CET Volume 86
DOI: 10.3303/CET2186142
Paper Received: 19 October 2020; Revised: 9 February 2021; Accepted: 23 April 2021
Please cite this article as: Santos Jr. J.M., De Souza G.F.B., Vidotti A.D.S., De Freitas A.C.D., Guirardello R., 2021, Optimization of Glycerol
Gasification Process in Supercritical Water Using Thermodynamic Approach, Chemical Engineering Transactions, 86, 847-852
DOI:10.3303/CET2186142
CHEMICAL ENGINEERING TRANSACTIONS
VOL. 86, 2021
A publication of
The Italian Association
of Chemical Engineering
Online at www.cetjournal.it
Guest Editors: Sauro Pierucci, Jiří Jaromír Klemeš
Copyright © 2021, AIDIC Servizi S.r.l.
ISBN 978-88-95608-84-6; ISSN 2283-9216
Optimization of Glycerol Gasification Process in Supercritical
Water using Thermodynamic Approach
Julles Mitoura dos Santos Juniora, Gustavo Furtado Bezerra de Sousab,
Annamaria Dória Souza Vidottib, Antonio Carlos Daltro de Freitasb, Reginaldo
Guirardelloa*
aSchool of Chemical Engineering, University of Campinas (UNICAMP), Av. Albert Einstein 500, 13083-852, Campinas-SP,
Brazil
bChemical Engineering Department, Exact Sciences and Technology Center, Federal University of Maranhão (UFMA), Av.
dos Portugueses, 1966, Bacanga, 65080-805, São Luís-MA
guira@feq.unicamp.br7
Glycerol is a byproduct of biodiesel production. An alternative to use of this byproduct is the gasification with
supercritical water (SCWG) for hydrogen generation. This work aims to analyze the conditions that enhance
the formation of hydrogen using the response surface methodology in combination with optimization
techniques. The results of the reaction in the equilibrium condition were obtained using Gibbs energy
minimization method for isothermic systems and entropy maximization method for adiabatic systems. The
proposed thermodynamic models were solved using GAMS 23.9.5 software in combination with CONOPT3
solver. As a result of the simulations, the final compositions of the gaseous phase and the thermal behavior for
the operational conditions of the reaction are presented. The reaction was characterized by the formation of
hydrogen, evaluating the temperature between 586.64 and 1259 K, pressures in the range of 216.36 to 283.64
bar and the glycerol / water molar ratio varying from 0.032 to 0.368 in the feeding according to the planning
experimental. Higher hydrogen formation was observed for isothermic reaction conditions, indicating that
glycerol SCWG was an endothermic reaction, this fact is justified by the results of adiabatic reactors.
Hydrogen formation is mainly influenced by the effects of temperature and glycerol composition, reaching the
maximum hydrogen formation (2.29 moles) operating in an isothermal manner for high temperatures (1123 K)
and high glycerol/water ratios in the feed (0.368). For both results, the pressure pressure on the amount of
internal hydrogen was not statistically significant at 95%.
1. Introduction
The conversion of biomass into energy, fuels or chemicals with high added value can be accomplished in a
number of ways. Biochemical and thermochemical conversion routes are possible ways to obtain biofuels
through the conversion of biomass. The thermochemical routes, such as gasification, pyrolysis and
combustion, deal with the conversion of biomass into gases, either directly by burning or using chemicals as
intermediates, these are more favorable because they present better efficiency in destroying organic
compounds in less reaction time (Lachos-Perez et al. 2015). An alternative route for converting biomass into
clean combustible gases is the gasification reaction with supercritical water (SCWG), this process has high
levels of hydrogen formation with a high degree of purity and does not require the pre-drying treatment that
would be necessary for the conventional gasification process (Freitas and Guirardello 2014).
The interest in using supercritical water as a reaction medium is in its transport and solubilization properties
(Calzavara et al. 2005). Supercritical water has the viscosity of the gas phase and density of the liquid phase,
approximately. The increase in temperature provides a decrease in its density, which causes the reduction of
its dielectric constant, which causes water to behave similarly to a non-polar solvent, making it a good solvent
for organic compounds (Houcinat et al. 2018).
847
The gasification reaction with supercritical water has the advantage of a high rate of hydrogen formation, an
effect resulting from the possibility of operating with biomass sources with a high moisture content. This trend
ocourss because the addition of water that favors the displacement of the water gas shift reaction, thus
generating more hydrogen.
