Highly selective CO removal by sorption enhanced Boudouard reaction for hydrogen production

Article abstract of DOI:10.1039/c9cy00851a

The development of an energy-efficient and cost-effective technology for the purification of a hydrogen rich stream to achieve a CO concentration below 10 ppm, suitable for low temperature fuel cell applications, is one of the major objectives of energy research. Here we report the sorption enhanced Boudouard (SEB) reaction as an effective route for pure hydrogen production on metal-free catalysts by the highly selective removal of CO and CO₂ from a hydrogen rich stream, where CaO serves as both a catalyst and a CO₂ sorbent. We reveal that the in situ generated oxygen vacancy by CO adsorption on CaO catalyzes the sorption enhanced Boudouard reaction, hence the Boudouard reaction and CO₂ removal occur simultaneously in a single step. The capture of CO₂ in the presence of H₂ by the solid sorbent shifts the chemical equilibrium towards complete CO conversion. The results demonstrate a remarkable decrease in the CO concentration from 0.5% in a hydrogen rich stream to less than 5 ppm, suitable for low temperature fuel cells. It avoids a subsequent preferential oxidation process or methanation reaction, as well as an additional CO₂ capture process. The feasibility of applying the sorption enhanced Boudouard reaction in the production of highly pure hydrogen (CO content less than 10 ppm) was demonstrated by multicycle tests of Boudouard-carbonation/regeneration on Ca-based oxides with stable cyclic operation.

Full text of DOI:10.1039/c9cy00851a

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: D. Chen, K. R.
Rout and V. Gil, Catal. Sci. Technol., 2019, DOI: 10.1039/C9CY00851A.
Volume
Number
21 January 2017
Pages 313-534
7
2
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
rsc.li/catalysis
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2044-4761
PAPER
Chaoqiu Chen, Yong Qin et al.
Highly dispersed Pt nanoparticles supported on carbon
nanotubes produced by atomic layer deposition for hydrogen
generation from hydrolysis of ammonia borane
rsc.li/catalysis

Please do not adjust margins
Catalysis Science & Technology
Page 1 of 8
View Article Online
DOI: 10.1039/C9CY00851A
ARTICLE
Highly Selective CO Removal by Sorption Enhanced Boudouard
Reaction for Hydrogen Production
Kumar R Routa*, María V Gilb and De Chenb*
Received 00th January 20xx,
Accepted 00th January 20xx
Development of an energy-efficient and cost-effective technology for purification of a hydrogen rich stream to achieve a CO
concentration below 10 ppm, suitable for low temperature fuel cell applications, is one of major objectives of the energy
research. Here we report the sorption enhanced Boudouard (SEB) reaction as an effective route for pure hydrogen
production on metal-free catalysts by highly selective removal of CO and CO2 from a hydrogen rich stream, where CaO serves
as both catalyst and CO2 sorbent. We reveal that the in-situ generated oxygen vacancy by CO adsorption on CaO catalyzes
the sorption enhanced Boudouard reaction, hence Boudouard reaction and CO2 removal occurring simultaneously in a single
step. The capture of CO2 in the presence of H2 by the solid sorbent shifts the chemical equilibrium towards the complete CO
conversion. The results demonstrate a remarkable decrease in the CO concentration from 0.5% in a hydrogen rich stream
to less than 5 ppm, suitable for low temperature fuel cells. It avoids a subsequent preferential oxidation process or
methanation reaction, as well as an additional CO2 capture process. The feasibility of applying the sorption enhanced
Boudouard reaction in the production of highly pure hydrogen (CO content less than 10 ppm) was demonstrated by
multicycle tests of Boudouard-carbonation/regeneration on Ca-based oxides with stable cyclic operation.
DOI: 10.1039/x0xx00000x
www.rsc.org/
the WGS reactors, CO reacts with water and produces CO2 and H2 by
Eq. 1:
1. Introduction
⟷
CO + H2O
CO2 + H2
(1)
It is widely accepted that hydrogen is an environmentally friendly
and clean energy carrier for a sustainable energy system [1]. World
consumption of hydrogen is expected to increase 3.5 percent
annually to more than 300 billion cubic meters, not only for
conventional industrial uses, such as petroleum refinery, chemical
production, food processing, pharmaceutical or metal processing,
but also for clean energy technologies, such as fuel cell vehicles and
bio-refinery. Global hydrogen production mostly comes from natural
gas (48%), oil (30%) and coal (18%), water electrolysis accounting for
only 4% [2]. The most established technology to produce hydrogen
on a commercial scale is nowadays the steam methane reforming
(SMR) of natural gas. The synthesis gas is the typical intermediate for
hydrogen production. High temperature- and low temperature
water-gas shift (WGS) reactions are extensively used to increase the
hydrogen content in the synthesis gas. Commercially, the syngas is
usually passed through a dual-stage WGS reaction, with two reactors
in series: a high temperature reactor (300-450 °C) and a low
temperature reactor (180-270 °C). The high temperature reactor
usually contains an iron oxide catalyst to convert bulk CO from ~15%
to ~3%, whereas the low temperature reactor normally contains a
copper catalyst for reducing the CO level to less than 0.4% [3, 4]. In
The product gas coming out from the low temperature WGS
reactor still contains a high amount of CO, which is not suitable for
certain uses, such as low temperature fuel cells [5]. It is well known
that the fuel cell technology offers a highly efficient conversion of
chemical energy into electrical energy without emission of
environmental pollutants, thereby making fuel cells one of the most
promising sources of power generation [5, 6]. Proton exchange
membrane (PEM) fuel cells are a promising type of low temperature
fuel cell being developed for transport applications as well as for
stationary and portable fuel cell applications. The major challenge for
low temperature fuel cell is that they are extremely sensitive towards
impurities such as COx. The CO concentration in the hydrogen gas
stream is typically required to be lower than 10 ppm [7]. Purification
of the hydrogen stream is then essential, and several technologies
are under development to produce fuel cell grade hydrogen, such as
preferential oxidation (PROX) [5, 8, 9], selective methanation [10-
12], sorption enhanced water gas shift (SEWGS) reaction [13-16] and
pressure swing adsorption (PSA) [17, 18]. In some cases, following
the shift reactors, the syngas is fed to a CO2 capture unit based on
physical solvents like selexol, rectisol, or chemical ones like amine-
based solvents. However, for a high purity hydrogen production, a
PSA unit is needed as final step.
