Synergistic Effect of MgH2 Doping With Ni and Carbon N 2022 International Jo
Synergistic Effect of MgH2 Doping With Ni and Carbon N 2022 International Jo
Synergistic Effect of MgH2 Doping With Ni and Carbon N 2022 International Jo
ScienceDirect
Article history: The synergistic effect of the nickel and carbon nanotubes addition on thermochemical
Received 10 August 2021 processes for CO2 recycling employing MgH2 as hydrogen source was demonstrated. By
Received in revised form reducing the metallic charge from 10wt.% Ni, replacing 5wt.% with CNTs, the amount of
13 September 2021 produced methane through thermal treatment at ~350e375 C under CO2 increased more
Accepted 1 October 2021 than 30%. A COx-free mixture of methane (70.9%) and hydrogen (29.1%) with a CH4 yield of
Available online 20 October 2021 79% was obtained by reaction at 350 C during 48 h. Ni-containing species act as selective
catalyst promoting the occurrence of the Sabatier reaction, which competes with the direct
Keywords: reduction of CO2 to generate CH4 via C as an intermediary. The role of CNTs is not catalytic
Hydrides but rather it could be related to the protection of the hydride. It is proposed that when
Carbon dioxide
* Corresponding author.
E-mail address: guillerminaamica@gmail.com (G. Amica).
https://doi.org/10.1016/j.ijhydene.2021.10.007
0360-3199/© 2021 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
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 4 7 ( 2 0 2 2 ) 4 2 8 e4 4 2 429
Hydrogen-methane mixtures MgH2 is neighbored by graphitic planes, its oxidation due to direct contact with CO2 and/or
Carbon nanotubes the generated H2O could be impeded or retarded.
© 2021 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
conditions upon 500 C to generate potassium for- To the best of our knowledge, there are no other previous
mylhydroborates, hydrogen, methanol, carbon monoxide and investigations involving catalyzed-hydrides systems for CO2
a small amount of methane [34]. Magnesium based-metal hy- conversion. Our current effort is orientated to the study of
drides containing Ni or Fe can be considered suitable dual hydrides-based systems that act as a source of hydrogen, which
conversion promoters and hydrogen sources for CO2 reduction in the presence of an active catalytic phase and under suitable
because the segregation of metallic nanoparticles increases experimental conditions can generate significant amounts of
the catalytic activity of the system [35,36]. It was postulated methane or CO2-free methane-hydrogen mixtures, through
that the reaction rate of the CO2 hydrogenation process on the processes with low operating temperature and fast reaction ki-
surface of Mg2NiH4 is determined by the dissociative adsorp- netics. The use of a solid matrix allows the safe storage of H2
tion of CO2 molecules instead of the dissociation of H2 mole- produced from renewable energies. As hydride-forming mate-
cules on the hydride [35]. Total conversion of CO2 was reached rials can be employed to transform in-situ a polluting gaseous
after heating Mg2FeH6 and Mg2NiH4 under CO2 with a 4H2:CO2 waste into products with added value, through this approach,
molar ratio at 400 C during 5 and 10 h, respectively. For the the H2 cycle can be integrated into that of CO2. In this work we
former, due to the catalytic action of Fe particles, the process analyze the feasibility of employing nickel-catalyzed MgH2 with
involves water-gas shift reaction followed by the methanation the incorporation of multi-walled carbon nanotubes for efficient
of CO in the presence of steam [36]. For the later, in addition to thermochemical CO2 conversion processes. Nickel and carbon
the catalytic reaction, the CO2 direct reduction has a substan- nanotubes synergy is demonstrated. The metal acts as an active
tial contribution due the poisoning of Ni-based compounds by catalyst to promote the Sabatier process, whereas it was sug-
CO during the early step of the reaction, hindering its further gested that CNTs protect the hydride from being oxidized by
catalytic action [36]. Methanation via-Mg2NiH4 synthesized by preventing its direct contact with the CO2 and/or H2O. Through
in situ monitoring mechanical milling under hydrogen atmo- reaction between Ni/CNTs edoped MgH2 and CO2 at 350 C
sphere from a 2MgH2:Ni stoichiometric mixture [37] was during 48 h, a gas phase free of CO2 or CO, composed of CH4
proved to be faster than via-Mg2NiH4 prepared by a two-step (70.9%) and H2 (29.1%) can be obtained, with a calculated a
procedure involving milling and sintering at high tempera- methane yield of 79%.
