Renewable Energy 194 (2022) 912e925

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

An advanced, comprehensive thermochemical equilibrium model of a

downdraft biomass gasifier
A. Ibrahim *, S. Veremieiev , P.H. Gaskell
Department of Engineering, Durham University, Durham, DH1 3LE, UK

a r t i c l e i n f o a b s t r a c t

Article history: A stoichiometric model is formulated for predicting the syngas yield from the reduction zone of a
Received 13 January 2022 downdraft biomass gasifier. It incorporates the thermodynamic equilibrium of the global gasification
Received in revised form reaction, predicts the concentration of the minor gasification products of hydrogen sulphide and
20 April 2022
ammonia as the sulphur-based and nitrogen-based contaminants, respectively, and implements a new
Accepted 12 May 2022
empirical correlation, formulated using existing pertinent experimental data, to account for the mass tar
Available online 26 May 2022
yield. The governing set of model equations is solved in a fully coupled manner, with the boudouard
reaction employed to predict char output and the ammonia synthesis reaction used to predict ammonia
Keywords:
Ammonia concentration
production. The model does not require the use of correction factors and satisfactorily predicts the
Biomass gasification concentration of methane, a shortcoming that has tended to plague existing equilibrium models. The
Methane concentration syngas composition, tar and char yields, gasification temperature, cold gas efficiency and lower heating
Equivalence ratio value are obtained for various biomass feedstock with a specific ultimate analysis, for different equiva-
Gasification temperature lence ratios and varying moisture content. Where possible, predictions are compared with corresponding
Thermodynamic equilibrium model experimental data and found to be in very good agreement.
Crown Copyright © 2022 Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).

1. Introduction residence time in a biomass gasifier, has prompted the develop-

ment of mathematical models in order to characterise and predict
The adverse climate effects accompanying the continued use of syngas production. Such models provide a reliable representation
fossil fuels are well known. The large quantities of CO2 and other of the chemical and physical phenomena occurring within the
emissions produced, coupled with the urgent desire for a more gasifier and are effective at providing qualitative guidance on the
sustainable energy sector, has prompted the need for and wide- influence of design, operating and feedstock parameters on overall
spread use of dependable, affordable and cleaner alternatives. gasifier performance. Various approaches have been used to model
Biomass has become one such promising avenue, as its modern biomass gasification systems, namely equilibrium, kinetic, CFD and
application is considered a very optimistic clean-energy alternative. neural network alternatives, with equilibrium and kinetic versions
Although biomass production is invariably accompanied by the being the most extensively implemented. Although the latter can
release of CO2, it is widely viewed as a renewable energy source and generate arguably more reliable results, they are more complicated
is presently the only renewable that can directly replace fossil fuels to implement and have received far less attention compared to
due to its abundant availability, simple storage and transportation their equilibrium counterpart. Kinetic models incorporate reactor
requirements, and its synthesis of different fuels and chemicals [1]. kinetics and hydrodynamics, suitable for long residence times
The complicated process of gasification together with the when the reaction rate is very slow at a low reaction temperature
sensitivity of product distribution to the rate of heating and and used to estimate the gas composition and temperature profiles
within a biomass gasifier [2e4].
Equilibrium models, though comparatively simpler, are found to
describe the gasification process well, especially for downdraft and
Abbreviations: ER, Equivalence Ratio; MC, Moisture Content; LHV, Lower
Heating Value; HHV, Higher Heating Value; CGE, Cold Gas Efficiency; GFE, Gibbs in particular fluidised bed gasifiers which operate at close to
Free Energy. equilibrium conditions. Although thermodynamic equilibrium
* Corresponding author. cannot be reached within a gasifier, this assumption provides a
E-mail addresses: [email protected] (A. Ibrahim), s.veremieiev@ reasonable prediction of the final syngas composition and are often
durham.ac.uk (S. Veremieiev), [email protected] (P.H. Gaskell).

https://doi.org/10.1016/j.renene.2022.05.069
0960-1481/Crown Copyright © 2022 Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

employed as a simulation tool for processes whose duration is modify the equilibrium constants of chemical reactions in order to
normally quite long with respect to the reaction time scale, or obtain better agreement with experimental data. The limitation of
involve high gasification temperatures (>800
C). Their main ad- such an approach is that it is restricted to one set of input param-
vantages is that they are relatively easy to implement with fast eters since correction factors are relevant to specific operating
convergence. Two types of model can be formulated: stoichio- conditions only. The equilibrium model developed by Costa et al.
metric and non-stoichiometric. The former are based on calculating [25] was optimised [26] based on the work of Jarungthammachote
the thermodynamic equilibrium constants of an independent set of and Dutta [17], correcting equilibrium constants through multi-
chemical reactions which can be associated with the Gibbs free plication factors representing the degree of approach to equilib-
energy (GFE); the latter, are based on direct minimisation of the rium, with their value determined by solving a multi-objective
GFE of the chemical reaction [5]. optimisation problem via the genetic algorithm MOGA II. Initially
Various experimental studies have been conducted to better guessing a gasification temperature and providing it as an oper-
understand the parameters affecting gasification kinetics. Dahou ating input variable, allowed for the chronological determination of
et al. [6] carried out a thermogravimetric analysis of different the equilibrium constants and the syngas composition.
biomass samples to investigate biomass type and char preparation A parametric study of hydrogen production from steam gasifi-
influences on steam gasification. The selected samples included cation was performed by Abuadala et al. [27], with unreacted char
agricultural residues with their inorganic element compositions assumed to equal 5% of the biomass carbon content and tar
measured according to solid fuel standards. The char was produced modelled as benzene via the empirical correlation of Corella et al.
from in-situ pyrolysis of the samples and results indicated that [28]. Steam gasification is simulated by first varying the amount of
biomass type had a much larger influence on steam gasification biomass with the quantity of steam and gasifier temperature fixed,
kinetics than the conditions used to prepare the char. The same followed by changing the operating temperature while maintaining
authors [7,8] reviewed the role of inorganics on char gasification a constant amount of steam and biomass content. An identical
reactivity and reaction kinetics, concluding that some of the modelling approach has been adopted by several authors [29e31]
inherent inorganic elements of the biomass, such as potassium, when incorporating an empirical correlation to describe the carbon
silicon and calcium, have a significant influence. Furthermore, they fraction representing the amount of char, based on the assumption
established the role of potassium during steam gasification and that not all the carbon participates in equilibrium reactions. The
analysed the influence of potassium carbonate on pyrolysis and equilibrium model of Gagliano et al. [23] fixes the tar and char
gasification reactions, demonstrating that K2CO3 increases the re- yields at 4.5% and 10.5%, respectively [32,33], while utilising mul-
action rate both with and without contact with the biomass, and tiplicative factors and correlations. Ferreira et al.’s [34] review
the interaction of K2CO3 with the biomass is decisive for the ulterior article offers a detailed description of previously developed stoi-
gasification. Tamasiunas et al. [9] investigated olive biomass waste chiometric and non-stoichiometric thermodynamic equilibrium
for energy recovery using thermal arc plasma gasification, finding models for biomass gasifiers.
that the charcoal derived from olive pomace, generated as a waste Some of the above models have proved effective, acceptably
from the olive oil industry, had great potential for syngas produc- determining the syngas composition compared with experimental
tion (around 55% of total produced gas). Khiari et al. [10] presented studies. However, a good number of them either fail to calculate the
a comprehensive survey of already-well-established or future po- tar and char yields, fix them at specific values based on experi-
tential energy applications, including gasification and combustion mental studies, or resort to empirical correlations to determine the
of biomass derived chars, showing how they can be utilised in production of char. Furthermore, none of the equilibrium models
boilers to generate heat and/or steam to produce electricity. available in the literature allow prediction of all the key gasification
Through reforming and cracking, tars can be purified resulting parameters simultaneously, including the gasification temperature
in a higher quality of products. As such, the experimental studies and concentrations of H2S and NH3, via a fully coupled system of
and reviews mentioned above reflect an inherent flaw in most governing equations. Additionally, most existing equilibrium
existing mathematical models which do not consider the effects of models fail to explore the effects of important working parameters,
inorganic species on the biomass gasification process. such as the equivalence ratio (ER), and do not represent the influ-
The first recognised stoichiometric equilibrium model was ence of some of the operating input variables on H2S and NH3
developed by Zainal et al. [11] and applied to various feedstock to concentrations.
determine the composition of the resulting syngas and the oxygen In this paper, a thermodynamic equilibrium model is presented
content. An identical model was developed by Mountouris et al. which is not inhibited by the need for correction factors and
[12], focusing on the thermodynamic analysis of plasma gasifica- satisfactorily predicts the amount of methane concentration in the
tion, involving estimation of the resulting gas composition and producer gas, an output whose previous prediction has posed a
energy and exergy analyses. The gasification process within a flui- major flaw. The char yield is obtained via the boudouard reaction,
dised bed was investigated by Prins et al. [13] in order to describe comparing well with experimental data for char output. The tar
the gasification of fuels with different compositions of organic yield is calculated using a new empirical correlation generated
matter, adopting a quasi-equilibrium temperature approach. Sub- through an exponential best fit curve to existing experimental tar
sequently, Melgar et al. [14] and Sharma [15] developed models to data for downdraft gasifiers; a thermodynamic equilibrium reac-
predict syngas composition and reaction temperature, with the tion describing the formation of tars within a biomass gasifier does
latter being provided as an initial guess and calculated iteratively. not exist, tar being a non-equilibrium product. The model de-
The limitation of all these models is that they do not account for the termines the gasification temperature via the energy balance
production of tar and char, which are important outputs. A model through a coupled system of equations, a robust approach offering
adding a pyrolysis stage based on semi-empirical correlations, better convergence properties. Furthermore, it embodies the pro-
instead of using thermodynamic equilibrium calculations, is re- duction of H2S and comprises the formation of NH3 via the
ported in Puig-Arnavat et al. [16] and used to estimate the forma- ammonia synthesis reaction. The effects of both moisture content
tion of gas, char and volatiles, considering tar and char leaving the (MC) and ER on the syngas composition, tar and char yields, gasi-
gasifier as a percentage of their value in the pyrolysis stage. fication temperature, the lower heating value (LHV) and the cold
Subsequently, several authors [17e24] have followed a very gas efficiency (CGE) are explored and the results compared with
similar modelling approach, introducing correction factors to those from the literature. Finally, particular care has been taken to
913
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

