gasification temperature and S/B.

LHV in MJ/kg with significant variables formulates the second ordered

Table 10 polynomial regression model, as shown below.
Optimization constraints and objectives.
LHV(MJ / kg) = 8.33 − 3.167 ∗ S / B + 0.01083 ∗ T + 0.8284 * S / B ∗ S / B
Parameters Range/objective
− 0.000004 ∗ T ∗ T − 0.000615 ∗ S / B ∗ T (17)
Gasification Temperature (oC) 750–900
S/B 0.7–0.81
LHV (MJ/kg) maximize
CGE (%) maximize 6.3. Comparison of the full model and simple model in statistical analysis

Both full model (having factors A, B, AA, AB and BB) and simple
two-way interaction coded AB, and square interaction AA and BB are not
model (containing only factors A and B) were analyzed and it was
significant on the response variable (CGE).
observed that the R2 and R2adj values of the full model are better than that
For gasification temperature, the p-value is 0.04 with corresponding
F-value of 5.95, which is significant as well. The high value of regression of the simple model as shown in Table 8, hence full model has been
coefficient 98.2% is observed which shows the excellent fitting of the presented in this paper.
model with the experimental results. The Radj 2 value of 96.91%, is in
great agreement with R2. 6.4. Effects of decision variables on CGE and LHV
The equation of the final model as a function of coded factors for the
CGE with significant variables formulates the second ordered poly­ Fig. 9 and Fig. 10 depict the effects of decision variables (S/B and T)
nomial regression model as shown below. on CGE and LHV respectively. As S/B increases, both CGE and LHV
decreases and with the rise in gasification temperature, both CGE and
CGE(%) = 140.9 − 47.4 ∗ S / B − 0.062 ∗ T + 2.94 ∗ S / B * S / B LHV increase.
+ 0.000033 ∗ T ∗ T +0.0201 * S / B * T (16) With increasing S/B, the amount of H2 and CO2 in the syngas in­
creases and CO decreases as shown in Fig. 5 resulting decrease in LHV.
Table 7 depicts the results from ANOVA for lower heating value as And as gasification temperature increases, the concentration of CO in­
the response variable. creases and CO2 concentration decreases, H2 remaining the same as
The p-value for the entire model is found to be zero having a F-value shown in Fig. 3, results in an increase in LHV. CGE decreases as S/B rises
of 715.93 which shows the importance of the model. S/B has a p-value of due to the decrease in the heating value of syngas. As gasification
zero and F-value of 2601.10 is the most dominating variable that in­ temperature rises, the heating value of the syngas increases, resulting
fluences LHV. For gasification temperature, the p-value is zero with increase in CGE.
corresponding F-value of 795.75 is also significant. Fig. 8 represents the
Pareto chart for variable influence on LHV. It is observed that all the bars 6.5. Interaction effect of decision parameters
do not cross the reference line 2.36, which implies that all the factors are
not statistically significant. Only the effect of two-way interaction coded The interaction consequence of decision parameters on CGE is shown
AB is not significant on the response variable (LHV). in Fig. 11. CGE is found to be maximum at low S/B and high gasification
The high value of R2 is observed to be 99.8%, which shows the temperature. In this condition, the CGE exceeds 90%. In the meantime,
excellent precision of fitting of the model with the experimental results. the minimum value of CGE is found at a condition of high S/B and low
The R2adj value of 99.67% is in excellent agreement with R2. gasification temperature, less than 60%.
The equation of the final model as a function of coded factors for the The interaction consequence of decision parameters on LHV is shown
in Fig. 12.

10
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Table 11 found to be 1.41 and 0.34 respectively.