ture [36]. This demonstrates that the complex hydride syn-
thesis method has big influence on the catalyst-support
interaction, affecting the methanation global mechanism [37]. Experimental
Alkali metal hydrides (LiH and NaH) or alkaline-earth metal
hydrides (CaH2 and MgH2) in the absence of a catalyst have Material preparation
shown their best performance for thermochemical CO2
reduction by heat treatment at 450 C for 48 h through a pro- For the experiments, magnesium hydride (MgH2, 98%),
cess considering elemental carbon as an intermediate for the metallic nickel (Ni, 99.99%) and multi-walled carbon nano-
formation of methane [38,39]. Moreover, it was demonstrated tubes (CNT, <95%) were purchased at Sigma-Aldrich. Due to
that COx-free H2-mixed CH4 fuels can be obtained thought the high reactivity of the samples, they were handled in an
thermochemical reaction between inorganic carbonates and MBraun Unilab argon filled glove box, with oxygen and
activated metal hydrides, as sources of CO2 and H2, with high moisture levels lower than 10 ppm. For all studies, high purity
yields and selectivity. According to the experimental tuning, argon (Linde, 99.999%) was used. Ball milling in a planetary
CH4 content can range from 8.6% to 88.1% [40]. Our previous mill (Fritsch Pulverisette 6) was employed to prepare four
investigation on the un-catalyzed reaction between MgH2 and samples: (i) MgH2 and (ii) MgH2e10wt.% Ni, (iii) MgH2-5wt.%Ni-
CO2, showed that a methane yield of 44.6% was obtained after 5wt.%CNT and (iv) MgH2-5wt.%CNT. All samples were ball
24 h of thermal treatment at 400 C employing a molar ratio milled for 5 h at 500 rpm with a ball to powder mass ratio of
MgH2:CO2 of 2:1, through a methanation mechanism that 53:1 using a sequence of 10 min milling and a 15 min pause.
involved the direct reduction of CO2 and the generation of CH4
via C as an intermediary [41]. When either the temperature or Characterization of samples
the reaction time was lowered, the yield values dropped
sharply due to partial reduction of CO2 to CO. Moreover, it was The reaction between the samples and CO2 under different
proved that the presence of a catalyst and its nature allows the experimental conditions was studied under static and dy-
methanation to follow another path of reaction, a mechanism namic conditions. Static measurements were performed in a
that responds to a Sabatier process. Through the incorporation stainless steel reactor coupled to Sieverts volumetric equip-
of an active catalyst, by heat treatment at 350 C for 48 h ment that allows the selection of the pressure and tempera-
considering a molar ratio MgH2:CO2 of 4:1, MgH2 doped with ture. A specific mass of each sample was introduced in the
10 wt% Co achieved a methane yield of 78% and the resulting reactor inside the glove box to avoid contamination with ox-
gas mixture was composed of CH4 (49.7%), CO2 (0.5%), CO ygen or air humidity and then heated up with a ramp of 10 C
(<1000 ppm) and the rest of H2 [41]. The calculated methane min1 to different temperatures (300, 350, 375 and 400 C)
yield was not only considerably higher than the one previously under CO2 pressure during different reaction times (6, 12, 24
reported at 350 C (17%) but, interestingly, it was also higher and 48 h). In all cases a molar ratio MgH2:CO2 of 4:1 was
than that reported at 450 C (68%) for un-catalyzed MgH2 [39]. considered. Under dynamic conditions, the samples were
Then, the use of cobalt allowed lowering the operational heated at 5 C min1 up to 450 C in a continuous flow of CO2,
temperature without affecting the methane yield [41]. using a TG Analyzer (TG-HP50, TA Instruments). Additionally,
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 4 7 ( 2 0 2 2 ) 4 2 8 e4 4 2 431
the experiment was carried out using a flow of He to analyze crystallite size reduction. The chosen milling time (5 h) was
the decomposition of the sample in an inert atmosphere. based on preliminary results, which indicated that during the
The composition of the gases after thermal treatment first 2 h only a reduction in particle size occurs without any
under CO2 was studied through gas-phase FTIR analyses real mixing of the components [47]. Between 2 and 4 h Ni is
(FTIR, PerkinElmer Spectrum 400 with an MCT detector) and dispersed in the material, allowing Ni to be distributed on the
gas chromatography (GC, Agilent Technologies GC System surface of the MgH2 or within its matrix, and thus can play its
7820A). For the GC experiments 1 ml of the released gas was role as a catalyst. After 4 h an effective mixing of the com-
taken using a syringe, which was injected into the GC equip- ponents is produced accompanied by microstructure refine-
ment. The molar amounts of gaseous CO2, CH4 and CO were ment due to additional reduction of particle sizes [47]. For
quantified using calibration curves. The methane yield was these reasons and to minimize possible contamination with
calculated as the ratio between the produced CH4 moles and Fe from the milling jar, a duration of 5 h was selected. SEM
the total CO2 moles at the beginning of the reaction. A nu- images of the sample after ball milling shows agglomerate
merical example is available in the Supplementary Material size distribution between 5 and 50 mm (Fig. 1A). The sample
(see Methane yield calculation S1). For IR spectroscopy mea- was characterized by its spongy shape, with some central
surements, the gases released were collected in a degassed solid blocks and a large quantity of small agglomerates
quartz optical cell with KBr windows and an optical length of adhering to its surface (Fig. 1B). Element mapping obtained by
9 cm. The gas phase spectra at room temperature were taken EDXS indicated a homogeneous distribution of Ni (Fig. S2). To
considering transmittance mode with a resolution of 0.5 cm1 investigate the conversion of CO2 in the presence of MgH2
and 10 scans. During each spectrum acquisition, part of the with catalyst addition, the as-milled Ni-doped sample was
optical path was in air atmosphere, so a baseline measure- heated up to different temperatures in static conditions
ment was required prior to each measurement. The structural under a CO2 atmosphere during a period of 24 h, considering a
characterization of the solid samples was studied using X-ray molar relation MgH2:CO2 of 4:1 (see Table 1). Thought
powder diffraction (XRPD, Bruker D8 Advance) and Fourier elemental mapping it was verified that after the sample was
transform infrared spectroscopy (FTIR, PerkinElmer Spectrum subjected to thermal treatment at 375 C under CO2 during
400 with an MCT detector). An airtight holder (which has a 24 h, there was no migration or agglomeration of Ni (Fig. S3).