ensure that everything is consistently defined to ensure


reproducibility. a b l
mgas ¼ 1 þ  þ þ d  ER: (4)
4 2 2
For completeness, the derivation of Eq. (4) is provided in
2. Problem formulation and method of solution
Appendix A.
The atom balance equations of C, H, O, N and S derived from the
The assumptions underpinning the model are that:
global gasification reaction, are:
1. The gasifier operates under steady state conditions;
2. The residence time is infinite; C : x2 þ x3 þ x4 þ 6x9 þ x10 ¼ 1; (5)
3. Nitrogen is considered inert;
4. The process is adiabatic; H : 2x1 þ 4x4 þ 3x6 þ 2x7 þ 2x8 þ 6:2x9 ¼ a þ 2w; (6)
5. The contribution of char and ash to the energy balance equation
is negligible;
6. The total pressure is assumed constant (i.e. P0 ¼ 101, 325 Pa), O : x2 þ 2x3 þ x8 þ 0:2x9 ¼ b þ w þ 2mgas ; (7)
however the partial pressure of the syngas components is not
and is determined via their chemical equilibrium;
N : 2x5 þ x6 ¼ l þ 7:52mgas ; (8)
7. reactants are at standard temperature (i.e. T0 ¼ 298.15 K), while
products are at the unknown T.
8. tar, at standard conditions, is a mixture of liquid and gas, with its S : x7 ¼ d; (9)
thermodynamic properties taken to be those of benzene;
9. char is assumed to have the same thermodynamic properties as respectively. The specific molar heat capacity for a chemical
graphite. component is determined from the following [36] empirical
relationship:
Starting from an ultimate analysis of the biomass feedstock, on a
dry basis its chemical composition CHaObNlSd is determined via the Cp;i ðTÞ ¼ RðAi þ Bi T þ Ci T 2 þ Di T 2 Þ; (10)
following expressions:
where subscript i denotes a particular chemical component and Ai,
yH  MC y  MC
a¼ ; b¼ O ; Bi, Ci and Di represent the dimensionless thermodynamic empirical
yC  MH yC  MO constants of each; their values are provided in Table 1. The enthalpy
(1)
y  MC y  MC for a specific chemical component [37], while ignoring the effect of
l¼ N ; d¼ S ;
yC  MN yC  MS pressure for an incompressible material (i.e. solid char), is obtained
from:
where yC, yH, yO, yN and yS are the percentage weights of C, H, O, N
and S present; a, b, l and d are the number of atoms of the

ðT
respective chemical species per one atom of carbon in the biomass,
B C D
hi ðTÞ ¼ hf ;i þ Cp;i dT ¼ Ji þ R Ai T þ i T 2 þ i T 3  i ; (11)
while MC, MH, MO, MN and MS are their molar masses in kg/mol. The 2 3 T
molar mass of biomass is expressed as: T0

MC where R is the ideal universal gas constant equal to 8.314 J/molK

Mbm ¼  100%: (2) and Ji is a constant resulting from integrating the right hand side of
yC

Eq. (11) and incorporating hf ;i , namely:
The following global gasification reaction forms the basis for
determining the various product species: !

Bi T02 Ci T03 Di
Ji ¼ hf ;i  R Ai T0 þ þ  ; (12)
CHa Ob Nl Sd þ wH2 O þ mgas ðO2 þ 3:76N2 Þ/x1 H2 þ x2 CO 2 3 T0
þx3 CO2 þ x4 CH4 þ x5 N2 þ x6 NH3 þ x7 H2 S þ x8 H2 O


þx9 C6 H6:2 O0:2 þ x10 C; where hf ;i is the enthalpy of formation at standard conditions for
(R1) reactants and products - see Table 1 - for biomass given in
Appendix A. (R1) is at a constant total pressure p0, therefore its
where w is the number of moles of H2O per 1 mol of biomass and is enthalpy balance is equal to zero, namely:
calculated from the biomass MC on a wet basis as follows:
X
DHðTÞ ¼ yi hi ðTÞ ¼ 0; (13)
Mbm  MC i
w¼ ; (3)
ð2MH þ MO Þ  ð1  MCÞ
where yi is the stoichiometric number which is positive for prod-
where mgas is the number of moles of gasifier input air per 1 mol of ucts, yi ¼ xi for i ¼ 1  10, and negative for reactants, yi ¼ (1,  w, 
P
biomass; the terms xi, for i ¼ 1  10, indicate the number of moles of mgas,  3.76mgas) and the summation i is repeated over all
the various chemical species produced, i ¼ 1  7 indicate compo- gaseous and solid components. However, note also the enthalpy of
nents of dry producer gas which forms the gasifying medium and moisture is taken in liquid state while the enthalpy of char and ash
consists of 79% N2 and 21% O2; the number of moles of N2 in the is ignored.
reactants is calculated as the amount of N2 relative to the amount of The entropy of a chemical component at temperature T can be
O2 in air. The chemical formula used to represent tar and char are written [37]:
C6H6.2O0.2 [35] and C, respectively and mgas is calculated as a
function of ER, namely: C for an incompressible material (i.e. solid char) as:
914
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

Table 1
Constants for the molar heat capacity of chemical species and their corresponding enthalpy of formation, and GFE of formation [36].