Performance comparison among the present and previously proposed gasifica­ The constraints for the optimization and objectives, as observed in
tion processes. Fig. 13, are presented in Table 10.
Study Performance Results Optimization Ref. From Table 9, it is observed that maximum values of LHV (12.28 MJ/
Approach kg) and CGE (91.50%) are obtained at a S/B value of 0.70 and at a
Performance assessment At an optimum RSM [38] gasification temperature of 775 ◦ C; and the second maximum values of
and optimization of condition (Gasification LHV (12.18 MJ/kg) and CGE (87.16%) are obtained at a S/B value of
sorption boosted temperature = 700 ◦ C 0.81 and a gasification temperature of 863.39 ◦ C. Thus, the LHV range
chemical looping and Steam to carbon
from 12 MJ/kg to 12.35 MJ/kg and CGE range from 88% to 90% are
biomass gasification ratio = 4.5), 70% CGE,
process. 56% Exergy efficiency selected to predict the optimum zone.
was obtained. The optimum zone is represented by the white region as shown in
Parametric optimization At optimum condition RSM [39] Fig. 13. Any point in the optimum zone can be chosen as the optimum
of catalytic co- (Gasification working point. For example, a point chosen arbitrarily in the region and
gasification of coconut temperature = 900 ◦ C),
shells and oil palm the maximum value of
representing the operating condition of 800.3 ◦ C and 0.8 S/B, gives a
fronds blends. H2 (around 21%) and predicted LHV value of 12.15 MJ/kg and cold gas efficiency (CGE) of
the maximum value of 88.92%, as obtained from Minitab. For the same operating condition
CO (around 19%) was (same T and S/B), the Aspen simulation run yields LHV and CGE values
achieved.
of 12.13 MJ/kg and 87.1% respectively, which are in good agreement
Optimization study for Maximum H2 yield was RSM [40]
high H2 yield for can observed at optimum with the Minitab-predicted values, thus validating the model with
lignite and sorghum condition (Gasification adequate accuracy. Corresponding to this optimum operating point,
biomass gasification. temperature around exergy efficiency is calculated as 76.12% for the gasification process.
890 ◦ C, 1.8 × 10− 3 m3/s Considering exergy of the separated char, the overall exergy efficiency
water flow rate and
stands at 77%.
biomass in coal/
biomass mixture around
25.9%) 6.7. Performance comparison with other optimized gasification processes
Optimization of palm Syngas composition RSM [41]
kernel shell using coal (31.38% H2, 26.44%
A performance comparison is necessary to show the worth of the
bottom ash as a CO, 15.67% CH4 and
catalyst. 25.59% CO2) have been present work in the development of optimization of biomass gasification
observed at optimized processes. Table 11 shows the summary of the present gasification
condition (Gasification process with previously proposed ones having different response vari­
temperature of 850 ◦ C, ables employing RSM. Table 11 shows encouraging results. Utilizing
2.5 L/min air flow rate
and 14.5 wt% catalytic
proposed gasifier, H2 rich syngas having high LHV with high CGE can be
loading). achieved at the optimized condition.
Optimization of 36.72% bio-oil yield RSM [42]
pyrolysis of rice husk, was observed at a 7. Conclusion
an initial assessment temperature 588 ◦ C and
for gasification. at a residence time of
0.25 s while 73.25% gas In this article, a biomass gasification model has been developed for
yield was achieved at a steam gasification. Gas compositions obtained from the model have
temperature of 798.8 ◦ C been validated with reported experimental results. Effects of important
and at a residence time gasification parameters like gasification temperature and S/B on the
of 15.47 s.
Optimization of the H2 rich syngas (upto RSM Present
performance of the gasifier were analyzed. Response surface method­
proposed steam- 58%) can be achieved study ology is employed to perform the optimization of the developed model
biomass gasification and optimum in Minitab software considering the synchronized effect of decision
process performance is obtained parameters. Accuracy of the regression model obtained from RSM is
for gasification
above 99%. Both energetic and exergetic performance have been
temperature in the
range of 750–900 ◦ C studied.
and S/B in the range of The study reveals that seam gasification can yield relatively clean,
0.70–0.81, when the H2-rich (up to 58%, dry basis) product gas and the RSM analysis suggests
CGE approaches 90% that optimum performance is obtained for gasification temperature in
and yields dry gas LHV
of 12 MJ/kg and more.
the range of 750–900 ◦ C and S/B in the range of 0.70–0.81, when the
cold gas efficiency (CGE) approaches 90% and yields dry gas LHV of 12
MJ/kg and more. An operating point has been chosen arbitrarily in the
The highest value of LHV is obtained at low S/B and high gasification region, represented by the operating condition of 800.3 ◦ C and 0.8 S/B,
temperature, which is greater than 12 MJ/kg. Meanwhile, the minimum yields LHV and CGE values of 12.13 MJ/kg and 87.1% respectively. The
value of LHV is obtained at high S/B value and lower value of gasifi­ values closely match the values predicted by Minitab, thus validating the
cation temperature, less than 10 MJ/kg. model results. The values of syngas exergy efficiency and model exergy
efficiency at the selected optimum point are found to be 76.12% and
6.6. Optimization 77% respectively.