bump in the zone of 2q ¼ 15e25 ) was employed during the A wider size distribution with agglomerates of up to 200 mm
XRPD data collection to prevent any reaction between the (Fig. 1C) and fewer exposed flat faces (Fig. 1D) were observed.
samples and air. For IR spectroscopy measurements, the The gas-state products of the reaction between MgH2 and
samples were ground with dry KBr under a purified argon CO2 were identified by FTIR (Fig. 2) and quantified by GC (Table
atmosphere, pressed into pellets and placed in a specially 1). Clear H2 and CH4 chromatography peaks were detected at
designed airtight cell. Solid state IR spectra were obtained in all temperatures. A minor excess of CO2 and co-generation of
the range of 4000e800 cm1 with a resolution of 4 cm1. a minor percent of CO were also detected in some cases. A
Morphological analysis of the samples was performed by direct relationship was observed between the operating tem-
scanning electron microscopy (SEM, SEM-FIB, Zeiss, Cross- perature and both the molar percentage and the yield of CH4.
beam 340), employing dispersed powders on carbon tape. As it can be observed in Table 1, when temperature was
Elemental analyses of the materials were also performed by reduced from 400 to 350 C, these values decreased from 50.2
energy-dispersive X-ray spectroscopy (EDXS) on the SEM. to 35.2 and from 64.5 to 44.9, respectively. At this temperature
range, gaseous mixtures composed mainly of CH4 and H2 were
Thermodynamic calculations obtained and the remaining amount of CO2 in the gas
remained below 2%. However, when the temperature was
The equilibrium compositions for the Ni catalyzed MgH2eCO2 lowered, this mixture became enriched in H2. As the amount
system as a function of temperature and total system pressure of CO2 remained practically unchanged, the lower content of
were calculated using HSC Chemistry Windows [46]. The CH4 suggests the formation of some other carbon-containing
program calculates the combination of the most stable species species in the solid phase [41]. Moreover, thermal treatment
and the phase composition to minimize the Gibbs free energy at 300 C was not enough to achieve a considerable rate of CO2
of the system. The calculation was conducted based on the conversion. In this condition the gas phase was mainly
gas phase containing CO, CH4, C2H6, CH3OH, H2O, H2, O2 and composed of unreacted CO2. The presence of CO was
CO2 and the solid phase containing MgCO3, MgO, Mg, MgH2, confirmed by FTIR analyses at 300 and 350 C, unlike obser-
Mg(OH)2, Ni, NiO, Ni (FCC), Ni(OH)2, Ni3C, Mg2Ni, MgNi2, C vations at higher temperatures. Regarding the operation time,
(graphite) and C (amorphous). it was observed that at 350 C the increase from 24 to 48 h of
thermal treatment allowed a large growth in the amount of
methane in the gas phase, which raised from 35.2 to 67.7% and
Results in the methane yield from 44.9 to 74.2%. The difference cannot
be explained by the variation of CO2 in the gas phase, but
CO2 conversion employing Ni-doped MgH2 under static rather by the solid gas reaction of H2 with some component of
conditions the solid phase. Gas-state FTIR was useful to identify the
presence of water in the gas phase in all the range of tem-
A sample composed of magnesium hydride with 10wt.% peratures (Fig. 2).
nickel was prepared through mechanical ball milling as The evolution of the phases in the solid product due to the
reactivity increase is expected with agglomerate and/or interaction with CO2 was studied by XRPD (Fig. 3) and FTIR
432 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 4 7 ( 2 0 2 2 ) 4 2 8 e4 4 2
Fig. 1 e SEM images of secondary electrons of the MgH2e10wt.% Ni sample as-milled (A and B) and after reaction with CO2 at
375 C during 24 h (C and D).
Fig. 4 e SEM image of secondary electrons and Mg, Ni mapping of the MgH2-5wt.%Ni-5wt.CNT sample: as-milled (A) and
after thermal treatment under CO2 at 375 C during 24 h (B).
peaks were narrower when the interference from the dome 1260 and 1130 cm1) [49]. As for the previous sample, a weak
was absent, it was possible to identify correctly all the phases band located at 1150 cm1 could not be assigned to carbon-
safely employing the airtight holder without exposing the ates. In addition, the resulting solid phase of samples after
solid sample to air. Moreover, solid-state FTIR analyses thermal treatments at 400 C under CO2 with different times
(Fig. S7) showed that of the presence of MgH2 (bands located at were analyzed (Fig. S8). XRPD spectra for 12 and 24 h were
1246 and 658 cm1) could not be ruled out. Moreover, biden- similar. At this temperature there was no signal of Mg2NiH4
tate superficial carbonates on MgO were identified (1670, 1590, formation, not even at low reaction times. The pattern for 6 h
did not show different phases. The metallic magnesium main
peak located at 2q ¼ 36.62 was considerably higher than for
the other cases and the relative intensity of this peak with
respect to the magnesium hydride main peak located at
2q ¼ 27.9 was also higher.