103B, K1 106C, K2 105D, K2

Chemical Species A Tmax hf ; J=mol gf ; J=mol

Hydrogen 3.25 0.422 e 0.083 3000 e e

Carbon Monoxide 3.38 0.557 e 0.031 2500 110,525 137,169
Carbon Dioxide 5.46 1.047 e 1.157 2000 393,509 394,359
Methane 1.7 9.081 2.164 e 1500 74,520 50,460
Nitrogen 3.28 0.593 e 0.04 2000 e e
Ammonia 3.58 3.02 e 0.186 1800 46,110 16,450
Hydrogen Sulphide 3.93 1.49 e 0.232 2300 20,630 33,560
Water Vapour 3.47 1.45 e 0.121 2000 241,818 228,572
Tar 2.06 39.064 13.3 e 1500 82,930 129,665
Char 1.77 0.771 e 0.867 2000 e e
Nitric Oxide 3.39 0.629 e 0.014 2000 90,250 86,550
Sulphur Dioxide 5.7 0.801 e 1.015 2000 296,830 300,194
Water Liquid 8.71 1.25 0.18 e 373.2 285,830 237,129

g 
*
i g R
Cp;i dh d ¼ d i þ dpi ; (21)
ds*i ¼ dT ¼ i ; (14) T T pi
T T


p
ðT ðT gi ðT; pi Þ ¼ gi* ðTÞ þ RTlog i : (22)

Cp;i
dhi p0
s*i ðTÞ ¼ sf ;i þ dT ¼ sf ;i þ ; (15)
T T If a reversible chemical reaction is at chemical equilibrium, then
T0 T0
its GFE balance equals zero, and as such:

Y
pi yi


where sf ;i is the entropy of formation at standard conditions and X X
DGðT; pÞ ¼ yi gi ðT; pi Þ ¼ yi gi* ðTÞ þ RTlog
i i i
p0
C for an ideal gas as:
¼ DG* ðTÞ þ RTlog kðTÞ ¼ 0: (23)
R
dsi ¼ ds*i  dpi ; (16) Accordingly, the thermodynamic equilibrium constant, k(T), of
pi the reaction is determined from Dalton's law as follows:

 Y
p i y i Y
xi yi
p log kðTÞ ¼ log ¼ log ; (24)
si ðT; pi Þ ¼ s*i ðTÞ  Rlog i ; (17) i
p0 i
Ntot
p0
Q
where the multiplication i is repeated over the reactions gaseous
where pi is the partial pressure of a gaseous component. The GFE of P9
a chemical component is calculated [37]: components only. Ntot ¼ i¼1 xi is the number of moles of the raw
producer gas at temperature T and the equilibrium constant is
C for an incompressible material (i.e. solid char) as: calculated via the standard gibbs free energy, DG*(T), of a reaction:



 DG* 1 X *
g* h h dh h log kðTÞ ¼  ¼ y g ðTÞ; (25)
d i ¼ d i  ds*i ¼ d i  i ¼  2i dT; (18) RT RT i i i
T T T T T
P
where the summation i is repeated over all the reactions gaseous
and solid components.
ðT
hi ðTÞ

BT Four independent equilibrium reactions are implemented to
gi* ðTÞ ¼ gf ;i  T dT ¼ Ji  RT Ai logðTÞ þ i
T2 2 model the gasification process: the methanation, water-gas shift,
T0 boudouard and ammonia synthesis reactions:
!
C T2 D
þ i þ i2 þ Ii ; (19) C þ 2H2 4CH4 ; (R2)
6 2T

where gf ;i is the GFE of formation at standard conditions given in CO þ H2 O4CO2 þ H2 ; (R3)
Table 1 and Ii is another integration constant determined from Eq.
(19) at standard conditions, namely: C þ CO2 42CO; (R4)

! N2 þ 3H2 42NH3 ; (R5)

Ji  gf ;i Bi T0 Ci T02 Di
Ii ¼  Ai logðT0 Þ þ þ þ 2 ; (20)
RT0 2 6 2T0 respectively. (R2) progresses rapidly with hydrogen being reduced
to form methane in the presence of carbon; (R3) increases the
hydrogen concentration at the expense of carbon monoxide and
describes the equilibrium between the two in the presence of
C for an ideal gas as: water. In the absence of steam and in the presence of air as a
915
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

gasifying medium, (R4) is dominant; therefore, it is implemented in

order to describe the conversion of char to carbon monoxide in the
presence of carbon dioxide. As reported in Gambarotta et al. [38],
ammonia is the most abundant nitrogen-based syngas contami-
nant and its production is accounted for via (R5) [39]. Accordingly,
the equilibrium constants, kR2, kR3, kR4 and kR5 for the above
chemical reactions are obtained from Eq. (24) as follows:

x4  Ntot x  x3
kR2 ðTÞ ¼ ; kR3 ðTÞ ¼ 1 ;
x21 x2  x8
ðx2 Þ2 x2  Ntot
2
kR4 ðTÞ ¼ ; kR5 ¼ 6 ; (26)
x3  Ntot x5  x31

and at temperature T are calculated from Eq. (25).

Tar is modelled by taking its thermodynamic properties to be those
of benzene. To account for tar production, a unique empirical cor-
relation is generated in the form of an exponential best fit curve
using appropriate tar data, gathered from experiments performed
on downdraft biomass gasifiers [40e44] using gas chromatography
and separation techniques. While corresponding data is available Fig. 1. Downdraft gasifier tar yield as a function of ER: showing corresponding
for other than downdraft gasifiers, it is either not provided in the experimental data and best-fit relationship, Eq. (28), based on the datasets [40e43].

preferred mass tar yield format or can be converted into the same.
Mass tar yield on dry basis (% d.b.) offers a coherent dimensionless
models, such as [26,27,31], which adopted Correla et al.’s [28]
form that removes any dependence on the dimensional properties
correlation, generated for a fluidised bed gasifier, and that of Kir-
of biomass and gasifier length scales. Accordingly, the above tar
sanovs et al. [46] who failed to show how the tar model fitted with
data (denoted in units of g/Nm3) was converted to mass tar yield
experimental data, the above curve-fit provides a reliable predic-
(wt.%) as follows:
tion of the tar yield as the experimental tar data was gathered from
the outputs of downdraft biomass gasifiers only.
tar yield
tar content ¼ ; The system of Eqs. (5)e(9) as well as Eqs. (13), (26) and (29)
syngas yield  100%
consist of 11 equations for the 11 unknowns x1, x2, x3, x4, x5, x6, x7,
Ngas  Vm x8, x9, x10 and T, and are solved numerically using Matlab's built-in
where syngas yield ¼ ; (27)
Mbm function fsolve - details of its implementation are provided in
P Ref. [47]. For all the results generated and discussed subsequently,
where Ngas ¼ 7i¼1 xi is the number of moles of the dry producer the 11 unknowns were assigned the same starting values namely
gas at standard temperature and Vm ¼ 22.4 Litres is the volume for (xi, i ¼ 1  10, T) ¼ (0.05, 0.06, 0.03, 0.03, 0.07, 0.05, 0.02, 0.07, 0.081,
an ideal gas at standard temperature and pressure. This was done 0.091, 1000.0 K). One of the advantages of the current approach is
for all the associated ER values, obtaining the amount of tar per unit determination of the gasification temperature, T, via a more robust
mass of biomass, thus achieving consistency in the determination fully coupled solver offering better convergence properties.
of the mass tar yield between the respective experimental studies. The molar LHV of the dry producer gas at standard temperature
The resulting data points are plotted in Fig. 1. The tar yield is (MJ/mol) is given by:
expressed as a function of ER only since it is the most important
operating condition influencing biomass gasification, affecting both
the producer gas composition and T; since T is an output parameter 1 X7
LHVgas ¼ x LHVi ;
i¼1 i
(30)
its effect is not reflected in the resulting mass tar yield relationship. Ngas
The effect of the other operating condition, MC, is much less sig-
nificant [39,45]. Shown also is the curve fit obtained using the where LHVi is the molar LHV of component i at standard temper-
experimental data of [40e43] and Matlab's Curve Fitting Toolbox ature calculated from their complete combustion reactions as
app; the data of [44] is shown for completeness, being clearly follows:
inappropriate for this purpose due to its uncorrelated nature. The
resulting empirical relationship in equation form is given by:




LHV1 ¼ hf ;1  hf ;8 ; LHV2 ¼ hf ;2  hf ;3 ; LHV3 ¼ 0;
tar yield ¼ 0:8212expð3:281ERÞ  100%; (28)