Therefore, it is deduced that the removal of COx up to obtain ppm
levels in a hydrogen stream makes the overall process quite complex,
lengthy and with low yield, which hinders the application of the
existing technology for hydrogen production in vehicular and small-
scale fuel cell applications. Furthermore, besides the consumption of
fossil fuels, the overall process entails high capital costs [19]. Thus,
aSINTEF Industry, Norway
bDepartment of Chemical Engineering, Norwegian University of Science and
Technology, Sem Sælands vei 4, Trondheim, NO-7491, Norway
E-mail: kumarranjan.rout@sintef.no
E-mail: de.chen@ntnu.no
This journal is © The Royal Society of Chemistry 2018
1
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 2 of 8
ARTICLE
the development of more energy-efficient and cost-effective
technologies for pure hydrogen production is highly desired. Hence,
it is essential to find new routes for the removal of CO from hydrogen
streams with the specific objective of application in fuel cell systems,
which are highly selective and low-cost processes that provide the
high yields of hydrogen needed. In the present work, the sorption
enhanced Boudouard (SEB) reaction is purposed as a one-step high-
purity hydrogen production process, which could greatly simplify the
procedure, increase the energy efficiency and reduce costs.
2. Experimental
2.1. CO2 sorbent
View Article Online
DOI: 10.1039/C9CY00851A
Arctic dolomite was supplied by Franefoss Miljøkalk As, Norway. This
has a purity of approximately 98.5 wt.% CaMg(CO3)2 and no sulfur
according to X-ray fluorescence analysis. The dolomite was used as a
precursor of CaO for the capture of CO2. The initial maximum CO2
capture capacity was estimated as being 0.46 g CO2/g sorbent. The
dolomite was ground and sieved to reach a particle size of 250-500
μm. It was calcined in an air flow (200 mL min-1) at 900
°C for 4 h
Nowadays, an important research work is being carried out on
the calcium looping process (CaL), which is a calcium sorbent-based
chemical looping process that has the potential to increase the H2/CO
ratio [20]. The CaL integrates the WGS reaction with in-situ CO2,
sulfur and halide removal at high temperatures in a single-stage
reactor. It utilizes a high temperature regenerable calcium oxide-
based sorbent (CaO), which apart from capturing the CO2, enhances
the yield of H2 and simultaneously captures sulfur and halide
impurities. However, it introduces issues and costs associated with
the separation of the sorbent and catalyst prior to calcination, with
the pretreatment of the catalyst to the active form after its
deactivation in the presence of CO2 in the calciner at high
temperatures, as well as with the replacement of the spent catalyst
[21]. Recently, the WGS reaction has been performed in the presence
of only a CaO sorbent [21]. The results obtained lead to the
conclusion of that the WGS reaction takes place to a considerable
extent even in the absence of a metal catalyst at relatively higher
temperatures than the conventional WGS process. Thus, high purity
H2 can be produced with very low levels of CO and CO2 during the
pre-breakthrough region of the sorption curve while the CaO sorbent
is active. However, the yield of H2 has not been reported in that study
[21]. It is believed that the high purity of H2 is due to the shifting of
the reaction equilibrium to the forward direction when the CO2
formed is removed by the CaO. However, the possible interaction
between CO and CaO has not been studied properly yet. Therefore,
in the present work an attempt has been made for the first time to
study the feasibility of the reaction between CO and CaO via the
Sorption Enhanced Boudouard (SEB) reaction to increase the ratio of
H2/CO in the syngas.
prior to its storage.
2.2. Experimental procedure
The SEB reaction experiments were carried out by passing the gas
mixture over dolomite in homemade fixed bed reactor
a
thermogravimetric analyzer (TGA), where a good gas-solid contact is
obtained. Gases were delivered by Bronkhorst® mass flow
controllers. The dolomite, 3-20 mg sample, was inserted in the
stainless tube fixed bed reactor, while the reactor is suspended from
an electronic balance. A mass spectrometer (MS) was used to analyze
the gas evolved during the experiments. The experiments were
performed at 420-550
carried out using a regeneration step at 800
°
C and 1 atm. A cyclic experiment was also
°
C in an N2 or air flow
(1 NL min-1). The regeneration temperature was selected taking into
consideration the thermodynamic limitations of the decarbonation
reaction and the kinetics of the decarbonation of dolomite [22].
The dolomite sample was exposed to the inlet gas mixture for a
time sufficiently long (approximately 250 minutes) to determine the
weight gain of the sorbent. Thus, the weight change of the dolomite
sample was continuously monitored.