Discussion
Fig. 9 e Molar fraction of CH4 and H2 (A, B) and methane yield (C) after reaction with CO2 during 24 h at different
temperatures for the MgH2-10wt.%Ni and MgH2-5wt.%Ni-5wt.%CNT samples.
widely studied [36,37,41]. In our previous work, by heat storage material with a high Ni content, for example Mg2NiH4
treatment at 400 C for 24 h, the MgH2 system with 10% co- solid matrix for reducing CO2, is not beneficial since the CO2
balt achieved a methane yield of 60.9% and the resulting gas adsorption and disproportion of the surface that occurs
mixture was composed of CH4 (38.2%), CO2 (0.8%), CO during its decomposition leads to the formation of MgH2,
(<1000 ppm) and the rest of H2 [41]. Under the same experi- MgNi2, MgO and carbonates and Ni-based compounds are
mental conditions, the nickel-doped MgH2 exhibited better poisoned with CO during the early step of the reaction,
performance (methane yield of 64.5% and molar percentage limiting the production of CH4 [36].
of CH4 of 50.2%). Even by heat treatment at 350 C for 24 h the In this work, the performance of nickel-catalyzed MgH2
results were better (methane yield of 59.7% with Co and with the incorporation of CNTs has been analyzed. Previously
61.5% with Ni). Moreover, it was proved that the length of the we have demonstrated that the simultaneous addition of
heat treatment influences the performance of the system. At metal (in particular cobalt) and CNTs to MgH2 had a positive
350 C, when increasing the time from 24 to 48 h, the effect on the hydrogen sorption properties compared to each
methane yield rose to 74.2%. This value is similar to that additive individually [51]. The improvements in the kinetics
previously obtained with Co under the same conditions were not attributed to CNTs but to metal. It was explained
(78%), but with a higher percentage of methane (67.7 vs that the presence of a transition metal allowed the dissocia-
49.7%). This shows that the nature of the catalyst has an tion of the H2 molecule at lower temperatures and when CNTs
important function. Theoretical studies have shown that are well distributed throughout the sample, they can play a
both Co and Ni show favorable thermodynamics and low CO2 microstructural role that affects reaction kinetics [51]. Our TG
decomposition barriers [17]. Cobalt is placed at the top of the analyses (Fig. 5) showed that the dehydrogenation tempera-
volcano curve which compares the activities of different ture of MgH2 (330 C) could be lowered 70 C or 80 C due to the
supported transition metals as a function of the reaction addition of Ni or Ni/CNTs, respectively. At low temperatures
energy for dissociative CO chemisorption but the diagram is (300 C) and also at high temperatures (400 C), the calculated
not useful when selectivity plays an important role [18]. In methane yield for the MgH2-10wt.%Ni and MgH2-5wt.%Ni-
this regard, our experimental results showed that metallic 5wt.%CNT samples after reaction with CO2 during 24 h did not
nickel is useful to improve CO2 conversion processes using show great differences. In fact, this observation is positive,
hydrides and that its performance is as good than that of since the reduction of the metallic charge, i.e. the amount of
cobalt or even slightly better in certain conditions. The nickel in the sample, did not cause a negative effect on the
replacement of Co by Ni has advantages. On the one hand, performance of the reaction. This means that by replacing
despite Co has a biologically necessary role as metal con- 5wt.% nickel with CNTs, a marginal improvement in methane
stituent of vitamin B12, excessive exposure has been shown yield from 1.7 to 4.5% was obtained at 300 C and no
to induce various adverse health effects. On the other hand, improvement was observed at 400 C (64.5 and 62.1%,
Ni is cheaper and highly available. In addition, the use of a H2 respectively). However, the differences were maximized at
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 4 7 ( 2 0 2 2 ) 4 2 8 e4 4 2 437
Fig. 10 e Molar percentage of CH4 after reaction with CO2 for the MgH2-10wt.%Ni and MgH2-5wt.%Ni-5wt.%CNT considering
different reaction times at (A) 400 C and (B) 350 C (CH4 yield is included).