Central composite design (CCD) is applied to screen design variables CRediT authorship contribution statement
based on two-level full factorial design. To find out the optimal opera­
tional condition, the combined effect of the parameters is explored. Sk Arafat Zaman: Conceptualization, Methodology, Software,
Table 9 shows the values of cold gas efficiency and LHV values at Validation, Formal analysis, Investigation, Data curation, Visualization,
different sets of gasification temperature and S/B. It also compares the Writing - original draft, Writing - review & editing. Dibyendu Roy:
results obtained from response surface methodology and Aspen Plus Conceptualization, Methodology, Software, Validation, Formal analysis,
software. Root mean square error for cold gas efficiency and LHV are Investigation, Visualization, Writing - review & editing. Sudip Ghosh:

11
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

Conceptualization, Supervision, Writing - review & editing. Department, Jadavpur University, Kolkata for access to their computa­
tional facilities and software.
Acknowledgement

The authors acknowledge the support of Mechanical Engineering

Nomenclature

A Ash
Adj MS Adjusted mean squares
Adj SS Adjusted sum of squares
C Carbon
CGE Cold gas efficiency
DECOMP Decomposer
DF Degrees of freedom
Ė Exergy rate, (kW)
H Hydrogen
HHV Higher heating value, (MJ/kg)
Hsteam Enthalpy of steam, (kJ/kg)
H2S SEP H2S separator
LHV Lower heating value, (MJ/kg)
ṁ Mass flow rate, (kg/s)
N Nitrogen
O Oxygen
Q̇ Rate of heat flow, (kW)
S Sulphur
Seq SS Sequential sum of squares
S/B Steam-to-biomass ratio
T Temperature, (◦ C, K)
VM Volatile matter
η Efficiency
ζ Exergy biomass coefficient