3. Thermodynamic analysis of sorption enhanced
Boudouard reaction on CaO
The Boudouard reaction is the redox reaction to form carbon dioxide
and solid carbon from carbon monoxide, as it is shown in Eq. 2 [23]:
⟷
2CO
CO2 + C
∆H = - 172 kJ/mol
(2)
In this work, we illustrate the feasibility of highly selective
removal of CO by the SEB reaction up to reach ppm levels without
consumption of hydrogen by thermodynamic analysis and
experimental demonstration. We further demonstrate the feasibility
of a continuous process by a cyclic carbonate looping process where
the carbon formed by the SEB reaction can be burned effectively in
air to supply heat for the regeneration of the saturated CaO
materials. We study the effects of operation conditions, such as the
concentration of H2 and the temperature, on the activity of the SEB
reaction. We also demonstrate a remarkable reduction of the CO
concentration from 0.5% in a syngas up to 5-10 ppm, which is
suitable for low temperature fuel cells and avoids a subsequent
preferential oxidation process. An oxygen vacancy mediated catalytic
cycle of Boudouard reaction on calcium oxide is proposed in this
study.
The Boudouard reaction is reversible and exothermic at all
temperatures. However, at high temperatures (typically>700 ºC),
when the large positive entropic term (T·ΔS) prevails the enthalpic
component, the Gibbs free energy becomes negative (ΔG=ΔH-T·ΔS)
and the reaction proceeds spontaneously to produce CO [24].
Likewise, at lower temperatures the forward exothermic reaction
toward the formation of CO2 is favored, but the conversion is limited
by thermodynamics. Based on Le Chatelier’s principle, the removal
of CO2 in-situ by carbonation of an alkaline metal oxide such as CaO
(Eq. 3) could enhance the conversion. The total reaction (Eq. 4) is
called sorption enhanced Boudouard reaction, which is highly
exothermic.
⟷
CaO + CO2
CaCO3
∆H = - 178 kJ/mol
(3)
The overall SEB reaction is shown in Eq. 4 as follows:
⟷
CaO + 2CO
CaCO3 + C
∆H = - 350 kJ/mol
(4)
2
Please do not adjust margins

Page 3 of 8
Catalysis Science & Technology
Thermodynamic analyses of the Boudouard and the SEB but also it lowers the temperature required for regeneration by
reactions were conducted between 400 and 700 °C. The equilibrium decreasing the CO2 pressure.
composition was estimated by minimization of the Gibbs free energy A relevant advantage of using the SEB rDeaOcIt:i1o0n.10fo3r9/tCh9eCrYe0m08o5v1aAl
View Article Online
of the system using Aspen Plus 8.6 software. The RGibbs reactor was of CO is that it does not require a metal catalyst or steam during the
specified as the reaction system and the Peng-Robinson reaction. The decomposition of CO by the Boudouard reaction to
thermodynamic package as the property method. The species form carbon is a very slow reaction and hence is usually catalyzed by
included in the product pool were CO, CO2, C (graphite as solid transition metals such as Fe, Ni or Co [25]. A CO2 sorbent has
carbon), CaO and CaCO3. The equilibrium partial pressure of CO as a previously been used in literature [26, 27] to enhance the purity of
function of temperature in a range of 400 and 700 °C is plotted in hydrogen from syngas using the Boudouard reaction in conjunction
Fig. 1 for the thermodynamic analysis of both the SEB reaction and of water gas shift reaction, but using metal catalysts and steam.
the Boudouard reaction. The results illustrate that the equilibrium However, it will be demonstrated in the present work that the SEB
partial pressure of CO for the SEB reaction is much lower than that reaction would avoid the need of using a conventional catalyst by the
for the Boudouard reaction at all the temperatures studied, clearly shifting of the reaction equilibrium through the in situ CO2 removal
indicating the sorption enhanced effects. Moreover, for the SEB by a CO2 sorbent. An additional advantage of the SEB reaction is that
reaction at 400 °C, the CO and CO2 partial pressure values are close the thermodynamic analysis (one mole of CO2 is formed from two
to zero, while between 400 and 450 °C only CO is present. This means moles of CO) indicates that it is favored at higher pressures, so the
that in the 400-450 °C temperature range almost all the CO2 formed efficiency of the CO disproportionation via the Boudouard reaction
is captured by the sorbent. It suggested that fuel cell grade hydrogen will be expected to increase at elevated pressures.
(CO content less than 10 ppm) can be obtained by fixation of CO to
solids through the SEB reaction at mediate temperatures. In addition,
4. Results and discussion
based on Le Chatelier’s principle high pressure shifts the reaction
towards CaCO3 and C formation, thus enhancing the conversion of
CaO.
4.1. Sorption enhanced Boudouard (SEB) reaction
For the SEB reaction, thermodynamic analysis in Fig. 1 indicates that
at 400 °C the CO level is 3 ppm and no CO2 present. The CO level
increases from 3 ppm at 450 °C to 10 ppm at 490 °C. This implies that
the SEB reaction can produce such low CO and CO2 levels as required
for low temperature fuel cells. Fig. 2 shows the experimental results
of the SEB reaction in a CO2/CO/H2/N2 gas mixture on dolomite at
550 °C during the carbonation stage. The concentrations of CO2, CO
and H2 in the gas mixture simulated the syngas composition.
Afterwards, decarbonation was carried out in N2 at 800 °C. The
weight change, ∆w, of the dolomite as a function of the time of
experiment (TOS: time on stream) is shown in Fig. 2a, while the outlet
gas concentrations during the carbonation stage as a function of the
TOS are shown in Fig. 2b. The gas consumption rates of CO and CO2
are shown as a function of TOS in Fig. 2c.