intermediate temperatures. At 350 and 375 C the sample Theoretical calculations based on Gibbs free energy mini-
with less metallic charge and with the addition of CNTs was mization method for the MgH2-Ni system with CO2 (4:0.1:1) at
significantly better than the sample that only contained Ni: 400, 375 and 350 C were performed. Carbon nanotubes were
the amount of methane in the gas phase increased 34% at not considered as an active phase. The possible reactivity of
350 C and 32% at 375 C and the methane yield increased 37% CNTs was tested with a synthetic mixture of 4H2:1CO2 through
at 350 C and 29% at 375 C. With a thermal treatment of 24 h, heat treatment during 24 h at 375 C. GC analyses indicated
the best results for sample MgH2-5wt.%Ni-5wt.%CNT were that after exposure to the gas mixture, the resulting gas phase
obtained at 375 C. Under these conditions, the gas phase was was mainly H2 (76.3%) and CO2 (22.0%), with very small
mainly composed of CH4 (60.2%) and H2 (38.0%) with small amount of CH4 (1.0%) and CO (0.7%) (Fig. S9). Gas analyses of
amount of CO2 (1.8%) and traces of CO and a methane yield of the synthetic mixture under the same conditions showed the
74.5% was calculated. Considering long reaction times (48 h) generation of CO (0.6%) and CH4 (0.1%). The difference in CH4
at 350 C, CH4 yield rose to 79%. The performance improve- content in the gas phase after heating may be due to the
ment observed with the increase in treatment time from 24 to presence of iron particles coming from the milling jar or,
48 h at 350 C for the Ni and CNTs-doped sample was weaker although being purified, to a very small amount of metallic
than that observed for the sample catalyzed only with Ni. It is nanoparticles from which the nanotubes were grown. This
worth mentioning that after 24 h treatment, the best CH4 observation indicates that the addition of CNTs does not affect
yields for the Co [41] or Ni doped-MgH2 have been obtained at the thermodynamics as it can be considered as an inert specie
400 C and resulted to be 60.9% and 64.5%, respectively, and for the system. Our simulations show that the gas phase is
that after 48 h treatment at 350 C these samples have shown expected to be a mixture of H2 and CH4 with no residual CO2 or
a methane yield of 78% and 74.2%, respectively. Then, the CO formation (Fig. 11A). Water is not expected since as this
above presented results indicate that the co-addition of CNTs method allows predicting the phases in the chemical equi-
allows obtaining better conversion rates, especially at shorter librium, it considers the direct reduction and not the Sabatier
reaction time. Interestingly, when comparing values for 24 h route with catalysis. At a given total pressure of the system, as
treatment, for the Ni and CNT-doped MgH2, there seems to the temperature increases, the amount of methane decreases.
exist an optimal operating temperature (Fig. 9). As the Then, from a thermodynamic point of view, low temperatures
Sabatier reaction is exothermic and is disadvantaged with are required to obtain high CH4 yields because if the temper-
increasing temperature, this behavior is expected for a reac- ature rises the proportion of H2 in the gas phase grows.
tion that undergoes this catalyzed mechanism. As for the However, experimentally, poor conversion is obtained at low
MgH2-10wt.%Ni sample a linear relationship between the temperature due to kinetic restrictions. Regarding the solid
temperature and methane yield was observed, it would seem phase at 375 C, Mg, MgO, unreacted MgH2, Mg2Ni and carbon
that in that case it was not possible to reduce the activation deposition is expected (Fig. 11B). It is interesting to observe
energy enough to observe an optimum point in the range of that at this temperature the hydride phase remains without
temperatures studied. totally decomposing.
438 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 4 7 ( 2 0 2 2 ) 4 2 8 e4 4 2
Fig. 11 e Equilibrium composition (mol%) as a function of the total pressure of the system considering the mixture
4MgH2:0.1Ni:1CO2 for the (A) gas phase at 350, 375 and 400 C and the (B) solid phase at 375 C.
Table 3 e Gas-phase composition determined by GC analyses for as-milled MgH2, as-milled MgH2-10wt.%Ni, as-milled
MgH2-5wt.%CNT and as-milled MgH2-5wt.%Ni-5wt.%CNT after thermal treatment at 375 C for 24 h.
Sample Molar percentage Molar percentage Molar percentage Molar percentage CH4 yield
of CH4 (%) of CO2 (%) of CO (%) of H2 (%) (%)
MgH2 21.3 4.3 6.5 67.9 28.6
MgH2-5wt.% CNT 46.8 2 0.1 51.1 56.9
MgH2-10 wt% Ni 45.5 1.9 <1000 ppm 52.6 57.5
MgH2-5% wt.Ni-5wt.% CNT 60.3 1.8 <1000 ppm 38 74.5
Fig. 12 e Gas-state FTIR patterns analyses for as-milled MgH2, MgH2-10wt.%Ni, MgH2-5wt.%CNT and MgH2-5wt%Ni-5wt%
CNT after thermal treatment at 375 C for 24 h.
Fig. 13 e Relative intensities of the phases of Mg/MgO and Mg/(Mg þ MgO þ MgH2) at different temperatures for samples
MgH2-10wt.%Ni and MgH2-5wt.%Ni-5wt.%CNT.
