i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Hydrogen production by a Fe-based oxygen carrier

and methane-steam redox process:
Thermodynamic analysis

pez-Ortiz, M.J. Mel
endez-Zaragoza, V. Collins-Martı́nez*

A. Lo
Centro de Investigaci

on en Materiales Avanzados, S.C., Miguel de Cervantes 120, Complejo Industrial Chihuahua,
Chihuahua, Chihuahua, Mexico

article info abstract

Article history: The performance of iron oxide FeO/Fe3O4 redox pair as an oxygen carrier was investigated
Received 7 June 2017 for hydrogen (H2) and/or syngas (H2þCO) production through a methane-steam chemical
Accepted 17 July 2017 looping process using thermodynamic analysis and process simulation by Aspen Plus to
Available online xxx find optimal reactions conditions (ORC). The reaction system was divided in two reactors.
In the fuel reactor (FR) the following reactions were targeted: ¼ CH4(g) þ Fe3O4 ¼ 3FeO þ ¼
Keywords: CO2(g) þ ½ H2O(g) and CH4(g) þ Fe3O4 ¼ 3FeO þ 2H2(g) þ CO(g) where the CH4/Fe3O4 molar
Methane-steam redox process ratio was varied from 0.25 to 1 and 300e900  C. In the steam oxidation reactor (SR), the solid
Oxygen carrier products from FR were combined with steam to produce: 3FeO þ H2O(g) ¼ Fe3O4 þ H2(g) in a
Thermodynamic analysis T range of 300e900  C. Results indicate that ORC in the reduction reactor consists in T ¼ 650
e700  C and CH4/Fe3O4 ¼ 0.75e1, While for the SR were T ¼ 520e600  C.
© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

scale, since small-scale plants make hydrogen production

Introduction
Hydrogen, apart from being an important industrial chemical
worldwide, can be considered as the choice of carrier for the
emerging renewable energy generation technologies. How-
ever, today hydrogen production relies on the use of fossil
fuels, being steam methane reforming (SMR) the preferred
route for the industrial production of hydrogen, since it ac-
counts for more than a half of the world hydrogen production.
SMR is a highly endothermic catalytic reaction
(DH 298K ¼ 206 kJ/mol), which takes place in a temperature
range of 700e850  C according to:
Reaction 1 : CH4ðgÞ þ H2 OðgÞ ¼ 3H2ðgÞ þ COðgÞ
The economic feasibility of SMR units are restricted by
uneconomical. Therefore, large centralized industrial plants
have been found to be the most cost-effective [1]. World
consumption of hydrogen is projected to increase 3.5%
annually through 2018 to more than 300 billion cubic meters
and is expected to grow to about 5e6% in the next five years
[2]. Hence, an alternate approach that may include the use of
near and remote natural gas reserves, especially those coming
from the recent shale gas sites being explored worldwide, is
needed for more cost-effective small scale H2 production.
Consequently, there is a renewed interest in the old and
mature steam-iron process [3]. This process, in principle, is
one of the few processes that do not require the step of gas
separation to produce a high purity hydrogen stream at small
scale. This process involves a series of the redox reactions,
where the reduction of iron oxide with syngas (a mixture of CO

* Corresponding author.
E-mail address: virginia.collins@cimav.edu.mx (V. Collins-Martı́nez).
http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
0360-3199/© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
2 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3

and H2) is followed by the oxidation of the reduced metallic structural stabilization of ceria-zirconia solid solutions sup-
iron with steam according to the following reactions: ported on alumina towards the CO oxidation catalysis. The
unique oxygen storage capacity of the CeO2eZrO2 combined
Reaction 2 : Fe3 O4 þ 4COðgÞ ¼ 3Fe þ 4CO2ðgÞ
with iron oxides produced an oxygen carrier that combines
this important feature with the thermal stability and activity
Reaction 3 : Fe3 O4 þ 4H2ðgÞ ¼ 3Fe þ 4H2 OðgÞ as confirmed by several research studies [12e14].
Moreover, several researchers [15e19] have proposed a
Reaction 4 : 3Fe þ 4H2 OðgÞ ¼ Fe3 O4 þ 4H2ðgÞ novel two-step SMR process for hydrogen production based on
This process make use of a metal oxide (Fe3O4) as an oxy- the chemical looping concept and a scheme of this is pre-
gen carrier to perform the redox reactions (2e4) and at the sented in Fig. 1.
same time serves as a heat transfer medium between the two In the fuel reactor (I) the oxygen carrier, typically a metal
reactors (reduction and oxidation) in a cycle reaction loop in oxide (MeO), is reacted with methane to produce syngas and
order to make a continuous process and to increase energy the reduced metal (Me) according to:
efficiency. Gaseous carbon oxides rich stream is achieved after Reaction 6 : CH4ðgÞ þ MeO ¼ Me þ 2H2ðgÞ þ COðgÞ
steam condensation from the reduction reactor (reactions 2
and 3) and a high purity hydrogen stream can be obtained
Reaction 7 : H2 OðgÞ þ Me ¼ MeO þ H2ðgÞ
after steam condensation from the oxidation reactor (reaction
4). Due to the significant characteristics of this process, this Finally, the regenerated oxygen carrier is sent back to the
has been used as a reference for the production of H2 in the fuel reactor to complete one full loop or reaction cycle. The
chemical looping combustion [4] and H2 storage [5]. combination of reactions (6) and (7) leads to reaction (1), which
For the specific case of the reduction of iron oxide with a is the SMR reaction. The reaction process indicated in Fig. 1 is
gas fuel such as methane and the redox iron pair being Fe3O4/ usually called as chemical looping steam methane reforming
FeO the following reaction applies: (CL-SMR). This process separate the feed of methane and
steam in two step reactors thus, producing a syngas stream in
1 1 1
Reaction 5 : Fe3 O4 þ CH4ðgÞ ¼ 3FeO þ CO2ðgÞ þ H2 O the first and a H2 gas product in the second that needs no
4 4 4
further purification.
Even though several other metal oxides have been pro- One critical issue of this proposed reaction scheme deals
posed under the chemical looping concept for hydrogen pro- with the relatively low reactivity of the oxygen carrier towards
duction, the Fe3O4/Fe redox system is one of the most studied the reoxidation of the reduced metal (reaction 7). Several
systems due to the fact that employs inexpensive materials metal oxides have been reported as oxygen carriers for the CL-
and because it produces a high yield of H2 per mass of iron SMR such as Fe3O4, WO3, SnO2, NieFe, (Zn, Mn)eFe, CueFe and
(48 g H2/kg Fe, from equation 4). One of the main challenges of Ce-based oxides [15]. Among all these oxygen carriers such as
a looping operation of the steam-iron process lies in the Fe-oxides have proven to be suitable for the CL-SMR.
temperature swings and different gaseous atmospheres that
the material needs to be exposed in order to perform a cycle
operation in the process. This operation mode, performed at
high temperatures, often produce sintering in the material
with the consequent loss of activity and this being reflected in
a significant decrease of lifetime, thus reducing the overall
efficiency of the process.
In order to overcome this issue, several studies have
concentrated in improving reactor kinetics and the stability of
the materials by modifying iron oxide with the addition of
metals like Al, Mo and Ce as reported by Otsuka et al. [6].
Takenaka et al. [7] employed a mixture of NieCreFeOx, which
reacted with methane followed by reaction with steam to
produce pure hydrogen, where Cr species in NieCreFeOx
were stabilized as octahedral Crþ 3 O6 in the ferrites CrxFe3xO4
during the redox reactions. It was concluded that Cr helped to
inhibit sintering of iron species. While Urasaki et al. [8] added
palladium and zirconia to the iron-based oxygen carrier to
accelerate both the reduction and oxidation rates and also
found that zirconia enhanced structural and thermal stability
in iron oxides. Zafar et al. [9] used SiO2 and MgAl2O4 as sup-
ports to reduce the sintering of iron oxide, while Galvita and
Sundmacher [10] proposed the use of Cr2O3eFe3O4eCeO2e
ZrO2 to provide stability to the oxygen carrier up to 100 redox
cycles. However, some carbon monoxide in the hydrogen
product stream was observed during the re-oxidation phase,
whereas Liotta et al. [11] examined the redox behavior and Fig. 1 e Conceptual scheme of the SMR chemical looping.

