Understanding Coke Deposition Vis-à-Vis DRM Activity over Magnesia-Alumina Supported Ni-Fe, Ni-Co, Ni-Ce, and Ni-Sr Catalysts

The catalytic conversion of CH4 and CO2 into H2-rich syngas is known as the dry reforming of methane (DRM). The dissociation of CH4 over active sites, coupled with the oxidation or polymerization of CH4−x (x = 1–4), plays a crucial role in determining in determining the DRM product yield and coke deposition. Herein, a series of bimetallic-supported catalysts are prepared by the dispersion of Ni-M (M = Ce, Co, Fe, and Sr) over 60 wt% MgO-40 wt% Al2O3 (60Mg40Al) support. Catalysts are tested for DRM and characterized with XRD, surface area and porosity, temperature-programmed reduction/desorption, UV−VIS−Raman spectroscopy, and thermogravimetry. 2.5Ni2.5Sr/60Mg40Al and 2.5Ni2.5Fe/60Mg40Al, and 2.5Ni2.5Ce/60Mg40Al and 2.5Ni2.5Co/60Mg40Al have similar CO2 interaction profiles. The 2.5Ni2.5Sr/60Mg40Al catalyst nurtures inert-type coke, whereas 2.5Ni2.5Fe/60Mg40Al accelerates the deposition of huge coke, which results in catalytic inferiority. The higher activity over 2.5Ni2.5Ce/60Mg40Al is due to the instant lattice oxygen-endowing capacity for oxidizing coke. Retaining a high DRM activity (54% H2-yield) up to 24 h even against a huge coke deposition (weight loss 46%) over 2.5Ni2.5Co/60Mg40Al is due to the timely diffusion of coke far from the active sites or the mounting of active sites over the carbon nanotube.


Introduction
The concentration of greenhouse gases has reached a critical level, and the effects are being seen everywhere.Apart from the rise in sea level, drastic seasonal changes, droughts, and wildfires are also becoming intensified across the globe.At this point, the catalytic conversion of the two leading greenhouse gases, CH 4 and CO 2 , into H 2 -rich syngas is important.The proper catalyst for DRM may be a game changer for depleting greenhouse gases from the environment, and one of the main products, H 2 , is considered to be clean energy.This reaction is known as the dry reforming of methane (DRM), and it is highly endothermic (CH 4 + CO 2 → 2H 2 + 2CO; ∆H • = 247.34kJ/mol).Pt, Pd, Ir, Ru, and Ni dispersed over different supports such as MgO, CeO 2 , La 2 O 3 , Y 2 O 3 , TiO 2 , Si 2 O 3 , Al 2 O 3, and ZrO 2 are largely investigated for this reaction [1].
The selection of Ni as the active site and MgO, Al 2 O 3, and TiO 2 as the support is cost effective.Among the supports, TiO 2 is known for its stronger metal−support interaction.Still, the catalytic performance of the Ni/TiO 2 catalyst was found to be inferior to the Ni/MgO and Ni/Al 2 O 3 catalysts due to the destabilization of titania phases, as well as the coverage of active sites of Ni by TiO x species [2][3][4][5][6][7].Ni/MgO forms a NiO-MgO solid solution, which becomes hard at higher temperatures and results in a lower reducibility of NiO.It turns the inferior catalysts into the Ni/Al 2 O 3 catalyst [4][5][6][7].The Ni/Al 2 O 3 catalyst was found to be superior compared with all of the above catalysts.However, an enriched surface acid profile vis-à-vis massive coke deposition limits its optimum catalytic performance on an industrial scale [8].On the other hand, the addition of MgO with Al 2 O 3 has been found to inhibit the diffusion of Ni active sites into Al 2 O 3 (by forming NiAl 2 O 4 ), and stabilizes Ni particles over MgAl 2 O 4 .MgO also depletes the acidity of Al 2 O 3 and resists carbon deposition [9].
How can the catalytic limitation over the Ni/Al 2 O 3 catalyst or Ni/MgO be overcome?Can we decrease the acidity of the support in the Ni/Al 2 O 3 catalyst by incorporating some amount of MgO? Can we retain the reducibility of NiO (above 700 • C) over MgO by adding Al 2 O 3 ?Keeping the major framework of basic MgO and rest by Al 2 O 3 , both DRM's favorable properties of inherent basicity and Ni stability can be retained above 700 • C DRM temperature.Such an approach was previously tested for DRM by Quan et al. by keeping a Mg/Al ratio of 1:3 [8].So, the dispersion of Ni over the MgO-Al 2 O 3 support may be a promising approach for DRM.Further, the catalytic performance can be elevated by using suitable promotors over the Ni/MgO-Al 2 O 3 catalyst.