Glycerol is a substrate of great potential for application in SCWG, since it is the by-product generated in
greater quantity in the production of biodiesel, making up approximately 10% of the volume of biodiesel
formed. The amount of this substance that forms as industrial waste is hardly fully utilized in traditional ways
such as the production of cosmetics and animal feed (Castello and Fiori 2011). Through gasification, glycerol
can be converted into hydrogen and other by-products.
This work uses the response surface methodology to verify the results of the glycerol SCWG simulations. The
equilibrium compositions were obtained with the aid of the GAMS 23.9.5 software, using the Gibbs
thermodynamic energy minimization and entropy maximization methods. This study aims to verify the
operational conditions that maximize the formation of hydrogen throughout the glycerol SCWG.
2. Methodology
2.1. Thermodynamic models
Under conditions of constant pressure (P) and temperature (T), the condition of thermodynamic equilibrium
can be formulated as a problem of minimizing Gibbs energy. The total Gibbs minimization problem of the
system can be described according to Equation 1.
1 1
min
NC NF
k k
i i
i k
G n μ
= =
= (1)
The system in the condition of Gibbs minimum energy must satisfy the condition of non-negativity of the
number of moles (Eq.2) and the conservation of the number of atoms (Eq.3).
0kin ≥ (2)
1 1 1
, 1,...,
NC NF NC
k o
mi i mi i
i k i
a n a n m NE
= = =
= =
(3)
Under constant pressure (P) and enthalpy (H) conditions, equilibrium can be determined by the maximum
entropy (Rossi et al. 2011). An entropy maximization problem can be written according to Equation 4. The
formulation of equilibrium as an entropy maximization problem is interesting for determining the system
equilibrium temperature (Teq) mainly in exothermic reactions (Freitas and Guirardello 2012).
1 1
max
NC NF kk
ii
i k
S n S
= =
= (4)
In addition to the restrictions imposed on the system in the condition of minimum Gibbs energy, the
maintenance of the enthalpy (Eq.5) of the system must be added as a restriction.
1 1 1
NC NF NCk ok o o
i ii i
i k i
n H n H H
= = =
= = (5)
The Gibbs energy minimization methodology provides as a result the equilibrium compositions of a reaction
system in a reactor operating in an isothermal way and the methodology of maximizing entropy represents the
operation of an adiabatic reactor.
To obtain the fugacity coefficients of the system components, the equations of the virial truncated in the
second term were used. The calculation of the second virial coefficient will be made using the correlation of
Pitzer and Curl (1957) , modified by Meng et al. (2004) according to Equation 6.
848
(6) ln ∅ = −
The methodology used in this work, using the virial equation to represent the non-ideality of systems of
interest, has been used by our research group to verify SCWG from several sources of biomass and previous
works from our research group present the validation of this methodology comparing the predictive capacity of
the virial equation with experimental data. Freitas and Guirardello (2014) presentes a comparison between
experimental and predicted data using the virial equation for glycerol SCWG showing good agreement.
The thermodynamic models covered in this work will be solved by the GRG (Generalized Reduced Gradient)
search method, using the CONOPT 3 solver, in the GAMS software.
Table 1 shows the thermodynamic properties of the chemical components involved and considered as
possible to form in the SCWG reaction of glycerol, over the simulations performed in the GAMS software.
Table 1: Thermodynamic properties of the chemical components used in the simulations
Component
3
c
m
V
kmol
( )cP bar ( )cT K ω
H2O 0.056 220.600 647.300 0.344
C3H8O3 0.264 75.000 850.000 0.513
O2 0.073 50.400 154.600 0.222
H2 0.064 13.000 33.000 0.000
N2 0.089 34.000 126.200 0.038
CH2O2 0.125 58.100 588.000 0.316
CH3COOH 0.171 57.900 594.500 0.445
CH4 0.099 45.800 191.100 0.011
CH3OH 0.118 81.000 512.600 0.565
C2H6 0.146 48.700 305.300 0.099
C2H6O 0.167 64.500 513.900 0.649
C3H8 0.200 42.500 369.800 0.152
C4H10 0.255 38.000 425.100 0.200
CO 0.058 64.800 180.000 0.582
CO2 0.082 101.500 431.002 0.851
Font: Poling et al. (2001)
The SCWG reaction of different sources of biomass produces various components such as methanol,
methane, ethanol, ethane and others besides hydrogen. It was chosen to analyze the formation of hydrogen in
view of the fact that it presents a major part in the formation of the synthesis gas (syngas), in addition to being
a product of high interest for application in the Fischer-Tropsch synthesis reaction to obtain fuels or for use
direct into fuel cells.