Fig. 2b shows that the outlet CO2 concentration increases rapidly
reaching the value of the inlet concentration (15 vol.%). Afterwards,
the outlet CO2 concentration becomes slightly higher than the inlet
CO2 concentration, indicating that additional CO2 is formed via the
Boudouard reaction. The outlet concentration of CO is much lower
than the inlet CO concentration and gradually decreases with TOS.
Fig. 2c shows the gas consumption rates of CO and CO2 with the time,
where the initial CO2 consumption rate is very high, but it
The equilibrium CO2 concentration for the SEB reaction increases
concurrently with increasing temperature because of the
unfavorable thermodynamics of the carbonation reaction at high
temperature. This led to a weaker sorption enhancement at higher
temperatures, which is reflected by the increase in the CO
concentration as temperature increases.
The sorbent can be regenerated by air or oxygen following the
Eq. (5):
⟷
CaCO3 + C + O2
CaO + 2CO2 ∆H = - 41 kJ/mol (5)
The highly exothermic oxidation of solid carbon is effectively
coupled with the highly endothermic decarbonation reaction. The
regeneration is weakly exothermic. The overall cycle of SEB and
regeneration is also an exothermic reaction. The temperature
required for the regeneration depends on the equilibrium of the
decarbonation of CaCO3.
1.0000
0.1000
0.0100
0.0010
0.0001
0.0000
experiences
a very fast decrease and is almost zero after
approximately 36 minutes of TOS. The weight of the dolomite
increases concurrently with TOS (Fig. 2a) even after 36 minutes. The
weight decreases after about 100 min, which is a result of the
decarbonation of the sorbent in N2 at 800 °C, as shown in Fig. 2a. It
is interesting to note that a mass of 0.17 g/g dolomite was remained
after the complete decarbonation (Fig. 2a). On an assumption of full
carbonation of CaO with the CO2 generated by the Boudouard
reaction, the weight ratio of CO2 and C is 3.66. The ratio between the
weight loss of CO2 release and the weight remained is 2.83, which is
lower than the theoretic value of 3.66 for the Boudouard reaction. It
suggests that only part of the CO2 generated by the Boudouard
reaction is captured by the carbonation reaction. The remained
material was later confirmed to be carbon by oxidation in air (Fig. 8),
400
500
600
700
Temperature [ºC]
Fig. 1. Equilibrium partial pressure of CO (▬ ▬) vs. temperature for
the Boudouard reaction, and equilibrium partial pressures of CO (▬)
and CO2 (•••) vs. temperature for the sorption enhanced Boudouard
(SEB) reaction.
Using air for regeneration not only has the advantage of low cost,
since it avoids expensive air purification for pure oxygen production,
3

Please do not adjust margins
Catalysis Science & Technology
Page 4 of 8
ARTICLE
SEM (Fig. 4a) and Raman spectroscopy analysis (Fig. 4b). This agrees experiment with only CO2 was completely regenerated, which was
View Article Online
with the observation of Mondal et al. [28] in experiments on the indicated by the complete weight loss, whileDthOeI:w10e.i1g0h3t9g/Cai9nCeYd0i0n8t5h1Ae
separation of hydrogen from syngas using solid oxides, where experiment with the CO/CO2 mixture was only partially lost, and the
significant carbon deposition due to the Boudouard reaction resulted ratio of maximum weight and the remained weight was about 2.09,
in the deactivation of the oxides surfaces as a result of an insulated again lower than the theoretic value of 3.66. These results suggest
layer of carbon. All the results reveal that sorption enhanced that the SEB reaction is more preferable than the direct CO2
Boudouard (SEB) takes place, where CO dissociates by the carbonation, although the reaction rate of SEB is lower than that of
Boudouard reaction and it is converted to CO2 and solid carbon, while direct carbonation.
CO2 formed is captured by the CaO in the dolomite, in turn enhancing
9
the Boudouard reaction.
calcination
8
7
6
5
4
3
2
1
0
calcination
14
12
10
8
30
25
20
15
10
5
calcination
6
4
2
0
100
200
300
TOS [min]
b
a
0
0
Fig. 3. Weight change, ∆w, of the dolomite sorbent vs. time on
stream (TOS) during experiments with PCO2=0.15 bar (▬) and with
PCO2=0.15 bar + PCO=0.65 bar (▬) (balance N2). Experimental
conditions: Tcarbonation=550 ºC, Tcalcination=800 ºC (PN2=1 bar),
dolomite=20 mg.
0
100
TOS [min]
200
300
0
50
100
150
TOS [min]
0.0020
0.0015
0.0010
0.0005
0.0000
By comparing Fig. 2a and Fig. 3, it can be deduced that the
presence of H2 in the gas mixture significantly enhanced the SEB
reaction. The average formation rate of carbon within 100 minutes
increased from 1.1 to 1.7 mg/gdolomite,min and the weight ratio of CO2
captured to C formed increased from 2.09 to 2.83, when hydrogen is
added. It means that the average SEB rate increased from 0.091 to
0.14 mmol/gdolomite,min.
c
Carbon formed by the SEB reaction was studied by SEM analysis
(Fig. 4a). After the SEB reaction, the solid material (sorbent and
carbon) was washed to remove CaCO3 and unconverted CaO, while
carbon was collected and analyzed by SEM. It can be seen that the
carbon formed by the SEB reaction shows very thin crystalline sheets.
Fig. 4a also shows that several graphene layers stacked with each
other. Fig. 4b shows the Raman spectra of the carbon produced from
the SEB reaction. The obtained Raman spectra is identical to that
reported in the literature for graphite material [29]. The G and D
peaks of Raman spectra lie around 1560 cm-1 and 1360 cm-1,
respectively. This indicates that the carbon formed by the SEB
reaction is highly graphitic.