Because of its inherent basic property vis-à-vis CO 2 interactions, alkaline earth metals are frequently used as promoters over alumina-supported Ni catalysts.The Sr-promoted alumina-supported Ni catalyst showed enhanced basicity and a more significant metal support interaction by forming NiAl 2 O 4 [10].Alipour et al. found that Mg, Ba, and Camodified Ni/Al 2 O 3 had a higher reducibility, coke resistance, and activity in the order of Mg > Ba > Ca (than Ni/Al 2 O 3 catalyst) [11].Karam et al. claimed the enhanced formation of small metallic Ni by spinel nickel-aluminate nano species by adding Mg over an aluminasupported Ni catalyst [12].Diyanath et al. showed an enhanced metal−support interaction of Ni over magnesia-alumina support [13].Akki et al. revealed that ceria-promoted MgAl 2 O 4 -supported Ni catalyst attained a highly reducible surface with enhanced basicity and effective active oxygen transfer [14].Iron incorporation over an alumina-supported Ni catalyst was found to inhibit the deposition of inactive carbon and the oxidation of metallic Ni [15].Previously, the promotional addition of Co over a magnesia-alumina -supported Ni catalyst resulted in enhanced coke elimination due to the high oxygen affinity of cobalt [16].
A further literature survey reveals the promotional role of Ce, Co, Fe, and Sr towards DRM.Herein, 2.5 wt% Ni-2.5 wt% M (M = Ce, Co, Fe, Sr) is supported over 60 wt% MgO-40 wt%Al 2 O 3 .The catalyst system is investigated for DRM and is characterized thoroughly via X-ray diffraction, surface area and porosity measurement, H 2 -temperature programmed reduction, CO 2 -temperature programmed desorption, UV−VIS spectroscopy, Raman spectroscopy, and thermogravimetry.Here, the good correlation of characterization results with the catalytic activity is aimed at bringing about more insight into two major competitive routes, DRM reaction and coke deposition, through four sets of bimetallic metal sites (Ni-Co, Ni-Fe, Ni-Sr, and Ni-Ce) over a cheap magnesia-alumina support.It may also pave the way for developing a robust DRM catalyst system using cheap chemicals like MgO, Al 2 O 3 nickel nitrate, iron nitrate, strontium nitrate, and cobalt nitrate.

Catalyst Preparation
Here, 0.95 g of 60 wt%MgO-40 wt%Al 2 O 3 support was dispersed in 100 mL of deionized water at room temperature for 20 min.The nitrate precursor of Ni (equivalent to 2.5 wt%) and nitrate precursor of one metal (equivalent to 2.5 wt%) were added to the dispersed support solution at 80 • C until dry.The obtained solid sample was first calcined in air at 600 • C for 3 h.The support was abbreviated as 60Mg40Al, and the Ni supported catalysts were abbreviated as 2.5Ni-2.5M/60Mg40Al(M = Ce, Co, Fe, and Sr).

Catalyst Activity Test
The dry reforming of the methane experiment was carried over 0.1g of promoted catalysts at 700 • C under 1atm pressure in a stainless steel vertical fixed tubular reactor (PID Eng.& Tech Micro Activity Reference, 9.1 mm i.d. and 30 cm long).The temperature of the reactor was monitored by an axially positioned thermocouple (K-type stainless sheathed) at the center of the catalyst bed.Before the catalytic tests, reductive pretreatment of the catalyst samples was carried out under the flow of hydrogen (20 mL/min) for 60 min at 600 • C. The mixture of gases fed CH 4 /CO 2 /N 2 in the respective volumes as 6:6:2, and the volume flow rate was 70 mL•min −1 and 42,000 mL(h•g cat ) −1 gas hourly space velocity was passed through the reactor.The product gas stream was analyzed with a GC (GC-2014 Shimadzu) unit equipped with a thermal conductivity detector and two columns, Porapak Q and Molecular Sieve 5A.H 2 yield % and H 2 /CO molar ratio are determined using the following expressions: The surface and porosity results of the 2.5Ni-2.5M/60Mg40Al(M = Ce, Co, Fe, Sr) catalysts are shown in Figure 2. All catalysts showed type IV isotherms with an H4 hysteresis loop, indicating the presence of both micropores and mesopores [17].The 2.5Ce2.5M/60Mg40Al(M = Ce, Co, Fe) catalyst had similar surface parameters (surface area 199-202 m 2 /g and pore volume 0.32-0.34cm 3 /g).The surface area and pore volume of the 2.5Ni2.5Sr/60Mg40Alcatalyst were the lowest (surface area: 174 m 2 /g, pore volume: 0.29 cm 3 /g) compared with the other catalysts.This indicates the serious accumulation of NiO and Sr compounds inside the pores.