440 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 4 7 ( 2 0 2 2 ) 4 2 8 e4 4 2
was richer in methane (60.3%) and the CH4 yield grew to 74.5%. (~350-375 C) the MgH2-5wt.%Ni-5wt.%CNT sample was
The Ni crystallite size calculation for the MgH2-10wt.%Ni and significantly better than the sample that only contained Ni: the
MgH2-5wt.%Ni-5wt.%CNT samples after mechanical milling amount of methane in the gas phase and the methane yield
applying the Scherrer method from X-ray diffraction patterns increased more than 30%. Its incorporation did not modify the
was useful to verify that co-addition of CNTs did not modify initial size of the nickel crystallites, nor did it have a catalytic
the size of the initial active catalytic phase, which is approx- role as it was proved that they do not interact with a gas
imately 30 nm. This allows ruling out that the performance mixture of H2 and CO2 at moderate temperatures. It is postu-
improvements were due to the catalytic phase size variation. lated that the role of CNTs could be related to the protection of
Since CO2 is easily adsorbed on the surface of MgH2, in the the magnesium hydride. While MgH2 crystallites are neigh-
absence of carbon nanotubes, it can be easily oxidized to MgO. bored by graphitic planes with open and interconnected
This discourages the occurrence of the Sabatier reaction that structure, its oxidation due to direct contact with CO2 and/or
requires H2 from the hydride and the obtained CH4 yield is the generated H2O could be impeded or retarded, promoting
lower. If the main peak relative intensities (i.e., integrated the occurrence of the Sabatier reaction. Through reaction be-
intensity) of the phases with magnesium (MgO, Mg and MgH2) tween Ni/CNTs doped MgH2 and CO2 at 350 C during 48 h, a gas
are compared in the XRD diffractograms after reaction at phase free of CO2 or CO, composed of CH4 (70.9%) and H2 (29.1%)
375 C, it can be observed that for the sample MgH2-5wt.%Ni- was obtained, with a calculated a methane yield of 79%. Then,
5wt.%CNT the relations Mg/MgO and Mg/(Mg þ MgO þ MgH2) this investigation provides a thermochemical method for CO2
are higher than for the sample MgH2-10wt.%Ni (see Fig. 13). recycling to synthesize COx-free mixture of methane and
Same trend is obtained at 350 and 400 C. Considering the hydrogen employing a well-known metal hydride (MgH2) as
MgH2 and MgH2-5wt.% CNT solid samples after reaction at hydrogen source, doped with a cheap and high available cat-
375 C, the later displayed a bigger proportion of metallic Mg alytic phase (Ni) and carbon nanotubes.
(Fig. S10). These observations suggest that regardless of the
reaction path followed, the role of CNT may be to function as a
protector of magnesium hydride, preventing or retarding its Declaration of competing interest
oxidation. It was previously proved that crystallites of MgH2
are neighbored by graphitic planes with open and inter- The authors declare that they have no known competing
connected structure [51]. It could be hypothesized that carbon financial interests or personal relationships that could have
nanotubes can surround the grain boundaries inside the hy- appeared to influence the work reported in this paper.
dride particles and this can hinder MgH2 from being oxidized
by preventing its direct contact with the CO2 and/or H2O that
is produced by methanation through the Sabatier reaction.
Acknowledgements
Moreover, for the process to occur via Sabatier reaction, it is
necessary to have an active catalytic species (in this case
The work was supported by CONICET (Consejo Nacional de
nickel) capable of dissociating the H2 and the CO2 adhered to cnicas), ANPCyT (Agencia
Investigaciones Cientı́ficas y Te
the surface of the hydride in the vicinity of the H activated n Cientı́fica y Tecnolo gica) and CNEA
Nacional de Promocio
form. These observations can justify the observed synergistic n Nacional de Energı́a Ato mica). The authors also
(Comisio
effect of the two additives.
thank Bernardo Pentke (Departamento Fisicoquı́mica de
Materiales) for the SEM micrographs and Pierre Arneodo Lar-
ochette (Departamento Fisicoquıı́mica de Materiales) for TG
Conclusions measurements.
production from steam methane reforming using carbon [22] Varun Y, Sreedhar I, Singh SA. Highly stable M/NiO-MgO
capture storage & utilization. Int J Hydrogen Energy (M¼Co, Cu and Fe) catalysts towards CO2 methanation. Int J
2021;46:22685e706. Hydrogen Energy 2020;45:28716e31.