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3
Nevertheless, thermodynamic constrains for the hydrogen
evolution make this oxygen carrier to behave in a complex
way, since in some cases the complete reduced Fe phase is
required for a high H2 production yield, while the complete
oxidation with steam beyond Fe3O4 is thermodynamically
limited [17]. However, this oxygen carrier can be considered
ideal for CL-SMR applications and this is based on its reduced
propensity to carbon formation and resistance to form ag-
glomerates. These last features are critical for an oxygen
carrier to retain its activity over continuous redox cycles.
Furthermore, iron oxide oxygen carriers are not susceptible to
form sulfide or sulfate compounds under sulfur-containing
atmospheres [20], while large natural reserves and low cost
make them suitable for implementation at large scales
[21e23]. Some drawbacks of these Fe-based oxygen carriers
include weak redox characteristics, relatively low oxygen
storage capacity (OSC), and limited reactivity towards gaseous
fuels like methane [24].
Additionally, Fe presents multiple oxidation states such as
Fe2O3, Fe3O4, FeO and Fe and as indicated above it presents
thermodynamic limitations, since only the transformation
from Fe2O3 to Fe3O4 is feasible for the total oxidation of
methane, whereas partial oxidation of methane (POX) is
favored in the transition reduction from FeO to Fe according to
Monzam et al. [25].
Thermodynamic considerations
Several thermodynamic analyses have studied different ap-
proaches to the CL-SMR process. Fraser et al. [25] performed a
thermodynamic analysis of the reformer sponge iron cycle
(RESC) process that consisted in the hydrogen production
including a hydrocarbon reformer and a sponge iron reactor
(SIR). The reformer output gas (H2 and CO) reduced the pellets
from magnetite (Fe3O4) and wuestite (FeO) into iron metal (Fe)
through reactions (2) and (3) and the reduced Fe is reoxidized
with steam by reaction (4). They found that the process offers
hydrocarbon to hydrogen conversion efficiencies in the order
of 75% vs LHV for methane and at a temperature range of
750  C and above. Steinfeld and Kuhn [26] presented a high
temperature thermochemistry analysis of the Fe3O4 reduction
with methane for the production of syngas. They studied the
reaction system at temperatures above 1000  C aimed for the
use of concentrated solar radiation as the energy source. They
found that a mixture of 66.7% H2 and 33.3% CO can be ach-
ieved with complete reduction to metallic Fe. Kang et al. [27]
performed a thermodynamic analysis of the three-reactor
chemical-looping process (TRCL), which consisted in a fuel
reactor, a steam reactor, and an air reactor. In the fuel reactor,
natural gas (mainly CH4) is oxidized to CO2 and H2O by the
lattice oxygen of the oxygen carrier (Fe2O3, WO3, and CeO2). In
the steam reactor, the steam is reduced to hydrogen through
oxidation of the reduced oxygen carrier, while in the air
reactor; the oxygen carrier is fully oxidized by air. Results
indicated a hydrogen production of 2.64 mol H2 per mol CH4
under thermo-neutral conditions and this was affected
mainly by the steam-conversion rate. Bohn et al. [28] studied
the hydrogen production with simultaneous capture of CO2
using the redox reactions of iron oxides in packed beds. They
examined the thermodynamics of exposing Fe2O3 to a syngas