LHV4 ¼ hf ;4  hf ;3  hf ;8 ; LHV5 ¼ 0; (31)







with the molar tar yield, as used in the model formulation, given LHV6 ¼ hf ;6  hf ;5  hf ;8 ; LHV7 ¼ hf ;7  hf ;8  hf ;SO2 :
by:
The volumetric LHV of producer gas at standard temperature
M (MJ/mol) is given by the following:
x9 ¼ 0:8212expð3:281ERÞ  bm for 0:155  ER  0:415;
Mtar
(29) LHVgas
LHVgasvol ¼ : (32)
Vm
where Mtar ¼ 6MC þ 6.2MH þ 0.2MO is the molar mass of tar.
The curve decreases with increasing ER, showing a low tar yield, The CGE [48], which is the ratio of the LHV of the syngas and the
which is to be expected since downdraft gasifiers normally produce LHV of the biomass feedstock, as calculated in Appendix A is given
a low tar content (<1g/Nm3) [39]. Unlike previous equilibrium as follows:
916
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

et al. [49] and Barrio et al. [50]; any differences can be attributed to
LHVgas the fact that the model incorporates tar, char and accounts for
CGE ¼  100%: (33)
LHVbm minor gasification products. Shown also are the corresponding
solutions obtained by Ref. [23] for the same feedstocks; noting that
both sets of predictions overestimate the hydrogen and underes-
timate the methane concentrations. Such behaviour is typical of
3. Results and discussion
equilibrium models which others [56,57], justify on the basis that
methane produced in the low temperature zone can bypass the
The chemical and physical properties of the feedstock consid-
reaction zone and avoid reduction - see also [24]. The predicted
ered, as a necessary pre-requisite of any biomass gasification study,
methane concentration for rubberwood and wood pellets is found
are obtained through characterisation tests; the ultimate and
to be 1.03% and 0.78%, respectively showing that the current model
proximate analysis of the feedstocks and their higher heating value
is better at predicting the methane concentration when compared
(HHV) calculated via Eq. (A.11) - see Appendix A - are provided in
with experimental data whilst demonstrating a more sophisticated
Table 2.
syngas composition, the objective of developing biomass gasifiers
being to increase the amount of H2 and CO while maintaining a
3.1. Model validation relatively low amount of CO2. For the results obtained by Ref. [23],
the model was calibrated to achieve a more favourable outcome,
First, the producer gas composition is compared with a set of accomplished by introducing correction factors for (R2), thus
experimental data and corresponding model predictions, high- moving the reaction equilibrium towards more CH4 and less H2
lighting the satisfactory prediction of the methane concentration, production and for (R3), thus moving the reaction equilibrium to-
with the species concentration of component i at standard tem- wards more CO and less H2 production. Nevertheless, the current
perature is given by: model is better able to predict the concentration of CH4, and in
some cases the syngas; also comparison with the experimental data
xi
concentration ¼  100%: (34) of [50] shown in Fig. 2b is better for the current model than for the
Ngas
equilibrium model of [23].
Next, the predicted temperature T is compared with experi- Fig. 3 compares the producer gas composition obtained using
mental data, showing its influence when the operating parameters the current model with experimental data from Refs. [51,52]. Fig. 3a
are varied. Finally, the char yield is calculated as a percentage of is for the case of rice husk at an ER of 0.45 and moisture free basis,
biomass on a dry basis (%d.b.) as follows: showing good agreement with the data of Yoon et al. [51], espe-
cially for the case of N2 concentration. The model leads to a slightly
x10  MC higher CO concentration but still aligns well with the experimental
char yield ¼  100%; (35)
Mbm study. Fig. 3b relates to the gasification of bamboo at an ER of 0.3
and MC of 10% showing good overall agreement with the data of
and compared with existing experimental data, clearly showing the Dutta et al. [52]. Fig. 3c is for the gasification of neem at an ER of 0.3
reliability of implementing the boudouard reaction. and an MC of 20%, showing good agreement between the pre-
dictions and the experiments of [52].
3.1.1. Producer gas composition A detailed comparison of the predicted syngas composition and
The results of this section compare predicted and experimen- those measured by Ref. [49] is provided in Table 3, from which it is
tally observed producer gas compositions for different feedstock as clear that as the MC decreases and the ER increases, the H2 and CO2
a function of MC and ER. The comparison is based on the main concentrations decrease. Obviously, the amount of N2 increases
output gases forming the producer gas of a typical downdraft with increasing ER, resulting from the higher amount of air present
gasifier (i.e. H2, CO, CO2, CH4 and N2). in the system. For the same operating conditions, the syngas
It can be seen from Fig. 2a and b that the predictions are in composition predicted by the model compares well with the cor-
reasonably good agreement with the experimental data of Jayah responding measured values, but with slightly higher production

Table 2
The ultimate and proximate analysis of different biomass feedstocks.

Biomass Feedstock Ultimate Analysis Proximate Analysis

yC(wt.%d.b.) yH(wt.%d.b.) yO(wt.%d.b.) yN(wt.%d.b.) yS(wt.%d.b.) yFC(wt.%d.b.) yVM(wt.%d.b.) yASH(wt.%d.b.) HHV(MJ/kg d.b.)

Rubberwood [49] 50.6 6.5 42 0.2 e 19.2 80.1 0.7 20.98

Wood pellets [50] 50.67 6.18 40.97 2 0.18 e e 1 20.7
Rice husk [51] 33.14 5.14 37.20 0.55 0.1 20.1 60 23.85 15.81
Bamboo [52] 48.39 5.86 39.21 2.04 e 15.2 80.3 4.5 19.62
Neem [52] 45.1 6 41.5 1.7 e 12.65 81.75 5.6 18.38
Pellets [53] 46.97 5.82 39.52 0.06 0.31 e e 0.85 19.18
Wood chips 1 [53] 49.99 5.24 41.07 0.17 0.67 e e 0.06 19.36
Wood chips 2 [53] 48.51 5.51 36.86 0.10 0.43 e e 0.89 19.64
Wood chips 3 [53] 46.83 5.92 39.84 0.06 0.33 e e 0.41 19.23
Wood chips 4 [54] 49.44 6.06 43.51 e e e e 1 19.87
Lignite [55] 37.80 4.93 40.394 1.625 0.141 31.03 42.07 15.11 16.37
Mixed wood chips [41] 48.77 5.85 44.52 0.05 0.01 12.8 75.8 0.8 17.3
Softwood pellets [41] 49.20 6.20 44.06 0.08 0.06 15.2 79.2 0.4 19
Rape straw pellets [41] 39.60 5.60 48.54 0.78 0.08 17.2 62.5 5.4 16.2
Poultry litter pellets [41] 43.98 5.16 31.98 4.63 0.75 15.3 63.6 13.5 16.8
Sewage sludge - sawdust pellets [41] 41.08 5.51 26.90 3.77 0.94 14.3 59.5 21.8 17.8
Forest waste [38] 53.1 6.2 36.62 1.11 0.07 e e 2.9 19.2

917
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

Fig. 2. Comparison between predicted ([23] and the model) and experimentally obtained [49,50], producer gas composition for (a) rubberwood (MC ¼ 16%, ER ¼ 0.314) and (b)
wood pellets (MC ¼ 8% and ER ¼ 0.266).

Fig. 3. Comparison of the producer gas composition obtained experimentally [51,52], and predicted by the model for (a): rice husk (MC ¼ 0%, ER ¼ 0.45), (b): bamboo (MC ¼ 10%,
ER ¼ 0.3) and (c): neem (MC ¼ 20%, ER ¼ 0.3).

levels of H2 and CO. equilibrium models, the ER has the biggest influence on the
Finally, the syngas composition predicted by the model is
compared with a wider range of experimental datasets performed
on state-of-the-art small-scale downdraft biomass gasifiers, Table 3
currently in operation [53], as shown in Fig. 4. This was achieved for Syngas composition (%), for different operating conditions, predicted by the model
the different feedstocks by evaluating the producer gas composi- and obtained experimentally [49].

tion at the respective ER and MC for the specific technology. The MC % ER H2 CO CO2 CH4 N2
predicted results are in fairly good agreement with those of the 18.5 0.33 17.2 19.6 9.9 1.4 51.9
experimental studies; the differences can reasonably be attributed Jayah et al [49] 16 0.35 18.3 20.2 9.7 1.1 50.7
to the fact that there are other major operating conditions which 14.7 0.38 17.2 19.4 9.7 1.1 52.6
affect the gas composition in operational downdraft gasifiers such 18.5 0.33 24.50 22.93 10.53 1.08 40.92
Prediction 16 0.35 23.83 23.59 10.01 0.96 41.57
as thermal and electrical efficiencies and gasifier design. For
14.7 0.38 22.21 26.33 7.96 0.67 42.80