References [12] Z. Chen, L. Gao, W. Han, L. Zhang, Energy and exergy analyses of coal gasification
with supercritical water and O2-H2O, Appl. Therm. Eng. 148 (2019) 57–63,
https://doi.org/10.1016/j.applthermaleng.2018.10.050.
[1] J.H. Pauls, N. Mahinpey, E. Mostafavi, Simulation of air-steam gasification of
[13] C. Galletti, V. Giomo, S. Giorgetti, P. Leoni, L. Tognotti, Biomass furnace for
woody biomass in a bubbling fluidized bed using Aspen Plus: a comprehensive
externally fired gas turbine: development and validation of the numerical model,
model including pyrolysis, hydrodynamics and tar production, Biomass Bioenergy
Appl. Therm. Eng. 96 (2016) 372–384, https://doi.org/10.1016/j.
95 (2016) 157–166, https://doi.org/10.1016/j.biombioe.2016.10.002.
applthermaleng.2015.11.085.
[2] D. Roy, S. Samanta, S. Ghosh, Techno-economic and environmental analyses of a
[14] A. Adnan, M.M. Hossain, Gasi Fi Cation Performance of Various Microalgae
biomass based system employing solid oxide fuel cell, externally fired gas turbine
Biomass – A Thermodynamic Study by Considering Tar Formation Using Aspen
and organic Rankine cycle, J. Clean. Prod. 225 (2019) 36–57, https://doi.org/
Plus, vol. 165, 2018, pp. 783–793, https://doi.org/10.1016/j.
10.1016/j.jclepro.2019.03.261.
enconman.2018.03.078.
[3] S Ghosh, Biomass based distributed energy systems:opportunities and challenges.
[15] L.P.R. Pala, Q. Wang, G. Kolb, V. Hessel, Steam gasification of biomass with
Energy,Environment and Sustainability. Springer Singapore. Ed.: Agarwal, Avinash
subsequent syngas adjustment using shift reaction for syngas production: an Aspen
Kumar.ISSN: 2522-8366.
Plus model, Renew. Energy 101 (2017) 484–492, https://doi.org/10.1016/j.
[4] M. Gassner, Thermo-economic Process Model for Thermochemical Production of
renene.2016.08.069.
Synthetic Natural Gas ( SNG ) from Lignocellulosic Biomass, vol. 33, 2009,
[16] W. Doherty, A. Reynolds, D. Kennedy, The effect of air preheating in a biomass CFB
pp. 1587–1604, https://doi.org/10.1016/j.biombioe.2009.08.004.
gasifier using ASPEN Plus simulation, Biomass Bioenergy 33 (2009) 1158–1167,
[5] S.E. Hosseini, M.A. Wahid, A. Ganjehkaviri, An overview of renewable hydrogen
https://doi.org/10.1016/j.biombioe.2009.05.004.
production from thermochemical process of oil palm solid waste in Malaysia,
[17] L.E. Arteaga-Pérez, Y. Casas-Ledón, R. Pérez-Bermúdez, L.M. Peralta, J. Dewulf,
Energy Convers. Manag. 94 (2015) 415–429, https://doi.org/10.1016/j.
W. Prins, Energy and exergy analysis of a sugar cane bagasse gasifier integrated to
enconman.2015.02.012.
a solid oxide fuel cell based on a quasi-equilibrium approach, Chem. Eng. J. 228
[6] R.Ø. Gadsbøll, J. Thomsen, C. Bang-møller, Accepted Manuscript, https://doi.org
(2013) 1121–1132, https://doi.org/10.1016/j.cej.2013.05.077.
/10.1016/j.energy.2017.05.044, 2017.
[18] E. Shayan, V. Zare, I. Mirzaee, Hydrogen production from biomass gasi fi cation ; a
[7] P. Basu, Introduction, Biomass Gasif. Des. Handb (2010) 1–25, https://doi.org/
theoretical comparison of using di ff erent gasi fi cation agents, vol. 159, 2018,
10.1016/b978-0-12-374988-8.00001-5.
pp. 30–41, https://doi.org/10.1016/j.enconman.2017.12.096.
[8] T.K. Patra, K.R. Nimisha, P.N. Sheth, A comprehensive dynamic model for
[19] A. Abuadala, I. Dincer, G.F. Naterer, Exergy analysis of hydrogen production from
downdraft gasifier using heat and mass transport coupled with reaction kinetics,
biomass gasification, Int. J. Hydrogen Energy 35 (2010) 4981–4990, https://doi.
Energy 116 (2016) 1230–1242, https://doi.org/10.1016/j.energy.2016.10.036.
org/10.1016/j.ijhydene.2009.08.025.
[9] S. Sharma, P.N. Sheth, Air-steam biomass gasification: experiments, modeling and
[20] R. Hasanzadeh, T. Azdast, A. Doniavi, R.E. Lee, Multi-objective optimization of
simulation, Energy Convers. Manag. 110 (2016) 307–318, https://doi.org/
heat transfer mechanisms of microcellular polymeric foams from thermal-
10.1016/j.enconman.2015.12.030.
insulation point of view, Therm. Sci. Eng. Prog. 9 (2019) 21–29, https://doi.org/
[10] H. Li, X. Zhang, L. Liu, R. Zeng, G. Zhang, Exergy and environmental assessments of
10.1016/j.tsep.2018.11.002.
a novel trigeneration system taking biomass and solar energy as co-feeds, Appl.
[21] S. Mishra, A. Sharma, A. Kumari, Sanjay, Response surface methodology based
Therm. Eng. 104 (2016) 697–706, https://doi.org/10.1016/j.
optimization of air-film blade cooled gas turbine cycle for thermal performance
applthermaleng.2016.05.081.
prediction, Appl. Therm. Eng. 164 (2020) 114425, https://doi.org/10.1016/j.
[11] X. Zhang, H. Li, L. Liu, C. Bai, S. Wang, J. Zeng, X. Liu, N. Li, G. Zhang,
applthermaleng.2019.114425.
Thermodynamic and economic analysis of biomass partial gasification process,
[22] M.M. Sarafraz, I. Tlili, Z. Tian, M. Bakouri, M.R. Safaei, Smart optimization of a
Appl. Therm. Eng. 129 (2018) 410–420, https://doi.org/10.1016/j.
thermosyphon heat pipe for an evacuated tube solar collector using response
applthermaleng.2017.10.069.