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
3
to produce a high purity CO2 stream ready for sequestration,
while the oxidation of FeO to Fe3O4 with steam produced a
high purity hydrogen stream. They found that reduction to Fe,
rather than FeO, in step 1 gave low levels of H2 and after 10
cycles of reduction and oxidation led to the deposition of
carbon at lower temperatures. Furthermore, Svoboda et al.
[29] studied the thermodynamic and chemical equilibrium of
the reduction of Fe3O4 by CH4 and oxidation of iron by steam.
The study was concentrated on finding convenient conditions
for reduction of Fe3O4 to iron at temperatures from 400 to
800 K, but also on the possible formation of undesired soot,
Fe3C and iron carbonate. Reduction of magnetite with
methane did not produce iron carbides or carbonates,
whereas conditions favored the formation of Fe and CO2, and
H2O as gas products.
From all previous thermodynamic studies related to the
use of iron oxides as oxygen carrier for the complete oxidation
of methane (reactions 2 and 3) or the partial oxidation of
methane, which is a special case of the so-called chemical
looping steam methane reforming (CL-SMR) that involve the
partial oxidation of methane through:
Reaction 8 : 3Fe2 O3 þ CH4ðgÞ ¼ 2Fe3 O4 þ 2H2ðgÞ þ COðgÞ
Reaction 9 : Fe3 O4 þ CH4ðgÞ ¼ 3FeO þ 2H2ðgÞ þ COðgÞ
Reaction 10 : FeO þ CH4ðgÞ ¼ Fe þ 2H2ðgÞ þ COðgÞ
Reaction 11 : 3FeO þ H2 OðgÞ ¼ Fe3 O4 þ H2ðgÞ
These last reactions present the advantage to produce a
syngas mixture in the iron reduction reactor (reactions 8e10),
while the regeneration of the reduced Fe with steam is able to
produce a high purity H2 stream. However, this regeneration
will only produce Fe3O4 as reaction (11) is thermodynamically
limited. Therefore, in a cyclic operation the redox reaction
between the two reactors of Fig. 1 will only be reactions (9) and
(11). However, carbon deposition during the operation is
another important aspect to be considered in this process
through the Boudouard and methane pyrolysis reactions:
Reaction 12 : 2COðgÞ ¼ CO2ðgÞ þ C
Reaction 13 : CH4ðgÞ ¼ 2H2ðgÞ þ C
Galvita and Sundmacher [30] studied the CL-SMR reaction
system and concluded that at a reduction of approximately
60% of the Fe2O3 carbon formation was avoided. Further
reduction towards Fe produced carbon deposits on the oxygen
carrier. Therefore, the restrictions in order to avoid carbon
formation and the steam oxidation of the reduced iron lead to
select the Fe3O4/FeO as the ideal redox pair for the CL-SMR
process.
Up to date, no thermodynamic analysis have been per-
formed to explore optimal reactions conditions. From one
end: complete oxidation of methane by Fe3O4 (steam iron
process, SIP, which produce only CO2 and FeO in the reduction
reactor) to the other end: partial oxidation of methane (POX-
MeO or CL-SMR, where syngas and FeO is produced) and
subsequent hydrogen production from reoxidation of FeO.
Furthermore, most of the previously reported thermodynamic
4 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3
analyses have not considered the very important carbon for-
mation in their calculations, which avoidance will certainly
mislead the entire reaction system predicted equilibrium
compositions and reactions yields. This proposed process
operation aims to restrict the oxidation states between the
two reactors of Fig. 1 to the Fe3O4/FeO redox pair, while
avoiding carbon formation to insure a high purity hydrogen
stream from the steam oxidation reactor.
Therefore, the aim of the present investigation is to
perform a thermodynamic and process simulation studies to
find optimal operating conditions (avoiding carbon formation)
close to equilibrium for a high H2 and/or syngas production. In
addition, these under the SIP/CL-SMR reaction system using
methane as a feed (for the fuel reactor) and steam (for the
oxidation reactor), while employing the Fe3O4/FeO redox pair
between the reduction and oxidation reactors. This analysis
will include the evaluation of the thermodynamic equilibrium
composition of the reaction system, the behavior of the Fe3O4/
CH4 and H2O/FeO feed molar ratios as a function of tempera-
ture and product compositions for the reduction and oxida-
tion reactors, respectively. Furthermore, it is expected that the
feed molar ratio of Fe3O4 to CH4 will determine the operation
mode of the process from complete oxidation (SIP) to a partial
oxidation of methane (CL-SMR). Consequently, a process
simulation will examine oxygen carrier recirculation from the
oxidation to the reduction reactors and its influence on the
process operating conditions.
Methods
Thermodynamics method
Thermodynamic calculations in the present study were per-
formed by the Gibbs free energy minimization technique
GFEM. In a reaction system where many simultaneous re-
actions take place, equilibrium calculations can be performed
through GFEM approach (also called the nonstoichiometric
method). Details of this technique can be found elsewhere
[31,32]. All calculations were performed using the ASPEN Plus®
RGIBBS reactor model and the RedlicheKwong equation of
state for correction of non-ideal conditions. In a RGIBBS
reactor the equilibrium composition of all possible combina-
tion of reactions that are able to take place within the ther-
modynamic system are considered. The RGIBBS program of
this reactor finds the most stable phase combination and es-
tablishes the phase compositions where the Gibbs free energy
of the reaction system reaches its minimum at a fixed mass
balance, constant pressure, and temperature. For the oxida-
tion of methane and regeneration of the iron oxide species the
gaseous species included were: CH4, CO, CO2, H2, and H2O,
while solid species were: C, Fe, FeO, Fe2O3, F3C, and FeCO3.
For the fuel reactor the temperature was varied in the range
of 300e900  C at 1 atm. While, the CH4/Fe3O4 feed molar ratio
was varied from 0.25 (stoichiometric value according to reac-
tion (5)) to 1 (stoichiometric value according to reaction (9)).
Whereas, for the oxidation reactor the temperature was also
varied in the range of 300e900  C at 1 atm. The solid product of
the fuel reactor was fed to the steam oxidation reactor. In this
reactor, the FeO is to be reacted with steam according to