The Raman spectra of the 2.5Ni2.5M/60Mg40Al(M = Ce, Co, Fe, Sr) catalysts are shown in Figure 3.The surface and porosity results of the 2.5Ni-2.5M/60Mg40Al(M = Ce, Co, Fe, Sr) catalysts are shown in Figure 2. All catalysts showed type IV isotherms with an H4 hysteresis loop, indicating the presence of both micropores and mesopores [17].The 2.5Ce2.5M/60Mg40Al(M = Ce, Co, Fe) catalyst had similar surface parameters (surface area 199-202 m 2 /g and pore volume 0.32-0.34cm 3 /g).The surface area and pore volume of the 2.5Ni2.5Sr/60Mg40Alcatalyst were the lowest (surface area: 174 m 2 /g, pore volume: 0.29 cm 3 /g) compared with the other catalysts.This indicates the serious accumulation of NiO and Sr compounds inside the pores.The 2.5Ni2.5Ce/60Mg40Alcatalyst showed an intense RAMAN band about 463 cm −1 , which is attributed to the symmetric F 2g stretching mode of oxygen surrounding the Ce 4+ in cubic CeO 2 [18,19] (Figure 3a).The XRD of 2.5Ni2.5Ce/60Mg40Alalso showed the prominent peak for cubic CeO 2 .The 2.5Ni2.5Sr/60Mg40Alcatalyst showed two sharp Raman peaks at 560 cm −1 and 1071 cm −1 for the Sr-O and CO 3 2− vibration of Sr 2+ , respectively [20].The absence of Sr-related phases in XRD may be due to the good dispersion of the Sr compound over the catalyst surface.The 2.5Ni2.5Co/60Mg40Alcatalyst showed Raman vibration bands at 282 cm −1 , 580 cm −1 , and 672 cm −1 .In the literature, the Raman peaks for Co(OH) 2 have been reported at 284 cm −1 , 462 cm −1 (OCoO bending mode), and 522 cm −1 (OH deformation mode) [21,22].Previously, in the Ni 1-x Co x TiO 3 sample, as per the varying degree of x from 0.05 to 0.80, the Raman band was found to shift from 289 to 272 cm −1 [23].This means for the 2.5Ni2.5Co/60Mg40Alcatalyst, the Raman peak of about 284 cm −1 was for the Co(OH) 2 vibration and was very sensitive to the degree of substitution.The Raman peak at 580 cm −1 was reported for the vibration mode of CoO(OH) (oxidation state of Co (III)) [21].The peak at 580 cm −1 was the most intense Raman peak in the 2.5Ni2.5Co/60Mg40Alcatalyst.The Raman band at 675-680 cm −1 was a typical band reported for Co-O vibration in CoO, Co 3 O 4 , and CoO•xCoO [22,[24][25][26].From the Raman analysis, an interacted-CoO, Co 3 O 4 , Co (OH) 2, and CoO (OH)-type composition can be claimed.Previously, the absence of the CoO (OH) pattern in XRD was due to its amorphous nature [21].It is noticeable that, here, the catalyst was prepared at 600 • C, activated at 800 • C, and employed for the DRM reaction at 700 • C.During this temperature range, various types of interconversion between cobalt compounds may be possible [22], as per the following reactions: 6Co(OH Here, H 2 -TPR was taken after the catalyst preparation.This means that over a fresh 2.5Ni2.5Co/60Mg40Alcatalyst, some complex structure of cobalt compound made up of 6Co (OH) 2 and Co 3 O 4 (or Co 2 O 3 •CoO) may be present.In XRD, the phase for NiO was observed, but no phases of the cobalt compound were found.So, it can be expected that the cobalt compound may be amorphous or the cobalt compound may be more dispersed than NiO over the catalyst.The 2.5Ni2.5Fe/60Mg40Alcatalyst had Raman bands at 280 cm −1 , 480 cm −1 , and 689 cm −1 .In the literature, peaks at 290 cm −1 , 490 cm −1 , and 659 cm −1 were reported for the Raman vibration of Fe 2 O 3 [27].The red and blue shift of the Raman band (concerning the standard Fe 2 O 3 band) was due to the insertion of the hetero-atom in the lattice of Fe 2 O 3 .The band at 689 cm −1 may be related to structural disorder due to the insertion of the Al cation into the iron oxide lattice [28].The Raman spectra of the 2.5Ni2.5M/60Mg40Al(M = Ce, Co, Fe, Sr) catalysts are shown in Figure 3.The 2.5Ni2.5Ce/60Mg40Alcatalyst showed an intense RAMAN band about 463 cm −1 , which is attributed to the symmetric F2g stretching mode of oxygen surrounding the Ce 4+ in cubic CeO2 [18,19] (Figure 3a).The XRD of 2.5Ni2.5Ce/60Mg40Alalso showed the prominent peak for cubic CeO2.The 2.5Ni2.5Sr/60Mg40Alcatalyst showed two sharp Raman peaks at 560 cm −1 and 1071 cm −1 for the Sr-O and CO3 2− vibration of Sr 2+ , respectively [20].The absence of Sr-related phases in XRD may be due to the good dispersion of the Sr compound over the catalyst