These results clearly demonstrated that the SEB reaction takes
place in the absence of a conventional metal catalyst. It is therefore
suggested that CaO present in the dolomite produces a catalytic
effect over the Boudouard reaction under the studied conditions.
Previous studies in the literature have reported that calcined
dolomite can act as both CO2 sorbent and catalyst on sorption
enhanced water gas shift experiments and that no addition of
conventionally employed catalysts (metals, metal oxides) is required
[30, 31]. In the present work, CaO in calcined dolomite could
therefore be responsible for the activity towards the Boudouard
reaction, which would mean that calcined dolomite (CaO) may act as
0
50
100
150
TOS [min]
Fig. 2. Weight change, ∆w, of the dolomite sorbent (▬) (a), outlet CO
(▬) and CO2 (▬) concentrations (b), and reaction rate of CO (▬) and
CO2 (▬) (c) vs. time on stream (TOS) during an experiment with
+ PCO=0.65 bar + PH2=0.12 bar (balance N2).
Experimental conditions: Tcarbonation=550 °C, Tcalcination=800 °C
(PN2=1 bar), dolomite=20 mg.
PCO2=0.15 bar
The results might suggest competitive reactions on CaO surface
between the direct carbonation of CO2 and the Boudouard reaction
followed by the carbonation reaction. The surface Boudouard
reaction possibly suppresses the direct carbonation of the CO2 in the
gas phase. To confirm this hypothesis, Fig. 3 shows a comparison of
the weight gains during experiments with CO2/N2 and CO2/CO/N2 gas
mixtures using identical CO and CO2 pressures as in the previous
experiment. The initial increase in the weight followed almost an
identical rate for both mixtures. Taking into account the results in Fig.
2b, showing a fast increase in the CO2 concentration, it can be
suggested that direct CO2 carbonation takes place initially. Thus, the
rate of weight increase was slower for the CO/CO2 mixture than in
the experiment with only CO2. A more significant difference was
observed in the regeneration step in N2. The CaCO3 formed in the
4
Please do not adjust margins

Page 5 of 8
Catalysis Science & Technology
both catalyst and CO2 sorbent in the SEB process. It could be argued all the experimental time of about 12 hours. The weight of the
that Fe and other metals in dolomites can serve as catalysts for the dolomite increased concurrently with time on stream. The detected
View Article Online
Boudouard reaction. In order to check the possible effect of the increase in the weight accounts well for the captured CO2 and the
DOI: 10.1039/C9CY00851A
metal impurities contained in dolomites on the SEB, a reference run deposited carbon generated by the SEB reaction based on the mass
was performed on a synthetic CaO material with a purity of 99.99% balance. The results clearly demonstrate the SEB reaction as a new
using a gas mixture of CO and H2 at 550 °C, as shown in Fig. 5. It shows effective route to produce fuel-cell grade H2 without loss of
that weight is gained with an almost constant rate of 7.1 hydrogen.
mg/gCaO,min. This confirms that the CaO present in the dolomite can
effectively be responsible for the activity towards the SEB reaction.
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
6
5
4
3
2
1
0
a
a
b
0
5
10
0
5
10
TOS [hour]
TOS [hour]
Fig. 6. Weight change, ∆w, of the dolomite sorbent (a) and outlet CO
concentration (b) vs. time on stream (TOS) during an experiment
with PCO=0.005 bar + PH2=0.995 bar. Experimental conditions:
Tcarbonation=420 °C, dolomite=3 mg.
b
The effect of the temperature on the SEB reaction in a CO/H2 gas
mixture is shown in Fig. 7. The weight change of the dolomite at 450
and 550 °C as a function of the time of experiment is shown in Fig.
7a, whereas the outlet gas concentrations for the experiment at
550 °C as a function of the time of experiment are shown in Fig. 7b.
1000
1500
2000
2500
3000
Raman Shift [cm-1
]
Fig. 4. Structural characterization of the carbon produced by the SEB
reaction: SEM image (a) and Raman spectra (b).
12
10
8
20
18
16
14
12
10
8
11
10
9
6
8
7
4
6
6
4
a
b
2
5
2
4
0
0
3
0
50
100
TOS [min]
150
200
0
50
100
TOS [min]
150
200
2
1
Fig. 7. (a) Weight change, ∆w, of the dolomite sorbent during
carbonation experiments at 450 °C (▬) and 550 °C (▬) with PCO=0.25
bar + PH2=0.75 bar; (b) Outlet CO (▬) and CO2 (▬) concentrations vs.
time on stream (TOS) during the carbonation experiment at 550 °C
0
0
100
TOS [min]
200
Fig. 5. Weight change, ∆w, of the CaO-based synthetic sorbent vs.
time on stream (TOS) during an experiment with PCO=0.25 bar +
PH2=0.75 bar. Experimental conditions: Tcarbonation=550 °C, synthetic
sorbent=10 mg.
with PCO=0.25 bar
dolomite=10 mg.
+ PH2=0.75 bar. Experimental conditions:
The dolomite experiences a much higher increase in its weight
when the experiment is performed at 550 °C (Fig. 7a), which means
that a higher amount of CO2 is captured by the sorbent and a higher
deposition of carbon is taking place, i.e. the reaction rate of the SEB
reaction increases with the increasing in temperature. Despite the
favorable thermodynamics of the carbonation reaction at low
temperature, which favors the SEB reaction, the kinetics of the
process at low temperature is slower. Fig. 7b shows that the outlet
CO2 concentration increases during the time of experiment as the
4.2. Selective removal of CO for pure hydrogen production
Selective removal of CO from the CO/H2 mixture (0.5/99.5 molar
ratio) was carried out on dolomite in the TGA-fixed bed reactor at
420 °C. Fig. 6a shows the weight change of the dolomite, while Fig.