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
reaction (11). In this reactor the FeO/H2O feed molar ratio was
varied from 3 (stoichiometric value according to reaction (11))
to 0.546. This represents a gradual increase from 1 kmol/h of
H2O per 3 kmol/h of FeO up 6 kmol/h of H2O per 3 kmol/h of
FeO. This feed steam variation was aimed to find the minimum
amount of steam needed in order to regenerate FeO to Fe3O4 at
a reasonable oxidation temperature, where kinetics are
favorable (500e600  C) [33]. It is important to notice that all the
present simulation calculations are based on theoretical
thermodynamic considerations and these are to be taken as a
guide to further experimental evaluation of the reaction sys-
tems, since no heat and mass diffusional limitations as well as
kinetics effects were taken into account for the conformation
of the present thermodynamic analysis.
Process simulation method
Process simulation calculations was performed using the
Aspen-Plus© Engineering Process Simulator. This is a program
for simulation of chemical processes in which the analysis of
chemical processes as well as heat integration can be made.
This simulator was employed for the analysis of the process
scheme of Fig. 2. The module of Aspen-Plus that was used to
evaluate the reaction systems were: the RGibbs (Gibbs
Reactor), wherein RGibbs method is based on the Gibbs free
energy minimization technique for multiphase reactions and
material balance, cyclone units were used to separate solid
and gas streams.
Results and discussion
Thermodynamic analysis
The fuel reactor
This section presents results from the equilibrium amounts of
H2, CO, CO2, CH4 and C, Fe, FeO, Fe2O3 and Fe3O4 from the fuel
reactor in a temperature range of 300e900  C and CH4/Fe3O4
feed molar ratios from 0.25 to 1. At the studied conditions the
conversion of methane was gradually increased from about
4% at 300  C, 90% at 500  C and up to 99.9% at 600  C. This is
associated to the intrinsic equilibrium reactivity of a solid
Fe3O4 with respect to a gaseous fuel such as CH4 and avail-
ability of the metal oxide lattice oxygen into the gas phase.
Fig. 3 shows the equilibrium production of H2 (3a) and CO (3b)
from the fuel reactor (dry basis) in kmol/h as a function of
temperature and CH4/Fe3O4 feed molar ratio.
At low temperatures (300e500  C) and where CH4/Fe3O4
ratio is small (below 0.5), there is no enough temperature and/
or oxygen available for the reaction between Fe3O4 and
methane to produce syngas (CO þ H2) and therefore, this
metal oxide is poorly reactive towards reaction (9). However,
the small reactivity in this region is attributed to the complete
oxidation of methane by Fe3O4 (reaction 5) where at the SIP
conditions (CH4/Fe3O4 ¼ 0.25 and higher but not greater than
0.5) the production of CO2 and H2O dominate. Greater tem-
peratures than 500  C and CH4/Fe3O4 ratios than 0.5 will lead to
the promotion of the partial oxidation of methane by Fe3O4 to
form syngas through reaction (9) according to results pre-
sented in Fig. 3(a) and (b).
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3
Fig. 2 e Process simulation flowsheet.
Fig. 4(a) and (b) present results of formed CO2 and
unreacted CH4 contents in the gas product from the fuel
reactor, respectively.
In Fig. 4(a) it is evident that at low temperatures (T < 500  C)
and small CH4/Fe3O4 ratios, CO2 formation is favored, which is
the region where complete oxidation of methane is promoted
and the CO2 content is reduced as the available oxygen con-
tent is increased as the CH4/Fe3O4 ratio is also grown. Also in
Fig. 4(b) it is evident that methane reactivity is high at low
oxygen lattice content (Fe3O4) and this is reflected in small
methane content in the product gas at temperatures greater
than around 500  C and small CH4/Fe3O4 ratios (smaller than
0.5). Here in this plot the gradual increase in methane con-
version is clear that increases as temperature rises and the
CH4/Fe3O4 ratio decrease, which is translated in the promotion
of the partial oxidation of methane reaction with Fe3O4 (oxy-
gen starving conditions, reaction 9).
Moreover, Fig. 5(a) and (b) describe the behavior of the FeO
and C solid content at the effluent stream from the fuel reactor
of Fig. 1.
In Fig. 5(a) the maximum conversion from Fe3O4 to FeO at a
feed of 1 kmol/h of Fe3O4 is according to reaction (5) of 3 kmol/
h and that means that the iron(II) oxide is the chemical species
that in the next reactor can be combined with the appropriate
amount of steam to produce a pure hydrogen stream. There-
fore, optimal reaction conditions will be those where com-
plete conversion of Fe3O4 to FeO is achieved and those are
where the amount of FeO in Fig. 5(a) is 3 kmol/h. From this
figure, it is clear that those conditions are achieved at tem-
peratures between 500 and 700  C at all CH4/Fe3O4 ratios. At
those conditions, methane is able to reduce magnetite to
iron(II) oxide either through complete or partial oxidation of
methane. However, not only at those conditions magnetite is
also thermodynamically stable, greater temperatures from
700 up to 900  C in a CH4/Fe3O4 range from 0.375 to 0.625 will
also insure the production of magnetite and these last con-
ditions match with partial oxidation of methane through
equation (9).
Also important in the present investigation is to find
operating conditions where no carbon formation is feasible,
since any carbon contamination over the oxygen carrier will
be gasified in the oxidation reactor, thus producing carbon
dioxide gases that will ultimately combine with the expected
hydrogen production and consequently become a gaseous