918
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

Fig. 4. Comparison between predicted and experimentally obtained [53] producer gas composition for (a): pellets (MC ¼ 6.32%, ER ¼ 0.26), (b): wood chips 1 (MC ¼ 3.39%,
ER ¼ 0.25), (c): wood chips 2 (MC ¼ 10.30%, ER ¼ 0.29) and (d): wood chips 3 (MC ¼ 7.65%, ER ¼ 0.26).

producer gas composition, noting the proportional relationship increasing ER, which is expected due to the higher amount of air
between ER and N2. The predicted CH4 agrees well with the cor- present in the system, thereby enhancing the extent of the com-
responding experimental values - especially in the case of feed- bustion reaction which ultimately increases the amount of heat
stocks (a) and (c) - highlighting the uniqueness of the current released, thus facilitating T. A mixture of lignite - sawdust bri-
model as this has never been achieved by previous equilibrium quettes (70:30 wt%) is used as feedstock and minor differences can
models without the introduction of empirical correction factors. As be attributed to the fact that in experimental studies such as [55],
for the H2 prediction, in some cases it is overestimated but for the various other gasifier conditions affect T, compared with equilib-
obvious reasons already mentioned. rium models where only the ER and MC have a strong influence.
Evidently, Fig. 5a shows a good comparison of the model's T with
that obtained experimentally by Ref. [55].
3.1.2. Gasification temperature Fig. 5b shows the influence of MC on the predicted T compared
Fig. 5a provides a comparison of the predicted T against its with that obtained experimentally by Ref. [54] at an ER of 0.25; both
counterpart from the experimental study of Upadhyay et al. [55]. show a decrease in T with increasing MC, which is attributable to
The two profiles have the same trend with T increasing with

Fig. 5. Variation of T with ER for (a) lignite (MC ¼ 12%) and with MC for (b) wood chips 4 (ER ¼ 0.25).

919
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

the fact that an increase in MC in the feedstock favours endo- of operating conditions such as MC and ER for rubberwood and
thermic behaviour which tends to decrease the reaction tempera- wood pellets. This is followed by considering the case of ammonia
ture, consequently decreasing the T [58]. There is a significant and hydrogen sulphide concentration as components of the pro-
difference in T at an MC of around 21%, but good agreement be- ducer gas composition.
tween the model and the experimental investigation can be
favourably argued for T between the MC range. As explained, the
3.2. Effect of moisture content
factors influencing T in experimental investigations constitute
additional operating parameters. For example [54], demonstrates
MC is an essential property of biomass and an important oper-
also the primary and secondary air flow, fuel supply rate as well as
the thermal capacity of the gasifier, as all having an affect on T. ating parameter when developing a gasifier since it can strongly
influence the conversion of biomass into energy. Increasing levels
of moisture affects the self-sustainability of the combustion pro-
3.1.3. Char yield cess, ultimately decreasing the heating value of the syngas and
(R4) is used to account for the char yield prediction, due to it reducing the efficiency of the process. Furthermore, high levels of
being the more dominant reaction when using air as gasifying moisture reduce the oxidation temperature leading to an incom-
medium [39], rather than using an empirical correlation to describe plete cracking of the hydrocarbons produced during pyrolysis.
the carbon fraction, a factor representing the amount of carbon that Tolerable biomass moisture level limits range from 15% to roughly
participates in equilibrium reactions, as utilised in previous ther- 55% [5].
modynamic equilibrium models - see for example [24,29e31]. In In Fig. 7, the effect of MC in rubberwood on the composition of
order to confirm the viability and reliability of implementing the the resulting producer gas is revealed. As MC increases from 0% to
boudouard reaction, the predicted char yield is compared with the 40%, the percentage of CO2 increases from around 5%e20%, while
results obtained from the experimental investigation of a down- the percentage CO decreases from approximately 31%e11%. The N2
draft biomass gasifier by Ref. [41] in Fig. 6 for a variety of biomass concentration remains almost constant with increasing MC, as
feedstock and different operating conditions. On the whole two are expected, while the CH4 produced varies marginally from 0.5% to
in comparatively good agreement. In thermodynamic equilibrium around 3.5%, indicating the improved prediction of CH4 by the
models, ER is capable of affecting main- and by-product yields current equilibrium model relative to experimental studies of
significantly and normally, the char yield decreases with ER since downdraft biomass gasifiers. The H2 concentration increases
increased ER favours enhanced bed temperature, enhancing char slightly from 22% to around 26% with increasing MC in agreement
reactions through (R3) and (R4) and consequently a higher amount with the findings of [11,23].
of gas is formed. As such, this would explain the extreme variation Fig. 8a and b consider the LHV and the CGE, for both rubber-
in the char yield comparison for softwood pellets as it was carried wood and wood pellets, respectively, as a function of the MC and a
out at an ER of 0.2. For both model and experiment, the lowest char fixed ER of 0.326. Both decrease with increasing MC. This is antic-
yield occurs for the gasification of mixed wood chips. For the ipated considering the greater reduction in CO concentration
remaining feedstocks of a more pelletised nature, the char yield compared to the slight increase in the H2 concentration as the MC
increases, but varies depending on the amount of ash in the raw increases, as shown in Fig. 7 for rubberwood. In the case of the CGE,
material, the ER and T. The amount of ash in mixed wood chips is increasing MC decreases the efficiency of the gasifier as T decreases
0.8, which is 24 times less than the amount of ash found in soft- with MC resulting in a weaker production of syngas.
wood pellets, thus explaining the extreme variation in the pre- The corresponding variation of tar content with MC is shown in
dicted char yield between the feedstocks. Fig. 9, that for rubberwood being slightly greater than that of wood
Attention is now turned to using the model to explore the effect pellets for the same operating conditions. The decrease in tar
content with MC is in line with the results of [59] who studied the
influence of MC on the tar characteristics of wood pellet feedstock
in a downdraft gasifier using gas chromatography, mass spec-
trometry and gravimetric analysis to identify and analyse the tar

Fig. 6. Comparison of predicted char yield with that obtained experimentally by

Ref. [41] for mixed wood chips (ER ¼ 0.21), softwood pellets (ER ¼ 0.20), rape straw
pellets (ER ¼ 0.29), poultry litter pellets (ER ¼ 0.41) and sewage sludge - sawdust Fig. 7. Predicted variation of syngas composition with MC for rubberwood
pellets (ER ¼ 0.39). (ER ¼ 0.326).

920
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

Fig. 8. Predicted variation with increasing MC for (a) LHVgas and (b) CGE for rubberwood and woodpellets (ER ¼ 0.326).

Fig. 10. Predicted variation of producer gas composition with ER for rubberwood
(MC ¼ 18.5%).

Fig. 9. Predicted variation of tar with MC for rubberwood and woodpellets

(ER ¼ 0.326).