6b shows the outlet CO concentration as a function of the time on
stream. A very low CO concentration (<5 ppm) was detected during
5

Please do not adjust margins
Catalysis Science & Technology
Page 6 of 8
ARTICLE
2-
sorbent is being saturated and loses capacity for the removal of CO2. formation of CO3 complexes is unclear. DFT calculations [34, 35]
View Article Online
The increase in the CO2 concentration is the result of its formation predicted that these CO molecules are adsorDbeOdI:v1i0a.1t0h3e9i/rCC9aCtYo0m08s5o1An
via the SEB reaction. Likewise, the outlet CO concentration increases Ca sites, but rather weakly. On the polycrystalline CaO surfaces, no
with time (Fig. 7b) and it is lower than the inlet CO concentration significant contributions from molecularly adsorbed CO were found
along the experiment.
in MIES or UPS spectra [32]. However, a UV-Visible and IR
Cyclic experiments were carried out with a CO/H2 gas mixture on spectroscopic study of the adsorption of CO on microcrystalline MgO
dolomite at 550 °C during the carbonation stage and at 800 °C in N2 and CaO showed two types of adsorbed species, such as lower order
during the regeneration step, which is followed by a combustion step non-reduced species incorporating surface oxide ions with
in air in order to eliminate the deposited carbon. The cycle stability Ca2+…(CO2)2- stoichiometry, and reduced polymeric species like cyclic
2-
of the CO2 sorbent for the SEB reaction was tested in four anion species (CO)n [36]. The formation of (CO3)2- was then
carbonation/decarbonation cycles. Fig. 8 shows the weight change of proposed like:
the dolomite as a function of the time along the cycles. The results
CO + O2- = (CO2)2- + VO
(6)
(7)
(8)
remain quite constant during the four cycles, which indicates no
significant decrease in the CO2 capture capacity of the sorbent.
Captured CO2 was calculated to be 4.6 mg /10 g dolomite in average
of all the cycles. The weight ratio of CO2 to C is almost constant with
a value of about 0.75 in all the cycles, which is much lower than the
stoichiometric value of 3.66 for the Boudouard reaction.
(CO2)2- + nCO = (Cn+1On+2
)
2-
(Cn+1On+2)
2- + O2- = (CnOn)2- + (CO3)2- + VO
where VO is the oxygen vacancy. The proposed mechanism involved
the formation of surface species (CO2)2- and the formation of reduced
polymeric species such as (Cn+1On+2)2-. By the intervention of a further
surface oxygen ion, ring closure occurs to give a resonance-stabilized
polymeric carbanion (CnOn)2- and a carbonate species is rejected in
the surface. This happened predominantly at n=5 on CaO [36].
Atomic-scale simulation, by means of density functional calculations
and molecular dynamic simulation, has also been applied to
understand the mechanism of carbonation of CaO. It pointed out O
and C vacancies as thermally important defects, together with
interstitial O. The joint C and O diffusion via vacancy mechanisms is
important for CO2 diffusion within CaCO3 and CaCO3 growth.
Formation and diffusion of (CO3)2- in the lattice of CaO generates the
surface oxygen vacancy. The in-situ generated O vacancy seems to
facilitate the CO dissociation, CO → C + O, and then O fills the surface
oxygen vacancy, that reacts further with an undissociated CO(ads) to
form surface CO2(ads) [37, 38]. Based on above discussion, a reaction
mechanism of SEB is proposed in the scheme in Fig. 9:
After the calcination step, the remaining weight change is 6.8 mg
in average, which means 0.68 gcarbon/gfresh sorbent. This amount of
carbon is higher than that shown in Fig. 2a, but in the cyclic
experiment no CO2 had been added in the gas mixture and the
concentration of H2 is quite higher.
16
1st cycle
2nd cycle
3rd cycle
4th cycle
14
12
10
8
calc.
calc.
calc.
calc.
6
4
2
0
0
200
400
600
TOS [min]
800
1000
1200
Fig. 8. Weight change, ∆w, of the dolomite sorbent vs. time on
stream (TOS) for several cycles during an experiment with PCO=0.25
bar + PH2=0.75 bar. Experimental conditions: Tcarbonation=550 °C,
Tcalcination=800 °C (PN2=1 bar or Pair=1 bar), dolomite=10 mg.
VO
4.3. On the reaction mechanism
Surface science and density functional theory (DFT) calculation on CO
and CO2 interaction with CaO have been reported in the literature
and they provide mechanistic insights [32]. CO2 and CO interaction
with CaO surfaces has been studied by surface science techniques
such as Metastable Induced Electron Spectroscopy (MIES),
Ultraviolet Photoelectron Spectroscopy (UPS) and X-ray
Photoelectron Spectroscopy (XPS) on CaO films [32]. Since CaO has
not density of states near the Fermi level, no CO2 and CO dissociation
Fig. 9. The proposed reaction mechanism of the SEB reaction.
The main feature of the mechanism is the oxygen vacancy as the
active site for the SEB reaction, which is generated in-situ by CO
adsorption (Fig. 9). Moreover, the results in the present work have
clearly revealed that hydrogen enhanced the SEB reaction. It could
be speculated a mechanism of hydrogen assisted CO activation on
the oxygen vacancy, and the detailed mechanism needs to be further
studied.
2-
is possible. The CO2 molecules impinge on the surface and form CO3
complexes with surface oxygen atoms. According to Kadossov et al.