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
5
byproduct. Fig. 5(b) show that the carbon free operating win-
dows of the fuel reactor are as follows: from CH4/
Fe3O4 ¼ 0.25e0.375 no carbon is present in a temperature
range of 480e900  C, CH4/Fe3O4 ¼ 0.5 from 580 to 900  C, CH4/
Fe3O4 ¼ 0.625 from 660 to 900  C and CH4/Fe3O4 ¼ 0.75e1 from
700 to 900  C. These free carbon regions that can be seen in
Fig. 5(b) is where either the Boudouard and/or the methane
decomposition reactions (12 and 13) are not thermodynami-
cally favorable. These results agree well with experimental
results reported by Tang et al. [34].
A more careful analysis of the generated data allows to
conclude that according to the previous thermodynamic
analysis optimum conditions in order to produce a maximum
hydrogen production and carbon free operation is at
T ¼ 650e700  C and CH4/Fe3O4 ¼ 0.75e1. It is important to
indicate that no metallic iron (Fe) was present at the studied
fuel reactor conditions.
The oxidation reactor
This section presents results from the equilibrium amounts of
H2 and Fe3O4 from the oxidation reactor in a temperature
range of 500e600  C and H2O feed molar flowrate range from 1
to 6 kmol/h. At the studied conditions, it is expected to find the
minimum amount of steam per three kmols of FeO at a
reasonable temperature from 500 to 600  C, where the reaction
kinetics are reported to be favorable [33]. Fig. 6(a) and (b)
presents results from the oxidation reactor at conditions
above described.
In plot 6 (a) the maximum hydrogen production can be
seen as a function of temperature and the feed molar flow to
the reactor. From this figure, it is evident that there is a
defined region where the maximum hydrogen production is
achieved, which happens to be of 1 kmol/h. These conditions
increase as the amount of steam and temperature both in-
crease. For example, at 500  C a minimum range of
3.4e6 kmol/h of steam is needed in order to produce the
maximum amount of hydrogen according to reaction (11),
while at 550  C a range of steam of 4.6e6 kmol/h are needed to
obtain the maximum possible H2 production. Finally, at 600  C
6 kmol/h of steam or greater are needed for this purpose.
However, research points out that H2 evolution via water
splitting reaction (11) requires Fe phase to be as fully reduced
as possible, while it is thermodynamically limited for Fe phase
to be completely oxidized beyond Fe3O4 using steam [34].
6 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3

Fig. 3 e Equilibrium content of H2 (a) and CO (b) for the fuel reactor.

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3 7

Fig. 4 e Equilibrium content of CO2 (a) and CH4 (b) for the fuel reactor.

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
8 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3

Fig. 5 e Equilibrium content of FeO (a) and C (b) for the fuel reactor.

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3 9

Fig. 6 e Equilibrium content of H2 (a) and H2O (b) for the oxidation reactor.

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
10 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3

Moreover, other research indicate that the reduction from
Fe3O4 to Fe promotes carbon deposition [30].
Furthermore, Fig. 6(b) presents the Fe3O4 content at
different steam feed to the oxidation reactor. Here in this plot,
it can be seen that the only conditions where complete con-
version of FeO to Fe3O4 is achieved at regions where the
amount of steam and temperature both increase. These re-
sults mimic the same conditions where the maximum
hydrogen production is achieved from Fig. 5(a). Therefore, this
regenerated 1 kmol/h of Fe3O4 is expected to be recirculated
back to the fuel reactor.
Process simulation
Fig. 1 presents a diagram of the process scheme employed
during the simulation of the SIP/CL-SMR reaction system with
either the production of pure CO2 or a syngas stream from the
fuel reactor, while in the oxidation reactor the H2 evolution
and iron oxide regeneration is achieved. Table 1 summarizes
results from the process simulation of selected streams from
Fig. 1.
In this Table, results from the extreme case where com-
plete production of CO2 is aimed (SIP) and the CH4/Fe3O4 feed
molar ratio is equal to 0.25 are presented. Here in this table
stream INPUT-1 denotes the fresh CH4 feed of 0.25 kmol/h,
while the RECIRC stream contains 1 kmol/h of Fe3O4. Stream
OUTPUT-1 contains the products from the fuel reactor at a
temperature of 700  C where no carbon formation is possible
and the complete oxidation of methane by magnetite is
favored. In this stream, a methane conversion of 99.99% is
achieved with the production of 0.23 kmol/h of CO2 compared
to the theoretical 0.25 kmol/h according to reaction (5) making
these operating conditions very convenient for carbon
sequestration purposes, while the other carbon specie pro-
duced is CO with 0.017 kmol/h accompanied with the pro-
duction of water. Furthermore, in this stream only 2.79 kmol/h
of FeO is generated, while only a small amount of unreacted
Fe3O4 is present (0.068 kmol/h). Also, in this table the dry-basis
(DB) gas mol fractions in GASP-1 stream coming from the fuel
reactor are presented. Here, CO2 mol fraction in dry-basis (DB)
is 0.77, while H2 is only 0.17 with traces of CO (0.06) with no
presence of unreacted CH4. These results are consistent with
the promotion of reaction (5) conducing to a complete oxida-
tion of methane to produce FeO (2.80 kmol/h) as observed in
Table 1.
According to Fig. 2, gases and solids are separated in
streams GASP-1 and INPUT-2, where this last is fed to the
oxidation reactor together with stream STEAM, which con-
sists of 6 kmol/h at a reactor temperature of 600  C. Stream
OUTPUT-2 contains the products from the oxidation reactor
where 0.93 kmol/h of H2 combined with 1 kmol/h of Fe3O4 and
5 kmols/h of excess steam are produced. This stream is
separated in H2PROD and RECIRC streams, where this last one
is stent back to the fuel reactor. The H2PROD stream contains a
very high concentration of H2 (mol fraction of 1, dry basis) and
once this water is separated in this stream, pure H2 can be
obtained. The yield of this process based on the hydrogen
product is of 3.73 mol of H2 produced per mols of methane
being fed. Kang et al. [28] reported similar a yield in their
simulation studies of the reduction of methane with Fe3O4,
thus producing a yield of 2.64 mol of H2 per 1 mol of CH4.
However, this study was performed under thermally neutral
conditions at 850  C with CO2 capture in the first reactor.
Moreover, Table 2 presents results from the simulation of
the CL-SMR process, which represents the other mode of
operation described in Fig. 1. In this simulation a CH4/Fe3O4
feed molar ratio equal to 1 was employed in the fuel reactor.
In this table stream INPUT-1 contains the fresh CH4 feed of
1 kmol/h, while the RECIRC stream contains also 1 kmol/h of
Fe3O4. Stream OUTPUT-1 comprises the products from the
fuel reactor at a temperature of 700  C where no carbon for-
mation is possible and the complete oxidation of methane by
magnetite is favored. In this stream a methane conversion of
90% is reflected with the production of 1.54 kmol/h of H2
compared to the theoretical 2 kmol/h according to reaction (9),
while the production of CO was 0.7 kmol/h compared to the
theoretical value of 1 kmol/h of the same reaction. With
respect to the reaction of the Fe3O4 with methane, it is

Table 1 e Summary of process simulation results of the SIP process.