[5] Sabatier P, Senderens JB. New synthesis of methane. Acad Sci [23] Jin L, Xie T, Ma B, Li Y, Hu H. Preparation of carbon-Ni/MgO-
Paris 1902;134:514e6. Al2O3 composite catalysts for CO2 reforming of methane. Int J
[6] Vogt C, Monai M, Kramer GJ, Weckhuysen BM. The Hydrogen Energy 2017;42:5047e55.
renaissance of the Sabatier reaction and its applications on [24] De S, Zhang J, Luque R, Yan N. Ni-based bimetallic
Earth and in space. Nat Catal 2019;2:188e97. heterogeneous catalysts for energy and environmental
[7] Iaquaniello G, Setinib S, Salladini A, De Falco M. CO2 applications. Energy Environ Sci 2016;9:3314e47.
valorization through direct methanation of flue gas and [25] Tada S, Shimizu T, Kameyama H, Haneda T, Kikuchi R. Ni/
renewable hydrogen: a technical and economic assessment. CeO2 catalysts with high CO2 methanation activity and high
Int J of Hydrogen Energy 2018;43:17069e81. CH4 selectivity at low temperatures. Int J Hydrogen Energy
[8] Go€ tz M, Lefebvre J, Mo € rs F, Koch AMD, Graf F, Bajohr S, 2012;37:5527e31.
Reimert R, Kolb T. Renewable Power-to-Gas: a [26] Guo M, Lu G. The effect of impregnation strategy on
technological and economic review. Renew Energy structural characters and CO2 methanation properties over
2016;85:1371e90. MgO modified Ni/SiO2 catalysts. Catal Commun
[9] Schaaf T, Grünig J, Roman Schuster M, Rothenfluh T, 2014;54(55e60).
Orth A. Methanation of CO2 - storage of renewable energy [27] Renda S, Ricca A, Palma V. Study of the effect of noble metal
in a gas distribution system. J Energy Sustain Soc promotion in Ni-based catalyst for the Sabatier reaction. Int J
2014;4:2e14. of Hydrogen Energy 2021;46:12117e27.
[10] Ro€ nsch S, Schneider J, Matthischke S, Schlüter M, Go € tz M, [28] Baldauf-Sommerbauera G, Luxa S, Anisera W, Bitschnaub B,
Lefebvre J, Prabhakaran P, Bajohr S. Review on methanation - Letofsky-Papstc I, Siebenhofera M. Steady-state and
from fundamentals to current projects. Fuel controlled heating rate methanation of CO2 on Ni/MgO in a
2016;166:276e96. bench-scale fixed bed tubular reactor. J CO2 Util 2018;23:1e9.
[11] Gutie rrez-Martı́n F, Rodrı́quez-Anto n LM. Power-to-SNG [29] Li X, Wang Y, Zhang G, Sun W, Bai Y, Zheng L, Han X, Wu L.
technologies by hydrogenation of CO2 and biomass Influence of Mg-promoted Ni-based catalyst supported on
resources: a comparative chemical engineering process coconut shell carbon for CO2 methanation. Chem Sel
analysis. Int J Hydrogen Energy 2019;44:12544e53. 2019;4:838e45.
[12] Vincenzo M, Jacopo D. Burning behaviour of high-pressure [30] Feng Y, Yang W, Chen S, Chu W. Cerium promoted nano
CH4-H2-air mixtures. Energies 2013;6:97e116. nickel catalysts Ni-Ce/CNTs and Ni-Ce/Al2O3 for CO2
[13] Ilbas M, Crayford AP, Yılmaza I, Bowen PJ, Syred N. Laminar- methanation. Integrated Ferroelectrics Int J: Int J
burning velocities of hydrogeneair and 2014;151(1):116e25.
hydrogenemethaneeair mixtures: an experimental study. [31] Hugelshofer CL, Borgschulte A, Callini E, Matam SK, Gehrig J,
Int J Hydrogen Energy 2006;31:1768e79. Hogand D, Züttel A. Gasesolid reaction of carbon dioxide
[14] Orhan Akansu S, Kahraman N, Çeper B. Experimental study with alanates. JPhysChemC 2014;118:15940e5.
on a spark ignition engine fueled by methaneehydrogen [32] Zhao J, Teng Y-L, Dong B-X. Thermal reduction of CO2 with
mixtures. Int J Hydrogen Energy 2007;32:4279e84. activated alkali metal aluminum hydrides for selective
[15] Klell M, Eichlseder H, Sartory M. Mixtures of hydrogen and methanation. Energy Fuels 2020;34:11210e8.
methane in the internal combustion engine -synergies, [33] Vitillo JG, Groppo E, Gil Bardajı́ E, Baricco M, Bordiga S. Fast
potential and regulations. Int J Hydrogen Energy carbon dioxide recycling by reaction with g-Mg(BH4)2. Phys
2012;37:11531e40. Chem Chem Phys 2014;16:22482.
[16] € Galata S, Karago
Sandalci T, Is‚ın O, _ Effect of
€ z Y, Güler I. [34] Picasso CV, Safin DA, Dovgaliuk I, Devred F, Debecker D,
hythane enrichment on performance, emission and Li HW, Filinchuk Y. Reduction of CO2 with KBH4 in solvent-
combustion characteristics of an ci engine. Int J Hydrogen free conditions. Int J Hydrogen Energy 2016;41:14377e86.
Energy 2019;44:3208e20. [35] Kato S, Borgschulte A, Ferri D, Bielmann M, Crivello J-C,
[17] Lui C, Cundari TR, Wilson AK. CO2 reduction on transition Wiedenmann D, Parlinska-Wojtan M, Rossbach P, Lu Y,
metal (Fe, Co, Ni and Cu) surfaces: in comparison with Remhof A, Züttel A. CO2 hydrogenation on a metal hydride
heterogeneous catalysis. J Phys Chem C 2012;116:5681e8. surface. Phys Chem Chem Phys 2012;14:5518e26.