[33], the interaction between CO molecules and CaO(100) single
crystal surfaces is very weak. Non-dissociative CO adsorption is found
only at irregular sites, e.g. edges [33]. However, the formation of
CO32- complexes was clearly observed by MIES and XPS when the CaO
surface was exposed to CO, although the mechanism for the
6
Please do not adjust margins

Page 7 of 8
Catalysis Science & Technology
[10] C. Galletti, S. Specchia, V. Specchia, CO selective methanation in
5. Conclusions
H2-rich gas for fuel cell application: Microchannel reactor
The proposed SEB process integrates the Boudouard reaction with in
situ CO2 removal in a single step on a CaO based material, which not
only catalyzes the Boudouard reaction but also removes the
generated CO2 in situ. The direct tandem reaction on CaO by the
formation of CaCO3 shifts the Boudouard reaction to a forward
direction following the Le Chatelier’s principle, thus enhancing the
CO conversion. Here we demonstrated the feasibility of the removal
of CO without consumption of hydrogen by the SEB reaction on
dolomite in the absence of conventional noble metal catalysts,
achieving a CO concentration lower than 5 ppm. The presence of
hydrogen enhanced the SEB reaction. This new process can replace
the hydrogen purification processes such as methanation, which
consumes hydrogen, or selective oxidation of CO, which requires
pure oxygen and also consumes a part of the hydrogen. In addition,
the SEB reaction combines the purification process with the CO2
removal process, such as pressure swing process, in one-step.
Moreover, the process allows the production of pure hydrogen with
ultra-low concentrations of CO and CO2. The low temperature PEM
fuel cell can benefit from such ultra-pure hydrogen. An oxygen
vacancy mediated catalytic cycle of Boudouard reaction on metal
oxide is proposed, including CO dissociation and CO reaction with
lattice oxygen. It is expected that the chemistry of CO reaction on
CaO can be also applied to other metal oxides as well as to process
beyond CO removal, such as CO hydrogenation on metal oxides.
View Article Online
performance with Ru-based catalysts, Chemical Engineering Journal,
DOI: 10.1039/C9CY00851A
167 (2011) 616-621.
[11] H. Yoshida, K. Watanabe, N. Iwasa, S.-i. Fujita, M. Arai, Selective
methanation of CO in H2-rich gas stream by synthetic nickel-
containing smectite based catalysts, Applied Catalysis B:
Environmental, 162 (2015) 93-97.
[12] K. Hayashi, T. Miyao, Y. Tabira, K. Higashiyama, Low-
temperature Selective CO Methanation over Unsupported Nickel
Catalyst Covered by Silica Thin Layer, Journal of the Japan Petroleum
Institute, 59 (2016) 65-71.
[13] T. Noor, M.V. Gil, D. Chen, Production of fuel-cell grade
hydrogen by sorption enhanced water gas shift reaction using Pd/Ni–
Co catalysts, Applied Catalysis B: Environmental, 150–151 (2014)
585-595.
[14] C.R. Müller, R. Pacciani, C.D. Bohn, S.A. Scott, J.S. Dennis,
Investigation of the Enhanced Water Gas Shift Reaction Using
Natural and Synthetic Sorbents for the Capture of CO2, Industrial and
Engineering Chemistry Research, 48 (2009) 10284-10291.
[15] L. He, D. Chen, Single-Stage Production of Highly Concentrated
Hydrogen from Biomass-Derived Syngas, ChemSusChem, 3 (2010)
1169-1171.
[16] R.W. Stevens Jr, A. Shamsi, S. Carpenter, R. Siriwardane,
Sorption-enhanced water gas shift reaction by sodium-promoted
calcium oxides, Fuel, 89 (2010) 1280-1286.
[17] B. Li, G. He, X. Jiang, Y. Dai, X. Ruan, Pressure swing
adsorption/membrane hybrid processes for hydrogen purification
with a high recovery, Frontiers of Chemical Science and Engineering,
10 (2016) 255-264.
[18] E.H. Majlan, W.R. Wan Daud, S.E. Iyuke, A.B. Mohamad, A.A.H.
Kadhum, A.W. Mohammad, M.S. Takriff, N. Bahaman, Hydrogen
purification using compact pressure swing adsorption system for fuel
cell, International Journal of Hydrogen Energy, 34 (2009) 2771-2777.
[19] B. Dou, C. Wang, H. Chen, Y. Song, B. Xie, Continuous sorption-
enhanced steam reforming of glycerol to high-purity hydrogen
production, International Journal of Hydrogen Energy, 38 (2013)
11902-11909.
Acknowledgements
This work was carried out with financial support from the iCSI center
for Research-based Innovation.
Notes and references
[1] B. Balasubramanian, A. Lopez Ortiz, S. Kaytakoglu, D.P. Harrison,
[20] L. Fan, F. Li, S. Ramkumar, Utilization of chemical looping
strategy in coal gasification processes, Particuology, 6 (2008) 131-
142.
[21] S. Ramkumar, L.-S. Fan, Calcium Looping Process (CLP) for
Enhanced Noncatalytic Hydrogen Production with Integrated Carbon
Dioxide Capture, Energy & Fuels, 24 (2010) 4408-4418.
[22] L. He, H. Berntsen, D. Chen, Approaching Sustainable H2
Production: Sorption Enhanced Steam Reforming of Ethanol, Journal
of Physical Chemistry A, 114 (2010) 3834-3844.
Hydrogen from methane in
Engineering Science, 54 (1999) 3543-3552.
[2] R.B. Gupta, Hydrogen fuel : production, transport, and storage
Boca Raton, FL : CRC Press (2009).