Stream name INPUT1 RECIRC OUTPUT-1 GASP-1 INPUT2 STEAM OUTPUT-2 H2PROD
Mole flow, Kmol/h
CH4 0.25 0.00 0.00 0.00 0.00 0.000 0.00 0.00
CO 0.00 0.00 0.02 0.02 0.00 0.000 0.00 0.00
CO2 0.00 0.00 0.23 0.23 0.00 0.000 0.00 0.00
H2 0.00 0.00 0.05 0.05 0.00 0.000 0.93 0.93
H2O 0.00 0.00 0.45 0.45 0.00 6.000 5.07 5.07
Fe3O4 0.00 1.00 0.07 0.00 0.07 0.000 1.00 0.00
FeO 0.00 0.00 2.80 0.00 2.80 0.000 0.00 0.00
Fe 0.00 0.00 0.00 0.00 0.00 0.000 0.00 0.00
Mole Flow, Kmol/h 0.25 1.00 3.61 0.75 2.86 6.000 7.00 6.00
Mass Flow, Kg/h 4.01 231.54 235.55 18.92 216.63 108.092 324.72 93.18
Volume Flow, m3/h 20.38 0.04 61.17 61.12 0.05 429.69 429.82 429.78
Temperature,  C 720 600 720 720 720 600 600 600
Mol fractions (DB)
CH4 1 e 0.00 0.00 e e 0 0
CO 0 e 0.06 0.06 e e 0 0
CO2 0 e 0.77 0.77 e e 0 0
H2 0 e 0.17 0.17 e e 1 1

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3 11

Table 2 e Summary of process simulation results of the CL-SMR process.

Stream Name INPUT1 RECIRC OUTPUT-1 GASP-1 INPUT2 STEAM OUTPUT-2 H2PROD
Mole flow, Kmol/h
CH4 1.00 0.00 0.10 0.10 0.00 0.00 0.00 0.00
CO 0.00 0.00 0.70 0.70 0.00 0.00 0.00 0.00
CO2 0.00 0.00 0.20 0.20 0.00 0.00 0.00 0.00
H2 0.00 0.00 1.54 1.54 0.00 0.00 1.36 1.36
H 2O 0.00 0.00 0.26 0.26 0.00 8.20 6.84 6.84
Fe3O4 0.00 1.00 0.00 0.00 0.00 0.00 1.00 0.00
FeO 0.00 0.00 2.64 0.00 2.64 0.00 0.00 0.00
Fe 0.00 0.00 0.36 0.00 0.36 0.00 0.00 0.00
Mole Flow, Kmol/h 1.00 1.00 5.80 2.80 3.00 8.20 9.20 8.20
Mass Flow, Kg/h 16.04 231.54 247.58 37.78 209.80 147.73 357.53 125.99
Volume Flow, m3/h 81.52 0.04 223.66 223.61 0.05 587.25 587.42 587.37
Temperature,  C 720 600 700 700 700 600 600 600
Mol fractions (DB)
CH4 1 e 0.04 0.04 e e 0 0
CO 0 e 0.28 0.28 e e 0 0
CO2 0 e 0.08 0.08 e e 0 0
H2 0 e 0.61 0.61 e e 1 1

significant that two reduced iron species were produced with
a combination of 2.64 and 0.36 kmol/h of FeO and Fe3O4,
respectively. Here, it is possible that the syngas formed
(H2 þ CO) is able to further reduce the FeO to a small extent.
Also in this Table, the dry-basis (DB) mol fractions of the
product gas from the fuel reactor is presented. Here, mol
fractions of 0.61 and 0.28 of H2 and CO, respectively were ob-
tained with only small concentrations of unreacted methane
(0.04) and CO2 (0.08) being present. These results represent
that reaction (9) is being dominant at this conditions. These
results agree well with data reported by Kang et al. [28] who
found that in the case of the Fe oxide redox pairs at T ¼ 700  C,
Fe/Fe3O4 showed higher H2 mole fractions (0.649) than FeO/
Fe3O4 (0.404), Fe0.947O/Fe3O4 (0.621) and Fe0.945O/Fe3O4 (0.444)
showed higher H2 mole fractions than FeO/Fe3O4, even though
FeO has a lower oxidation state than Fe0.947O and Fe0.945O.
Although wüstite has the ideal formula FeO, it is a non-
stoichiometric compound. Therefore, Fe0.945O and Fe0.947O
are types of wüstite. Consequently, at a lower oxidation state
of the metal oxide the H2 mole fraction would be higher.
Furthermore, Tong et al. [35] calculated the phase diagram of
the iron oxidation states when subjected to varying partial
pressures of H2O/H2 and temperatures. At a given temperature
their phase diagram can describe the relationship between
iron oxide and fuel conversions where the operating line for
each mode of operation can be derived from an analytical
oxygen balance between the gas and solid species in the
reducer reactor. According to their data at 600  C the corre-
sponding H2 mol fraction is of 0.62 (highly consistent with
results found in the present research), while a maximum mol
fraction of 0.72 will indicate the formation of Fe, which the
present research aimed to avoid, since experimental results
from Galvita and Sundmacher [30] reported that reduction
towards Fe produced carbon deposits on the oxygen carrier.
Moreover, no other chemical species were found during
the equilibrium calculations such as iron carbides or carbon-
ates, which their content in the solid product from the
reducing reactor were negligible. This behavior agrees well
with findings from Svoboda et al. [30], who found that even
though iron carbide (Fe3C) is feasible to be formed under the
reducing conditions of the fuel reactor at temperatures greater
than 427  C, the presence of water-steam in the product
(which is the case of the present simulation results) contrib-
utes to a high instability of the carbide species.
In Fig. 2, coming from the fuel reactor, gases and solids are
separated in streams GASP-1 (syngas) and INPUT-2 (solid
product), where this last is fed to the oxidation reactor
together with stream STEAM, which according to results from
Table 2 consists of 8.2 kmol/h at a reactor temperature of
600  C. This excess of steam being fed at the oxidation reactor
is needed to re-oxide FeO and Fe reduced species back to
Fe3O4. Also, from Table 2 OUTPUT-2 stream contains the
products from the oxidation reactor where 1.36 kmol/h of H2
combined with 1 kmol/h of Fe3O4 and 6.8 kmols/h of excess
steam. This stream is separated in H2PROD and RECIRC
streams, where this last one is stent back to the fuel reactor.
The yield of this process based on the hydrogen product is
1.36 mol of H2 produced per 1 mol of methane being fed. While
based on the syngas produced is 2.24 mol of syngas (CO þ H2)
per mol of methane fed along with a H2/CO molar ratio of 2.19.
Go et al. [36] reported a synthetic gas (syngas) having the H2/
(2CO þ 3CO2) molar ratio of 0.65, which is consistent with a
value of 0.77 calculated from OUPUT-1 stream from results of
Table 2. Furthermore, the value of H2/CO ¼ 2.19 is very
convenient, since this streams can be used in the Fischer-
Tropsch process for the production of liquid fuels (GTL).
Conclusions
In the present research, the redox performance of iron oxide
FeO/Fe3O4 redox pair as an oxygen carrier was investigated for
hydrogen (H2) and/or syngas (H2þCO) production through a
methane-steam redox process using a thermodynamic anal-
ysis and process simulation to find most favorable reactions
conditions. The studied reaction system was divided in two
reactors. In the fuel reactor the complete and/or partial
oxidation of methane with Fe3O4 were studied. For the case
where complete oxidation of CH4 along with production of
CO2 and steam was aimed (SIP), optimal reaction conditions in