12
S.A. Zaman et al. Biomass and Bioenergy 143 (2020) 105847

surface methodology (RSM), Phys. A Stat. Mech. Its Appl. 534 (2019) 122146, [33] M. Shahbaz, S. Yusup, A. Inayat, D.O. Patrick, A. Pratama, M. Ammar,
https://doi.org/10.1016/j.physa.2019.122146. Optimization of hydrogen and syngas production from PKS gasification by using
[23] M. Christwardana, D. Frattini, G. Accardo, S.P. Yoon, Y. Kwon, Optimization of coal bottom ash, Bioresour. Technol. 241 (2017) 284–295, https://doi.org/
glucose concentration and glucose/yeast ratio in yeast microbial fuel cell using 10.1016/j.biortech.2017.05.119.
response surface methodology approach, J. Power Sources. 402 (2018) 402–412, [34] D. Fozer, B. Kiss, L. Lorincz, E. Szekely, P. Mizsey, A. Nemeth, Improvement of
https://doi.org/10.1016/j.jpowsour.2018.09.068. microalgae biomass productivity and subsequent biogas yield of hydrothermal gasi
[24] C. Loha, H. Chattopadhyay, P.K. Chatterjee, Thermodynamic analysis of hydrogen fi cation via optimization of illumination, Renew. Energy 138 (2019) 1262–1272,
rich synthetic gas generation from fluidized bed gasification of rice husk, Energy https://doi.org/10.1016/j.renene.2018.12.122.
36 (2011) 4063–4071, https://doi.org/10.1016/j.energy.2011.04.042. [35] J.A. Okolie, S. Nanda, A.K. Dalai, J.A. Kozinski, Optimization and modeling of
[25] W. Doherty, A. Reynolds, D. Kennedy, Aspen Plus simulation of biomass process parameters during hydrothermal gasification of biomass model compounds
gasification in a Steam blown dual fluidised bed, in: A. Mendez-Vilas (Ed.), Mater to generate hydrogen-rich gas products, Int. J. Hydrogen Energy (2019), https://
Process Energy Commun Currres Technol Dev, Formatex Research Centre, 2013. doi.org/10.1016/j.ijhydene.2019.05.132.
[26] Z. Zhang, S. Pang, Experimental investigation of tar formation and producer gas [36] C. Loha, P.K. Chatterjee, H. Chattopadhyay, Performance of fluidized bed steam
composition in biomass steam gasification in a 100 kW dual fluidised bed gasifier, gasification of biomass - modeling and experiment, Energy Convers. Manag. 52
Renew. Energy 132 (2019) 416–424, https://doi.org/10.1016/j. (2011) 1583–1588, https://doi.org/10.1016/j.enconman.2010.11.003.
renene.2018.07.144. [37] S. Rapagnà, N. Jand, A. Kiennemann, P.U. Foscolo, Steam-gasification of biomass
[27] T. Renganathan, M.V. Yadav, S. Pushpavanam, R.K. Voolapalli, Y.S. Cho, CO 2 in a fluidised-bed of olivine particles, Biomass Bioenergy 19 (2000) 187–197,
utilization for gasification of carbonaceous feedstocks: a thermodynamic analysis, https://doi.org/10.1016/S0961-9534(00)00031-3.