2.2. Statistical analysis
The statistical treatment of the results presented in this work was done with the aid of the TIBCO®
STATISTICA™ software, using the response surface methodology applied to the results of the simulations
made in GAMS for the glycerol SCWG. The reaction was characterized for the formation of hydrogen by
evaluating temperatures between 586 and 1259 K, pressures in the range of 216.36 to 283.64 bar and the
glycerol/water molar ratio varying from 0.032 to 0.368 in the feeding (ranges determined by the planning
carried out in the STATISTICA software).
3. Results and discussion
The results of Table 2 were obtained through simulations made in the GAMS software using methods of Gibbs
energy minimization and entropy maximization for hydrogen formation during the reaction and the system
equilibrium temperature (Teq), for the conditions established in the planning matrix provided by the
STATISTICA software. For both methodologies, the results presented in Table 2 indicate that the formation of
hydrogen is favored with the increase in temperature. The yield of hydrogen production increases with
increasing temperature, mainly above 873 K, similar results have been reported by Withag et al. (2012).
849
It is possible to conclude that lower rates of hydrogen formation are observed with the increase in system
pressure, however, this difference has an insignificant value, at a 95% confidence level (F test). Similar results
for temperature and pressure effects are reported in the literature for the glucose and cellulose SCWG
reaction (Freitas and Guirardello 2012). Figure 1 presents the results for the effects of temperature and
glycerol feeding under the formation of hydrogen throughout the reaction for Gibbs energy minimization and
entropy maximization method by fixing the pressure at 250 bar.
Table 2: Experimental design and results of simulations of hydrogen formation and equilibrium temperature of
the system fixing the water supply to 5 moles and varying the conditions of temperature, pressure and glycerol
composition in the system feed
minG maxS
Temperature (K) Pressure (bar) Glycerol (moles) H2 (moles) H2 (moles) Teq (K)
723.000 230.000 0.500 8.62E-02 6.13E-02 687.340
723.000 230.000 1.500 1.19E-01 6.39E-02 661.290
723.000 270.000 0.500 7.48E-02 5.68E-02 687.680
723.000 270.000 1.500 1.05E-02 5.92E-02 661.500
1123.000 230.000 0.500 1.48E+00 3.30E-01 972.730
1123.000 230.000 1.500 2.29E+00 2.61E-01 836.980
1123.000 270.000 0.500 1.38E+00 3.07E-01 973.680
1123.000 270.000 1.500 2.11E+00 2.42E-01 837.630
586.641 250.000 1.000 1.25E-02 2.73E-02 607.930
1259.359 250.000 1.000 3.90E+00 3.90E-01 987.450
923.000 216.364 1.000 6.57E-01 1.58E-01 780.070
923.000 283.636 1.000 5.53E-01 1.39E-01 780.980
923.000 250.000 0.159 3.83E-01 1.85E-01 876.040
923.000 250.000 1.841 7.48E-01 1.31E-01 739.630
923.000 250.000 1.000 6.00E-01 1.48E-01 780.570
923.000 250.000 1.000 6.00E-01 1.48E-01 780.570
The results presented in Figure 1 indicate greater efficiency of the Gibbs energy minimization methodology
when compared to the entropy maximization for hydrogen formation. Under conditions of temperature equal to
923K, pressure equal to 250 bar and molar ratio of glycerol/water equal to 0.2 in the feed, the values of gas
formation found were 0.60032 and 0.14783 moles of H2, respectively for each method. This result indicates a
better operationalization of the reactors in isothermal mode in comparison to adiabatic reactors, being still a
proof of the endothermic behavior of this reaction.
a) b)
Figure 1: Hydrogen formation along the glycerol SCWG using Gibbs energy minimization methodologies (a)
and entropy maximization (b)
850
Figure 2 shows the system equilibrium temperature as a function of the biomass concentration to be fixed at
250 bar. The results indicate that a glycerol SCWG presents a slightly endothermic behavior for both
conditions analyzed. For this reason, a glycerol SCWG presents better rates of H2 formation operating in
isothermal reactors, as the temperature support offered by the system compensates for the endothermic effect
of the reaction. This behavior is an indication that the reactions that result in the formation of hydrogen during
a glycerol SCWG are mostly endothermic, probably associated with the glycerol decomposition routes.
a) b)
c)
Figure 2: Equilibrium temperature as a function of the biomass concentration (a: 723 K; b: 923 K; c: 1123 K)
4. Conclusions
The Gibbs energy minimization and entropy maximization methods proved to be efficient for solving the
equilibrium calculations during SCWG of glycerol. The thermodynamic models solved in the GAMS 23.9.5
software with the aid of the CONOPT solver are presented as a quick and effective way to solve the proposed
thermodynamic problems, with computational time less than 1 second in all cases prevented by this work.