Fig. 11a explores the variation of the LHV for rubberwood with
samples. MC for different ER values. It is shown that the reduction of high
heating value gases such as H2 and CH4 and heavier hydrocarbons,
in addition to the dilution effects of N2, decreases the LHV of the
3.3. Effect of equivalence ratio
producer gas with increasing ER. The same outcome was obtained
by Cho et al. [62], who reported that increasing ER from 0.21 to 0.41
ER is the main operating parameter influencing biomass gasi-
led to a reduction of LHV from 13.42 to 7.05 MJ/Nm3. The variation
fication, which is considered as a fuel rich combustion when the ER
of CGE with MC for increasing values of ER, Fig. 11b, exhibits a
is less than 1. Typical values usually range from 0.2 to 0.4, and
similar trend to the LHVgas with CGE decreasing with increasing ER.
strongly affect the gasification process as it determines the tem-
A similar result was reported by Ref. [63] with the heating value
perature of the system, oxygen availability, the syngas output and
decreasing from 11.3 to 5.17 MJ/Nm3 for an increase of ER from 0.2
thus the gas composition and heating value in addition to the tar
to 0.45. At a low ER (0.25), the low quality syngas results in an
yield [5,39].
increase followed by a sudden decrease of H2 values, in line with
The influence of the ER on the syngas composition at 18.5% MC
the results of [11,15] and with the reasoning of (R3). As this hap-
for rubberwood is shown in Fig. 10. The H2 concentration decreases
pens, and since the syngas LHV is partly influenced by H2 produc-
while that of CO increases with increasing ER, which is due to the
tion, the syngas LHV demonstrates this trend.
fact that a higher ER will ultimately result in a higher T which fa-
The predicted variation of tar content with increasing ER for
cilitates the endothermic reaction (i.e. the formation of CO2 and
both rubberwood and wood pellets is considered in Fig. 12. As
CO). The amount of CO2 correspondingly decreases due to the
mentioned earlier, increasing ER enhances T as a result of a higher
increased T and the boudouard reaction. The CH4 concentration is
input air within the gasifier. Consequently, a higher quality syngas
found to decrease fractionally from around 2.5%e0.75%; this is due
is produced resulting in a reduction of the tar content. The
to the fact that the higher T facilitates the rate of the water-gas shift
increased T also facilitates tar cracking and thus the tar may
and the boudouard reactions but decreases the rate of the metha-
decompose into lighter gases which may assist in increasing the
nation reaction. The decrease in H2 concentration is in-line with the
combustible products in the syngas, ultimately decreasing the tar
results of [55,60,61]. Finally, the N2 concentration in the producer
content [55,64].
gas increases because the N2 is mainly in air and at a higher ER,
more air is present in the system.
921
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

Fig. 11. Predicted variation for different ER values for (a) LHVgas and (b) CGE, for rubberwood.

production of H2S [18], the production of NH3 is novel in stoi-

chiometric equilibrium modelling.
Fig. 13 show the variation in NH3 and H2S concentration for
different ER values and increasing MC, respectively, for forest waste
residue. Fig. 13a provides a comparison with the modelling results
of [38] showing that, at an MC ¼ 40%, there is extremely good
agreement between the NH3 and H2S concentrations predicted by
the model and those estimated by Ref. [38]. The NH3 concentration
is about 0.0122% for the model and approximately 0.015% for [38] at
an ER ¼ 0.1275 and displays a sharp decrease for both models as the
ER increases towards 0.255. The H2S concentration, however, is far
less and remains almost constant with changes in ER as depicted,
decreasing slightly with each ER increase. For increasing ER, T is
enhanced and more of the biomass is transformed into syngas
which in turn decreases the amount of volatiles, tar, char and
contaminating gases such as NH3 and H2S. Also, note that H2S is
present in very small quantities in the output of the gasifier. This is
because sulphur in feedstock is normally absent or contained in
traces, which has prompted authors to neglect it as an output since
its value does not contribute significantly to the main products of
Fig. 12. Predicted variation of tar production for different values of ER for both rub- the gasifier. Fig. 13b shows the variation of NH3 and H2S concen-
berwood and wood pellet feedstocks (MC ¼ 18.5%). trations with increasing MC, from which it is evident that the NH3
concentration increases with increasing MC, while that of H2S re-
mains almost constant.
3.4. Minor gasification products

This section focuses on the minor gasification products encap- 4. Conclusion

sulated within the model, H2S and NH3. The purpose of including
them is to estimate the concentrations of the primary sulphur- This paper provides a comprehensive equilibrium model for
based and primary nitrogen-based contaminants within biomass understanding the gasification process in a downdraft gasifier,
gasification. Although some previous authors have considered the incorporating a global reaction which includes all the gaseous

Fig. 13. Predicted variation in NH3 and H2S concentration for forest waste residue for (a) different ER values compared with the model results of [38] (MC ¼ 40%) and (b) increasing
MC (ER ¼ 0.326).

922
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

species and the yields of tar and char, the latter of which is deter- Appendix A. Stoichiometric Combustion Reaction
mined by implementing the boudouard reaction. In addition, the
gasification temperature is determined via the solution of a fully The derivation of the ER equation [65,66] comes from the global
coupled equation set resulting in a solver with improved conver- reaction for combustion of biomass in oxygen and is defined in
gence properties. To account for tar production in the model a new Section 2, where mgas is the actual number of molecules of oxygen
exponential best fit curve is generated, based on previous experi- in the system. According to the global combustion reaction for a
mental data for tar creation in downdraft gasifiers, and the particular feedstock:
resulting correlation is implemented to account for the molar tar
yield. CHa Ob Nl Sd þ mcomb O2 /y1 CO2 þ y2 H2 O þ y3 SO2 þ y4 NO;
The syngas composition, gasification temperature, tar and char (R6)
yields and the concentrations of the contamination gases predicted
by the model shows very good agreement with experimentally the stoichiometric balance of which consists of elemental mass
obtained values and investigations and other recent comparable balances for each of the following species C, H, O, N, S:
gasification models, suggesting that the current model can be
reliably used to perform engineering simulations of downdraft C : y1 ¼ 1; (A.1)
gasification systems and undertake process design, evaluation and
optimisation of gasification technology. a
H : y2 ¼ ; (A.2)
From the results obtained, the following conclusions are drawn: 2

1. The concentrations of H2 and CO2 increase with MC, while the O : 2y1 þ y2 þ 2y3 þ y4 ¼ b þ 2mcomb ; (A.3)
CO concentration decreases. N2 concentration remains constant
while CH4 concentration gradually increases with increasing N : y4 ¼ l; (A.4)
MC.
2. T increases with ER and decreases with MC for gasification
S : y3 ¼ d: (A.5)
reasons mentioned in 3.1.2.
3. The tar yield decreases with increasing ER and MC due to a By substituting Eqns. (A.1), (A.2), (A.4) and (A.5) into Eqn. (A.3),
better quality syngas. the following expression for mcomb is obtained:
4. The boudouard reaction is a reliable approach for predicting

char yield. a b l
mcomb ¼ 1þ  þ þd : (A.6)
5. The LHV and CGE decrease with an increase in ER and T. 4 2 2
6. The concentration of NH3 decreases with increasing ER values
Next, using the definition of the ER:
and increases with increasing MC, while the concentration of
H2S remains almost constant, decreasing slight with increasing mgas mgas
ER and MC. ER ¼ ¼ ; (A.7)
mcomb 1 þ a4  2b þ 2l þ d
7. It is shown that equilibrium models are able to provide reliable
predictions of syngas composition.
and accordingly:


a b l
mgas ¼ 1 þ  þ þ d  ER: (A.8)
Funding 4 2 2
Thus the combustion reaction simplifies as follows:
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.

a b l a
CHa Ob Nl Sd þ 1þ  þ þ d O2 /CO2 þ H2 Oþ dSO2 þ lNO:
4 2 2 2
Data availability (R7)
The biomass formation enthalpy is computed as a function of
Datasets related to this article can be found at https://data.
the molar LHV (J/mol) of the biomass [5,37,65], according to:
mendeley.com/datasets/jtwtrbhfcb/3, hosted at Mendeley Data
[47].