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
12 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3
the first reactor include: a CH4/Fe3O4 molar feed of 0.25,
T ¼ 700  C (with no carbon formation) and the complete
oxidation of CH4 by Fe3O4 with a methane conversion of
99.99%, producing 2.79 kmol/h of FeO. While optimal condi-
tions in the second reactor comprise a steam feed of 6 kmol/h
and a reactor T ¼ 600  C, producing 0.93 kmol/h of H2 com-
bined with 1 kmol/h of Fe3O4 and 5 kmols/h of excess steam.
The yield of this process (steam reactor) in molar basis was
3.72H2 per CH4 feed.
For the case where the production of syngas was aimed in
the fuel reactor, optimal reaction conditions include: a CH4/
Fe3O4 molar feed of 1, T ¼ 700  C (with no carbon formation)
and the partial oxidation of CH4 with small amounts of CO2
and unreacted CH4, along with 1 kmol/h of Fe3O4 with a
methane conversion of 90%. While optimal conditions in the
steam oxidation reactor comprise a steam feed of 8.20 kmol/h
and a reactor T ¼ 600  C, producing 1.36 kmol/h of H2 com-
bined with 1 kmol/h of Fe3O4. The yield of this process (fuel
reactor) in molar basis was 1.54 mol H2 per CH4 feed.
Finally, these results can be taken as a basis for future
experimental and theoretical studies in search for a suitable
catalyst and experimental conditions to evaluate the present
proposed technology.
references
[1] Mueller-Langer F, Tzimas E, Kaltschmitt M, Peteves S.
Techno-economic assessment of hydrogen production
processes for the hydrogen economy for the short and
medium term. Int J Hydrogen Energy 2007;32:3797e810.
[2] Markit HIS. Hydrogen. Chemical economics handbook. June
2015. Available at: https://www.ihs.com/products/hydrogen-
chemical-economics-handbook.html.
[3] Yang J, Cai N, Li Z. Hydrogen production from the steam-iron
process with direct reduction of iron oxide by chemical
looping combustion of coal char. Energy Fuels
2008;22:2570e9.
[4] Bayham SC, Kim HR, Wang D, Tong A, Zeng L, McGiveron O,
et al. Iron-based coal direct chemical looping combustion
process: 200-h continuous operation of a 25-kwth subpilot
unit. Energy Fuels 2013;27:1347e56.
[5] Lee DH, Cha KS, Lee YS, Kang KS, Park CS, Kim YH. Effects of
Ceo2 additive on redox characteristics of Fe-Based mixed
oxide mediums for storage and production of hydrogen. Int J
Hydrogen Energy 2009;34:1417e22.
[6] Otsuka K, Kaburagi T, Yamada C, Takenaka S. Chemical
storage of hydrogen by modified iron oxides. J Power Sources
2003;122:111e21.
[7] Takenaka S, Hanaizumi N, Son VTD, Otsuka K. Production of
pure hydrogen from methane mediated by the redox of Ni-
and Cr-Added iron oxides. J Catal 2004;228:405e16.
[8] Urasaki K, Tanimoto N, Hayashi T, Sekine Y, Kikuchi E,
Matsukata M. Hydrogen production via steam-iron reaction
using iron oxide modified with very small amounts of
palladium and zirconia. Appl Catal A 2005;288:143e8.
[9] Zafar Q, Mattisson T, Gevert B. Redox investigation of some
oxides of transition-state metals Ni, Cu, Fe, and Mn
supported on Sio2 and MgAl2O4. Energy Fuels 2006;20:34e44.
[10] Galvita V, Sundmacher K. Cyclic water gas shift reactor
(CWGS) for carbon monoxide removal from hydrogen feed
gas for PEM fuel cells. Chem Eng J 2007;134:168e74.
[11] Liotta LF, Longo A, Pantaleo G, Di Carlo G, Martorana A,
Cimino S. Alumina supported Pt (1%)/Ce0.6Zr0.4O2 monolith:

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130
remarkable stabilization of ceria-zirconia solution towards
CeAlO3 formation operated by Pt under redox conditions.
Appl Catal B 2009;90:470e7.
[12] Sim A, Cant NW, Trimm DL. Ceria-zirconia stabilized
tungsten oxides for the production of hydrogen by the
methane-water redox cycle. Int J Hydrogen Energy
2010;35:8953e61.
[13] Trovarelli A. Catalytic properties of ceria and CeO2-
containing materials. Catal Rev Sci Eng 1996;38:439e520.
[14] Sosa Va
zquez MI, Delgado Vigil MD, Salinas Gutie
rrez J,
Collins-Martı́nez V, Lo
pez Ortiz A. Synthesis gas production
through redox cycles of bimetallic oxides and methane. J
New Mater Electrochem Syst 2009;12:29e34.
[15] Zhu X, Wei Y, Wang H, Li K. CeeFe oxygen carriers for
chemical-looping steam methane reforming. Int J Hydrogen
Energy 2013;38:4492e501.
[16] Zheng Y, Zhu X, Wang H, Li K, Wang Y, Wei Y. Characteristic
of macroporous CeO2-Zro2 oxygen carrier for chemical-
looping steam methane reforming. J Rare Earths
2014;32:842e8.
[17] Zhu X, Li K, Wei Y, Wang H, Sun L. Chemical-looping steam
methane reforming over a CeO2-Fe2O3 oxygen carrier:
evolution of its structure and reducibility. Energy Fuels
2014;28:754e60.
[18] Lee DH, Cha KS, Kim HS, Park CS, Kim YH. Syngas and
hydrogen production via stepwise methane reforming over
Cu-Ferrite/YSZ. Int J Energy Res 2014;38:1522e30.
[19] He F, Wei Y, Li H, Wang H. Synthesis gas generation by
chemical-looping reforming using Ce-Based oxygen carriers
modified with Fe, Cu, and Mn oxides. Energy Fuels
2009;23:2095e102.
[20] Adanez J, Abad A, Garcia-Labiano F, Gayan P, De Diego LF.
Progress in chemical-looping combustion and
reforming technologies. Prog Energy Combust Sci
2012;38:215e82.
[21] Mattisson T, Johansson M, Lyngfelt A. Multicycle reduction
and oxidation of different types of iron oxide particles e
application to chemical-looping combustion. Energy Fuels
2004;18:628e37.
[22] Mattisson T, Lyngfelt A, Cho P. The use of iron oxide as an
oxygen carrier in chemical-looping combustion of methane
with inherent separation of CO2. Fuel 2001;80:1953e62.
[23] Nasr S, Plucknett K. Kinetics of iron ore reduction by
methane for chemical looping combustion. Energy Fuels
2014;28:1387e95.
[24] Cabello A, Abad A, Garcı́a-Labiano F, Gaya
n P, de Diego LF,
Ada
nez J. Kinetic determination of a highly reactive
impregnated Fe2O3/Al2O3 oxygen carrier for use in gas-fueled
chemical looping combustion. Chem Eng J 2014;258:265e80.
[25] Monazam ER, Breault RW, Siriwardane R. Kinetics of
magnetite (Fe3O4) oxidation to hematite (Fe2O3) in air for
chemical looping combustion. Ind Eng Chem Res
2014;53:13320e8.
[26] Fraser SD, Monsberger M, Hacker V. A thermodynamic
analysis of the reformer sponge iron cycle. J Power Sources
2006;161:420e43.
[27] Steinfeld A, Kuhn P. High-temperature solar
thermochemistry: production of iron and synthesis gas by
Fe3O4-reduction with methane. Energy 1993;18:239e49.
[28] Kang K, Kim C, Bae K, Cho W, Kim S, Park C. Oxygen-carrier
selection and thermal analysis of the chemical-looping
process for hydrogen production. Int J Hydrogen Energy
2010;35:12246e54.
[29] Bohn CD, Muller CR, Cleeton JP, Hayhurst AN, Davidson JF,
Scott SA, et al. Production of very pure hydrogen with
simultaneous capture of carbon dioxide using the redox
reactions of iron oxides in packed beds. Ind Eng Chem Res
2008;47:7623e30.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 3 13

[30] Svoboda K, Slowinski G, Rogut J, Baxter D. Thermodynamic
possibilities and constraints for pure hydrogen production by
iron based chemical looping process at lower temperatures.
Energy Convers Manag 2007;48:3063e73.
[31] Galvita V, Sundmacher K. Hydrogen production from
methane by steam reforming in a periodically operated two-
layer catalytic reactor. Appl Catal A 2005;289:121e7.
[32] Collins-Martı́nez V, Escobedo Bretado M, Mele
ndez
Zaragoza M, Salinas Gutie
rrez J, Lo

pez Ortiz A. Absorption
enhanced reforming of light alcohols (methanol and ethanol)
for the production of hydrogen: thermodynamic modeling.
Int J Hydrogen Energy 2013;38:12539e53.
[33] Yamaguchi D, Tang L, Wonga L, Burke N, Trimm D,
Nguyen K, et al. Hydrogen production through methane-
steam cyclic redox processes with iron-based metal oxides.
Int J Hydrogen Energy 2011;36:6646e56.
[34] Tang M, Xu L, Fan M. Progress in oxygen carrier development
of methane-based chemical-looping reforming: a review.
Appl Energy 2015;15:1143e56.
[35] Tong A, Bayham S, Kathe MV, Zeng L, Luo S, Fan L. Iron-
based syngas chemical looping process and coal-direct
chemical looping process development at Ohio State
University. Appl Energy 2014;113:1836e45.
[36] Go KS, Son SR, Kim SD, Kang KS, Park CS. Hydrogen
production from two-step steam methane reforming in
a fluidized bed reactor. Int J Hydrogen Energy
2009;34:1301e9.

pez-Ortiz A, et al., Hydrogen production by a Fe-based oxygen carrier and methane-steam redox
Please cite this article in press as: Lo
process: Thermodynamic analysis, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.07.130