The temperature and composition of glycerol in the feed have a major influence on the formation of hydrogen
throughout the reaction. Conversely, pressure is not significant under this result, at a level of 95% statistical
confidence. The results indicate that the highest hydrogen formation rates are obtained during a glycerol
SCWG when the system is conditioned at constant temperatures, which means, when an equilibrium condition
is solved as a Gibbs energy minimization problem. This result is justified by the endothermic behavior of the
glycerol SCWG when the isothermal reactor provides the thermal support favoring the formation of products
throughout the reaction, compensating for its endothermic effect.
851
Under similar conditions of pressure, temperature and glycerol composition in the reactor supply, the results of
hydrogen formation obtained by the Gibbs energy minimization methodology are on average 63.82% higher
than the results obtained by the entropy maximization methodology for the formation of this product, indicating
better operationalization of the reactors in isothermal mode compared to adiabatic reactors for an SCWG
glycerol.
In an isothermal reactor, higher rates of hydrogen (2.29 moles) are formed during the SCWG reaction of
glycerol when the system is conditioned to the high temperatures (1123K) and high glycerol/water ratios in the
feed (0.368). Thus, the glycerol SCWG reaction proved to be an effective route for the production of hydrogen.
Acknowledgments
The authors gratefully acknowledge the financial support from CNPq – Conselho Nacional de
Desenvolvimento Científico e Tecnológico, Brazil (Processes: 402882/2016-4 and 830535/1999-3).
References
Calzavara, Y., Joussot-Dubien, C., Boissonnet, G., Sarrade, S. J. E. c. and management (2005) 'Evaluation of
biomass gasification in supercritical water process for hydrogen production', 46(4), 615-631.
Castello, D. and Fiori, L. J. B. t. (2011) 'Supercritical water gasification of biomass: Thermodynamic
constraints', 102(16), 7574-7582.
Freitas, A. and Guirardello, R. J. C. E. T. (2012) 'Supercritical water gasification of glucose and cellulose for
hydrogen and syngas production', 27, 361-366.
Freitas, A. C. and Guirardello, R. J. I. j. o. h. e. (2014) 'Comparison of several glycerol reforming methods for
hydrogen and syngas production using Gibbs energy minimization', 39(31), 17969-17984.
Houcinat, I., Outili, N. and Meniai, A. H. J. B. (2018) 'Optimization of gas production and efficiency of
supercritical glycerol gasification using response surface methodology', 9(5), 625-633.
Lachos-Perez, D., Prado, J. M., Torres-Mayanga, P., Forster-Carneiro, T. and Meireles, M. A. A. J. F. P. H.
(2015) 'Supercritical water gasification of biomass for hydrogen production: Variable of the process', 5(3),
92-101.
Meng, L., Duan, Y.-Y. and Li, L. (2004) 'Correlations for second and third virial coefficients of pure fluids', Fluid
phase equilibria, 226, 109-120.
Pitzer, K. S. and Curl Jr, R. (1957) 'The Volumetric and Thermodynamic Properties of Fluids.: III. Empirical
Equation for the Second Virial Coefficient' in Molecular Structure And Statistical Thermodynamics:
Selected Papers of Kenneth S Pitzer, World Scientific, 311-312.
Poling, B. E., Prausnitz, J. M. and O'connell, J. P. (2001) The properties of gases and liquids, Mcgraw-hill New
York.
Rossi, C., Berezuk, M., Cardozo-Filho, L., Guirardello, R. J. C. and engineering, c. (2011) 'Simultaneous
calculation of chemical and phase equilibria using convexity analysis', 35(7), 1226-1237.
Withag, J. A., Smeets, J. R., Bramer, E. A. and Brem, G. J. T. J. o. s. f. (2012) 'System model for gasification
of biomass model compounds in supercritical water–a thermodynamic analysis', 61, 157-166.
852