Chem. Eng. Sci. 83 (2012) 159–170, https://doi.org/10.1016/j.ces.2012.04.024. [38] T. Detchusananard, K. Im-orb, F. Maréchal, A. Arpornwichanop, Analysis of the
[28] T. Nakyai, S. Authayanun, Y. Patcharavorachot, A. Arpornwichanop, sorption-enhanced chemical looping biomass gasification process: performance
S. Assabumrungrat, D. Saebea, Exergoeconomics of hydrogen production from assessment and optimization through design of experiment approach, Energy 207
biomass air-steam gasification with methane co-feeding, Energy Convers. Manag. (2020), https://doi.org/10.1016/j.energy.2020.118190.
140 (2017) 228–239, https://doi.org/10.1016/j.enconman.2017.03.002. [39] M. Inayat, S.A. Sulaiman, J.C. Kurnia, Catalytic co-gasification of coconut shells
[29] A.M. Parvez, I.M. Mujtaba, T. Wu, Energy, exergy and environmental analyses of and oil palm fronds blends in the presence of cement, dolomite, and limestone:
conventional, steam and CO2-enhanced rice straw gasification, Energy 94 (2016) parametric optimization via Box Behnken Design, J. Energy Inst. 92 (2019)
579–588, https://doi.org/10.1016/j.energy.2015.11.022. 871–882, https://doi.org/10.1016/j.joei.2018.08.002.
[30] K.J. Ptasinski, M.J. Prins, A. Pierik, Exergetic evaluation of biomass gasification, [40] A. Seçer, A. Hasanoğlu, Evaluation of the effects of process parameters on
Energy 32 (2007) 568–574, https://doi.org/10.1016/j.energy.2006.06.024. co–gasification of Çan lignite and sorghum biomass with response surface
[31] P. Mojaver, S. Khalilarya, A. Chitsaz, Multi-objective optimization using response methodology: an optimization study for high yield hydrogen production, Fuel 259
surface methodology and exergy analysis of a novel integrated biomass gasification (2020) 116230, https://doi.org/10.1016/j.fuel.2019.116230.
, solid oxide fuel cell and high-temperature sodium heat pipe system, Appl. Therm. [41] A. Inayat, M. Inayat, M. Shahbaz, S.A. Sulaiman, M. Raza, S. Yusup, Parametric
Eng. 156 (2019) 627–639, https://doi.org/10.1016/j. analysis and optimization for the catalytic air gasification of palm kernel shell
applthermaleng.2019.04.104. using coal bottom ash as catalyst, Renew. Energy 145 (2020) 671–681, https://doi.
[32] M. Inayat, S.A. Sulaiman, B.A. Bhayo, M. Shahbaz, Application of response surface org/10.1016/j.renene.2019.06.104.
methodology in catalytic co-gasification of palm wastes for bioenergy conversion [42] R. Bakari, T. Kivevele, X. Huang, Y.A.C. Jande, Simulation and optimisation of the
using mineral catalysts, Biomass Bioenergy 132 (2020) 105418, https://doi.org/ pyrolysis of rice husk: preliminary assessment for gasification applications, J. Anal.
10.1016/j.biombioe.2019.105418. Appl. Pyrolysis 150 (2020) 104891, https://doi.org/10.1016/j.jaap.2020.104891.

13