[18] Frontera P, Macario A, Ferraro M, Antonucci PL. Supported [36] Grasso ML, Puszkiel J, Ferna ndez Albanesi L, Dornheim M,
catalysts for CO2 methanation: a review. Catalysts Pistidda C, Gennari FC. CO2 reutilization for methane
2017;7e35:59. production via catalytic process promoted by hydrides. Phys
[19] Li W, Wang H, Jiang X, Zhu J, Liu A, Guo X, Song C. A short Chem Chem Phys 2019;21:19825e34.
review of recent advances in CO2 hydrogenation to [37] Grasso ML, Puszkiel J, Gennari FC, Santoru A, Dornheim M,
hydrocarbons over heterogeneous catalysts. RSC Adv Pistidda C. CO2 reactivity with Mg2NiH4 synthesized by in
2018;8:7651e69. situ monitoring mechanical milling. Phys Chem Chem Phys
[20] Younas M, Kong LL, Bashir MJK, Nadeem H, Shehzad A, 2020;22:1944e52.
Sethupathi S. Recent advancements, fundamental [38] Dong B-X, Wang LZ, Song L, Zhao J, Teng YL.
challenges, and opportunities in catalytic methanation of Thermochemical reduction of carbon dioxide with alkali
CO2. Energy Fuels 2016;30:8815e31. metal hydrides, producing methane and hydrogen fuels at
[21] Fumihiko K, Toshiaki Y, Yuji A, Takehisa M, Hideyuki T, moderate temperature. Energy Fuels 2016;30:6620e5.
Koichi M, Yoshinobu F, Koji K. Effect of Ni content on CO2 [39] Zhao J, Wei Y-F, Cai Y-L, Wang L-Z, Xie J, Teng Y-L, Zhu W,
methanation performance with tubular-structured Ni-YSZ Shen M, Dong B-X. Highly selective and efficient reduction of
catalysts and optimization of catalytic activity for CO2 to methane by activated alkaline earth metal hydrides
temperature management in the reactor. Int J Hydrogen without a catalyst. ACS Sustainable Chem Eng
Energy 2020;45:12911e20. 2019;7:4831e41.
442 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 4 7 ( 2 0 2 2 ) 4 2 8 e4 4 2
[40] Zhang S, Chen H-Q, Kan X-T, Tai Y-L, Liu W-L, Dong B-X, purification and CO methanation. Int J Hydrogen Energy
Teng Y-L. Storage and in-situ preparation of H2-mixed CH4 2020;45:19493e504.
fuel by thermochemical reduction of inorganic carbonates [46] Outokumpu HSC. Chemistry for windows, version 6.1.
with activated metal hydrides. Fuel 2021;292. Finland: Outokumpu Research Oy; 2009.
[41] Amica G, Rozas Azcona S, Aparicio S, Gennari FC. Catalysis [47] Cova F, Arneodo Larochette P, Gennari FC. Hydrogen sorption
effect on CO2 methanation using MgH2 as a portable in MgH2-based composites: the role of Ni and LiBH4
hydrogen medium. Phys Chem Chem Phys 2020;22:14720e30. additives. Int J of Hydrogen Energy 2012;37:15210e9.
[42] Kato S, Kumar Matam S, Kerger P, Bernard L, Battaglia C, [48] Gennari FC, Castro F, Urretavizcaya G. Hydrogen desorption
Vogel D, Rohwerder M, Züttel A. The origin of the catalytic behavior from magnesium hydrides synthesized by reactive
activity of a metal hydride on CO2 reduction. Angew Chem mechanical alloying. J Alloys Compd 2001;321:46e53.
2016;128:6132e6. [49] Du H, Williams CT, Ebner AD, Ritter A. In situ FTIR
[43] Burr JG, Weldon G, Brown WG, Heller HE. The reduction of spectroscopic analysis of carbonate transformations during
carbon dioxide to formic acid. Am Chem Soc adsorption and desorption of CO2 in K-promoted HTlc. Chem
1950;72:2560e2. Mater 2010;22:3519.
[44] Gribble GW, Nutaitis CF. Sodium borohydride in carboxylic [50] Selvam P. Evidence for the formation of surface carbonates
acid media. A review of the synthetic utility of on some hydrogen storage intermetallic compounds: an XPS
acyloxyborohydrides. Org Prep Proced Int: New J Org Synth study. Int J Hydrogen Energy 1990;15:133e7.
1985;17:317e84. [51] Vero n M, Troiani H, Gennari FC. Synergetic effect of Co and
[45] Gamba N, Puszkiel JA, Larochette PA, Gennari FC. Dual carbon nanotubes on MgH2 sorption properties. Carbon
application of Ti-catalyzed Li-RHC composite for H 2 2011;49:2413e23.