[3] P. Häussinger, R. Lohmüller, A.M. Watson, Hydrogen, Ullmann’s
Encyclopedia of Industrial Chemistry, (2000).
[4] A.F. Ghenciu, Review of fuel processing catalysts for hydrogen
production in PEM fuel cell systems, Current Opinion in Solid State
and Materials Science, 6 (2002) 389-399.
a single-step process, Chemical
[23] O. Boudouard, C.R. Acad. Sci, Paris 128, 824 and 1521, (1899).
[24] P. Lahijani, Z.A. Zainal, M. Mohammadi, A.R. Mohamed,
Conversion of the greenhouse gas CO2 to the fuel gas CO via the
Boudouard reaction: A review, Renewable and Sustainable Energy
Reviews, 41 (2015) 615-632.
[25] E.T. Turkdogan, J.V. Vinters, Catalytic effect of iron on
decomposition of carbon monoxide: I. carbon deposition in H2-CO
Mixtures, MT, 5 (1974) 11-19.
[26] T. Wiltowski, K. Mondal, A. Campen, D. Dasgupta, A. Konieczny,
Reaction swing approach for hydrogen production from
carbonaceous fuels, International Journal of Hydrogen Energy, 33
(2008) 293-302.
[5] T. Choudhary, Goodman, DW., CO-free fuel processing for fuel cell
applications, Catalysis Today, 77 (2002) 65-78.
[6] A. Appleby, Foulkes, FR., Fuel Cell Handbook, Van Nostrand
Reinhold, New York,, (1989).
[7] N.A. Koryabkina, A.A. Phatak, W.F. Ruettinger, R.J. Farrauto, F.H.
Ribeiro, Determination of kinetic parameters for the water–gas shift
reaction on copper catalysts under realistic conditions for fuel cell
applications, Journal of Catalysis, 217 (2003) 233-239.
[8] J. Saavedra, T. Whittaker, Z. Chen, C.J. Pursell, R.M. Rioux, B.D.
Chandler, Controlling activity and selectivity using water in the Au-
catalysed preferential oxidation of CO in H2, Nat Chem, 8 (2016) 584-
589.
[27] A. Campen, K. Mondal, T. Wiltowski, Separation of hydrogen
from syngas using a regenerative system, International Journal of
Hydrogen Energy, 33 (2008) 332-339.
[9] E. Newson, T.B. Truong, N. De Silva, A. Fleury, R. Ijpelaar, 85 High
selectivity preferential oxidation (PROX) catalysts for CO removal
from hydrocarbon derived reformates for PEM fuel cells, in: M.O.
Masakazu Anpo, Y. Hiromi (Eds.) Studies in Surface Science and
Catalysis, Elsevier, 2003, pp. 391-394.
[28] K. Mondal, K. Piotrowski, D. Dasgupta, E. Hippo, T. Wiltowski,
7

Please do not adjust margins
Catalysis Science & Technology
Page 8 of 8
ARTICLE
Hydrogen from Coal in a Single Step, Industrial & Engineering
Chemistry Research, 44 (2005) 5508-5517.
View Article Online
DOI: 10.1039/C9CY00851A
[29] A.C. Ferrari, Raman spectroscopy of graphene and graphite:
Disorder, electron–phonon coupling, doping and nonadiabatic
effects, Solid State Communications, 143 (2007) 47-57.
[30] M.A. Escobedo Bretado, M.D. Delgado Vigil, J.S. Gutiérrez, A.
López Ortiz, V. Collins-Martínez, Hydrogen production by absorption
enhanced water gas shift (AEWGS), International Journal of
Hydrogen Energy, 35 (2010) 12083-12090.
[31] C. Han, D.P. Harrison, Simultaneous shift reaction and carbon
dioxide separation for the direct production of hydrogen, Chemical
Engineering Science, 49 (1994) 5875-5883.
[32] F. Voigts, F. Bebensee, S. Dahle, K. Volgmann, W. Maus-
Friedrichs, The adsorption of CO2 and CO on Ca and CaO films studied
with MIES, UPS and XPS, Surface Science, 603 (2009) 40-49.
[33] E. Kadossov, U. Burghaus, Adsorption kinetics and dynamics of
CO, NO, and CO2 on reduced CaO(100), Journal of Physical Chemistry
C, 112 (2008) 7390-7400.
[34] W.S.A. Halim, CO adsorption on MgO, CaO and SrO crystals
periodic Hartree-Fock calculations, Applied Surface Science, 253
(2007) 8974-8980.
[35] M. Bajdich, J.K. Norskov, A. Vojvodic, Surface energetics of
alkaline-earth metal oxides: Trends in stability and adsorption of
small molecules, Physical Review B, 91 (2015).
[36] E. Garrone, A. Zecchina, F.S. Stone, CO ADSORPTION ON MGO
AND CAO - SPECTROSCOPIC INVESTIGATIONS OF STAGES PRIOR TO
CYCLIC ANION CLUSTER FORMATION, Journal of the Chemical
Society-Faraday Transactions I, 84 (1988) 2843-2854.
[37] A.K. Kandalam, B. Chatterjee, S.N. Khanna, B.K. Rao, P. Jena, B.V.
Reddy, Oxidation of CO on Fe2O3 model surfaces, Surface Science,
601 (2007) 4873-4880.
[38] G. Lanzani, A.G. Nasibulin, K. Laasonen, E.I. Kauppinen, CO
dissociation and CO+O reactions on a nanosized iron cluster, Nano
Research, 2 (2009) 660-670.
8
Please do not adjust margins