X4

hf ;bm ¼ LHVbm þ i¼1
yi hf ;i ; (A.9)

CRediT authorship contribution statement P

where 4i¼1 is repeated over all products of complete combustion
reaction (R6). The molar LHV can be calculated from the specific
A. Ibrahim: Conceptualization, Methodology, Software, Valida- HHV (J/kg) of the biomass on a d.b. [5,37,65] as:
tion, Formal analysis, Investigation, Data curation, Writing e orig-
a
inal draft, Visualization. S. Veremieiev: Conceptualization,
LHVbm ¼ HHVbmspec  Mbm  hvap ; (A.10)
Methodology, Writing e review & editing. P.H. Gaskell: Concep- 2
tualization, Methodology, Writing e review & editing.
where a2 is the number of moles of water produced per mole of
biomass as per the combustion reaction (R7) and hvap ¼ 44, 000J/
Declaration of competing interest mol is the enthalpy of vaporisation of water at standard tempera-
ture. The specific HHVbmspec (MJ/kg) is calculated via an empirical
The authors declare that they have no known competing correlation proposed by Channiwala and Parikh [67]:where yASH is
financial interests or personal relationships that could have the percentage weight of ash in biomass on a d.b. as reported in the
appeared to influence the work reported in this paper. proximate analysis of the feedstocks.
923
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

HHVbmspec ¼ 0:3491yC þ 1:1783yH þ 0:1005yS  0:1034yO  0:0151yN  0:0211yASH ; (A.11)

References fluidized bed-effect of the type of feedstock, in: Pyrolysis and Gasification,
Elsevier Applied Science London, 1989, pp. 618e623.
[29] E. Azzone, M. Morini, M. Pinelli, Development of an equilibrium model for the
[1] S. Sansaniwal, K. Pal, M. Rosen, S. Tyagi, Recent advances in the development
simulation of thermochemical gasification and application to agricultural
of biomass gasification technology: a comprehensive review, Renew. Sustain.
residues, Renew. Energy 46 (2012) 248e254.
Energy Rev. 72 (2017) 363e384.
[30] Y.-i. Lim, U.-D. Lee, Quasi-equilibrium thermodynamic model with empirical
[2] U. Kumar, M.C. Paul, Cfd modelling of biomass gasification with a volatile
equations for airesteam biomass gasification in fluidized-beds, Fuel Process.
break-up approach, Chem. Eng. Sci. 195 (2019) 413e422.
Technol. 128 (2014) 199e210.
[3] M. Puig-Arnavat, J.C. Bruno, A. Coronas, Review and analysis of biomass
[31] S. Rupesh, C. Muraleedharan, P. Arun, A comparative study on gaseous fuel
gasification models, Renew. Sustain. Energy Rev. 14 (9) (2010) 2841e2851.
generation capability of biomass materials by thermo-chemical gasification
[4] A.M. Salem, M.C. Paul, An integrated kinetic model for downdraft gasifier
using stoichiometric quasi-steady-state model, Int. J.Energy.Environ. Eng. 6
based on a novel approach that optimises the reduction zone of gasifier,
(4) (2015) 375e384.
Biomass Bioenergy 109 (2018) 172e181.
[32] T. Yamazaki, H. Kozu, S. Yamagata, N. Murao, S. Ohta, S. Shiya, T. Ohba, Effect
[5] M. La Villetta, M. Costa, N. Massarotti, Modelling approaches to biomass
of superficial velocity on tar from downdraft gasification of biomass, Energy
gasification: a review with emphasis on the stoichiometric method, Renew.
Fuels 19 (3) (2005) 1186e1191.
Sustain. Energy Rev. 74 (2017) 71e88.
ry, M. Grateau, M. Campargue, S. Bennici, [33] A.V. Bridgwater, Review of fast pyrolysis of biomass and product upgrading,
[6] T. Dahou, F. Defoort, S. Thie
Biomass Bioenergy 38 (2012) 68e94.
M. Jeguirim, C. Dupont, The influence of char preparation and biomass type on
[34] S. Ferreira, E. Monteiro, P. Brito, C. Vilarinho, A holistic review on biomass
char steam gasification kinetics, Energies 11 (8) (2018) 2126.
gasification modified equilibrium models, Energies 12 (1) (2019) 160.
[7] T. Dahou, F. Defoort, B. Khiari, M. Labaki, C. Dupont, M. Jeguirim, Role of in-
[35] A.K. Sharma, Modeling and simulation of a downdraft biomass gasifier 1.
organics on the biomass char gasification reactivity: a review involving re-
model development and validation, Energy Convers. Manag. 52 (2) (2011)
action mechanisms and kinetics models, Renew. Sustain. Energy Rev. 135
1386e1396.
(2021) 110136.
[36] H.P. Robert, W.G. Don, O.M. James, Perry's Chemical Engineers' Handbook,
[8] T. Dahou, F. Defoort, M. Jeguirim, C. Dupont, Towards understanding the role
Nova Iorque, 1984.
of k during biomass steam gasification, Fuel 282 (2020) 118806.
[37] R.T. Balmer, Modern Engineering Thermodynamics-Textbook with
[9] A. Tamosiu  nas, A. Chouche ne, P. Valatkevi
cius, D. Gim _ M. Aikas,
zauskaite,
Tables Booklet, Academic Press, 2011.
R. Uscila, M. Ghorbel, M. Jeguirim, The potential of thermal plasma gasification
[38] A. Gambarotta, M. Morini, A. Zubani, A non-stoichiometric equilibrium model
of olive pomace charcoal, Energies 10 (5) (2017) 710.
for the simulation of the biomass gasification process, Appl. Energy 227
[10] B. Khiari, M. Jeguirim, L. Limousy, S. Bennici, Biomass derived chars for energy
(2018) 119e127.
applications, Renew. Sustain. Energy Rev. 108 (2019) 253e273.
[39] P. Basu, Biomass Gasification and Pyrolysis: Practical Design and Theory, Ac-
[11] Z. Zainal, R. Ali, C. Lean, K. Seetharamu, Prediction of performance of a
ademic press, 2010.
downdraft gasifier using equilibrium modeling for different biomass mate-
[40] F. Guo, Y. Dong, L. Dong, C. Guo, Effect of design and operating parameters on
rials, Energy Convers. Manag. 42 (12) (2001) 1499e1515.
the gasification process of biomass in a downdraft fixed bed: an experimental
[12] A. Mountouris, E. Voutsas, D. Tassios, Solid waste plasma gasification: equi-
study, Int. J. Hydrogen Energy 39 (11) (2014) 5625e5633.
librium model development and exergy analysis, Energy Convers. Manag. 47
[41] N. Striu gas, K. Zakarauskas, A. Dziugys, R. Navakas, R. Paulauskas, An evalu-
(13e14) (2006) 1723e1737.
ation of performance of automatically operated multi-fuel downdraft gasifier
[13] M.J. Prins, K.J. Ptasinski, F.J. Janssen, From coal to biomass gasification: com-
for energy production, Appl. Therm. Eng. 73 (1) (2014) 1151e1159.
parison of thermodynamic efficiency, Energy 32 (7) (2007) 1248e1259.
rez, H. Laget, A. Horillo, Thermochemical equilibrium [42] M. Dogru, C. Howarth, G. Akay, B. Keskinler, A. Malik, Gasification of hazelnut
[14] A. Melgar, J.F. Pe
shells in a downdraft gasifier, Energy 27 (5) (2002) 415e427.
modelling of a gasifying process, Energy Convers. Manag. 48 (1) (2007)
[43] M. Dogru, A. Midilli, C.R. Howarth, Gasification of sewage sludge using a
59e67.
throated downdraft gasifier and uncertainty analysis, Fuel Process. Technol.
[15] A.K. Sharma, Equilibrium modeling of global reduction reactions for a
75 (1) (2002) 55e82.
downdraft (biomass) gasifier, Energy Convers. Manag. 49 (4) (2008) 832e842.
[44] P. Lv, Z. Yuan, L. Ma, C. Wu, Y. Chen, J. Zhu, Hydrogen-rich gas production from
[16] M. Puig-Arnavat, J.C. Bruno, A. Coronas, Modified thermodynamic equilibrium
biomass air and oxygen/steam gasification in a downdraft gasifier, Renew.
model for biomass gasification: a study of the influence of operating condi-
Energy 32 (13) (2007) 2173e2185.
tions, Energy Fuels 26 (2) (2012) 1385e1394.
[45] R. Wei, H. Li, Y. Chen, Y. Hu, H. Long, J. Li, C.C. Xu, Environmental Issues
[17] S. Jarungthammachote, A. Dutta, Thermodynamic equilibrium model and
Related to Bioenergy, Comprehensive Renewable Energy, 2020.
second law analysis of a downdraft waste gasifier, Energy 32 (9) (2007) 
[46] V. Kirsanovs, A. Zandeckis, C. Rochas, et al., Biomass gasification thermody-
1660e1669.
namic model including tar and char, Agron. Res. 14 (2016) 1321e1331.
[18] H.-J. Huang, S. Ramaswamy, Modeling biomass gasification using thermody-
[47] A. Ibrahim, S. Veremieiev, P. Gaskell. Equilibrium Model Solver (ems) [online]
namic equilibrium approach, Appl. Biochem. Biotechnol. 154 (1e3) (2009)
(2022). Mendeley Data Version 3, https://doi.org/10.17632/jtwtrbhfcb.3.
14e25.
[48] V. Maxim, C.-C. Cormos, A.-M. Cormos, S. Agachi, Mathematical modeling and
[19] R. Karamarkovic, V. Karamarkovic, Energy and exergy analysis of biomass
simulation of gasification processes with carbon capture and storage (ccs) for
gasification at different temperatures, Energy 35 (2) (2010) 537e549.
energy vectors poly-generation, in: Computer Aided Chemical Engineering,
[20] V.B. Silva, A. Rouboa, Using a two-stage equilibrium model to simulate oxygen
vol. 28, 2010, pp. 697e702. Elsevier.
air enriched gasification of pine biomass residues, Fuel Process. Technol. 109
[49] T. Jayah, L. Aye, R.J. Fuller, D. Stewart, Computer simulation of a downdraft
(2013) 111e117.
wood gasifier for tea drying, Biomass Bioenergy 25 (4) (2003) 459e469.
[21] N.S. Barman, S. Ghosh, S. De, Gasification of biomass in a fixed bed downdraft
[50] M. Barrio, J. Hustad, M. Fossum, A small-scale stratified downdraft gasifier
gasifierea realistic model including tar, Bioresour. Technol. 107 (2012)
coupled to a gas engine for combined heat and power production, Progress in
505e511.
thermochemical biomass conversion (2001) 426e440.
[22] A.Z. Mendiburu, J.A. Carvalho Jr., C.J. Coronado, Thermochemical equilibrium
[51] S.J. Yoon, Y.-I. Son, Y.-K. Kim, J.-G. Lee, Gasification and power generation
modeling of biomass downdraft gasifier: stoichiometric models, Energy 66
characteristics of rice husk and rice husk pellet using a downdraft fixed-bed
(2014) 189e201.
gasifier, Renew. Energy 42 (2012) 163e167.
[23] A. Gagliano, F. Nocera, F. Patania, M. Bruno, D.G. Castaldo, A robust numerical
[52] P. Dutta, V. Pandey, A. Das, S. Sen, D. Baruah, Down draft gasification
model for characterizing the syngas composition in a downdraft gasification
modelling and experimentation of some indigenous biomass for thermal
process, Compt. Rendus Chem. 19 (4) (2016) 441e449.
applications, Energy Proc. 54 (2014) 21e34, 0.
[24] E.S. Aydin, O. Yucel, H. Sadikoglu, Development of a semi-empirical equilib-
[53] F. Patuzzi, D. Basso, S. Vakalis, D. Antolini, S. Piazzi, V. Benedetti, E. Cordioli,
rium model for downdraft gasification systems, Energy 130 (2017) 86e98.
M. Baratieri, State-of-the-art of small-scale biomass gasification systems: an
[25] M. Costa, N. Massarotti, G. Cappuccio, C. Chang, A. Shiue, C. Lin, Y. Wang,
extensive and unique monitoring review, Energy 223 (2021) 120039.
Modeling of syngas production from biomass energy resources available in
[54] V. Kirsanovs, D. Blumberga, I. Veidenbergs, C. Rochas, E. Vigants, G. Vigants,
taiwan, Chem. Eng. 37 (2014).
Experimental investigation of downdraft gasifier at various conditions, Energy
[26] M. Costa, M. La Villetta, N. Massarottia, Optimal tuning of a thermo-chemical
Proc. 128 (2017) 332e338.
equilibrium model for downdraft biomass gasifiers, Chem. Eng. 43 (2015)
[55] D.S. Upadhyay, A.K. Sakhiya, K. Panchal, A.H. Patel, R.N. Patel, Effect of
439e444.
equivalence ratio on the performance of the downdraft gasifierean experi-
[27] A. Abuadala, I. Dincer, G. Naterer, Exergy analysis of hydrogen production
mental and modelling approach, Energy 168 (2019) 833e846.
from biomass gasification, Int. J. Hydrogen Energy 35 (10) (2010) 4981e4990.
[56] X. Li, J. Grace, C. Lim, A. Watkinson, H. Chen, J. Kim, Biomass gasification in a
[28] J. Corella, J. Herguido, J. Gonzalez-Saiz, Steam gasification of biomass in
circulating fluidized bed, Biomass Bioenergy 26 (2) (2004) 171e193.

924
A. Ibrahim, S. Veremieiev and P.H. Gaskell Renewable Energy 194 (2022) 912e925

[57] L. Damiani, A. Trucco, Biomass gasification modelling: an equilibrium model, G: Total Gibbs free energy, (J/mol)
modified to reproduce the operation of actual reactors, in: ASME Turbo Expo H: Total enthalpy, (J/mol)
2009: Power for Land, Sea, and Air, American Society of Mechanical Engineers M: Molecular mass of species, (kg/mol)
Digital Collection, 2009, pp. 493e502. Ntot: Number of moles of raw gas at temperature
[58] M. Kumar, B. Paul, D.S. Yadav, Effect of moisture content and equivalence ratio T: per 1 mol of biomass, (mol)
on the gasification process for different biomass fuel, Int. J. Mech. Eng. Ngas: Number of moles of dry gas at temperature
Technol. 7 (6) (2016). T0: per 1 mol of biomass, (mol)
[59] R.F. Naryanto, H. Enomoto, A. Vo Cong, K. Fukadu, Z. Zong, M.K. Delimayanti, R: Universal gas constant, (J/molK)
C. Chunti, R. Noda, The effect of moisture content on the tar characteristic of T: Gasification temperature, (K)
wood pellet feedstock in a downdraft gasifier, Appl. Sci. 10 (8) (2020) 2760. T0: Standard temperature, (K)
[60] E. Shayan, V. Zare, I. Mirzaee, Hydrogen production from biomass gasification; Vm: Molar volume of ideal gas at temperature
a theoretical comparison of using different gasification agents, Energy Conv- T0: and pressure p0, (m3/mol)
ers. Manag. 159 (2018) 30e41.
[61] Z. Yao, S. You, T. Ge, C.-H. Wang, Biomass gasification for syngas and biochar Lower case letters
co-production: energy application and economic evaluation, Appl. Energy 209
(2018) 43e55.
[62] M.-H. Cho, T.-Y. Mun, J.-S. Kim, Air gasification of mixed plastic wastes using g: Gibbs free energy, (J/mol)

calcined dolomite and activated carbon in a two-stage gasifier to reduce tar, gf : Gibbs free energy of formation (J/mol)
Energy 53 (2013) 299e305. h: Enthalpy, (J/mol)

[63] R. Xiao, B. Jin, H. Zhou, Z. Zhong, M. Zhang, Air gasification of polypropylene hf : Formation enthalpy of species (J/mol)
plastic waste in fluidized bed gasifier, Energy Convers. Manag. 48 (3) (2007) k: Equilibrium constant
778e786. m: Number of moles of air per 1 mol of biomass (mol)
[64] S. Salaudeen, P. Arku, A. Dutta, Gasification of plastic solid waste and p:: Partial pressure (Pa)
competitive technologies, in: Plastics to Energy, Elsevier, 2019, pp. 269e293. p0: Standard pressure (Pa)
[65] M.L. de Souza-Santos, Solid Fuels Combustion and Gasification: Modeling, s: Entropy, (J/mol)

Simulation, CRC Press, 2010. sf : Entropy of formation (J/molK)

[66] J. Ancheyta, Modeling of Processes and Reactors for Upgrading of Heavy Pe- w: Number of moles of water per 1 mol of biomass (mol)
troleum, CRC Press, 2013. x: Number of moles, (mol)
[67] S. Channiwala, P. Parikh, A unified correlation for estimating hhv of solid,
liquid and gaseous fuels, Fuel 81 (8) (2002) 1051e1063. Greek letters

Nomenclature a: Number of atoms of hydrogen in the biomass feedstock

b: Number of atoms of oxygen in the biomass feedstock
l: Number of atoms of nitrogen in the biomass feedstock
d: Number of atoms of sulphur in the biomass feedstock
Upper case letters y: stoichiometric number of moles, (mol)

Cp:: Molar heat capacity, (J/molK)

925