surface.The 2.5Ni2.5Co/60Mg40Alcatalyst showed Raman vibration bands at 282 cm −1 , 580 cm −1 , and 672 cm −1 .In the literature, the Raman peaks for Co(OH)2 have been reported at 284 cm −1 , 462 cm −1 (OCoO bending mode), and 522 cm −1 (OH deformation mode) [21,22].Previously, in the Ni1-xCoxTiO3 sample, as per the varying degree of x from 0.05 to 0.80, the Raman band was found to shift from 289 to 272 cm −1 [23].This means for the 2.5Ni2.5Co/60Mg40Alcatalyst, the Raman peak of about 284 cm −1 was The bandgap of the UV−VIS results is shown in Figure 3b.Among all of the catalysts, the band gap between the valance band and conduction band was found to be lowest for the 2.5Ni2.5Co/60Mg40Alcatalyst and highest for 2.5Ni2.5Sr/60Mg40Al.The H 2 -temperature programmed reduction profile (H 2 -TPR) of 2.5Ni2.5M/60Mg40Al(M = Ce, Co, Fe, Sr) catalysts are shown in Figure 3c.The negative H 2 -TPR in the temperature range of 200 • C was attributed to hydrogen spillover into the mesopores [29].Upon the incorporation of Ce or Sr along with Ni over 60Mg/40Al, there was a single intense peak of about 900 • C, which was attributed to reducible "strongly interacted NiO" species [30].The H 2 -TPR profile of the 2.5Ni2.5Fe/60Mg40Alcatalyst was unique.It had merge peak at 500 • C and 550 • C and a peak at 850 • C.These peaks were attributed to the reduction in "moderately interacted NiO" species (into Ni), reduction in Fe 3 O 4 (into FeO), and "strongly interacted NiO" species (into Ni), respectively [30].It may be possible that the reduction peak of FeO to Fe may be merged with a peak of about 850 • C over the 2.5Ni2.5Fe/60Mg40Alcatalyst.So, the peak intensity at 850 • C may be due to reducible strongly-interacted-NiO and reducible FeO species.

Catalytic Activity Result and Discussion
The catalytic activity of the pure NiO and support-60Mg40Al towards dry reforming of methane were tested.The H 2 yield and CO yield with pure NiO were found to be only 0.75% and 1.25%, respectively, at a reaction temperature of 700 • C. The support "60 wt%MgO40 wt%Al 2 O 3 " was also found to exhibit a catalytic activity, but it was much more selective towards CO than H 2 .60Mg40Al showed 0.06% H 2 yield and 4.10 CO yield.However, when Ni was dispersed over the 60Mg40Al support, the H 2 yield eventually increased more than 60%.This suggests that the size and stability of Ni against high temperatures over proper support is essential for achieving a high DRM activity.The size and morphology of Ni can be controlled by proper calcination methodology and mixing a suitable matrix with Ni during the catalyst preparation [35,36].The XRD results show that the Al 2 O 3 and MgO metal oxide matrixes mixed, and the support obtained a new phase (orthorhombic MgAl 2 O 4 ) identity that was composed of Mg, O, and Al.That means a homogenous matrix of Mg, O, and Al was ready to carry the NiO-MO x (M = Ce, Co, Fe, Sr) metal oxide to catalyze the dry reforming of methane.The catalytic active sites of "metallic Ni" were derived from the reduction in "strongly interacted NiO" species over the 2.5Ni-2.5M/60Mg40Al(M = Ce, Co, Fe, Sr) catalyst system.The catalytic activity results of 2.5Ni-2.5M/60Mg40Al(M = Ce, Co, Fe, Sr) catalysts are shown in Figure 6.Among all of the catalysts, 2.5Ni2.5Co/60Mg40Alhad a 57% H 2 yield, which decreased to 54% after 24 h.The CO yield was above the H 2 yield, and it remains between 62-60% 24 h TOS.The relatively higher CO yield than H 2 yield indicates the presence of H 2 consuming reactions.Over the catalytic surface, H 2 can interact with carbon monoxide (CO) or carbon dioxide (CO 2 ) and form water (H 2 O) and carbon (H 2 + CO → H 2 O + C; 2H 2 + CO 2 → 2H 2 O + C).However, with a positive value for Gibb's free energy, these reactions were not thermodynamic feasible [37].On the other hand, the reaction of H 2 and CO 2 forming CO and water is thermodynamically feasible and is known as reverse water gas shift reaction.This reaction increases the concentration of CO at the expense of the H 2 concentration, resulting into a higher CO yield.It is worth noting that we did not detect H 2 O in the gas-chromatogram, indicating the involvement of water in some other reactions.Thermodynamic equilibrium studies of DRM via Gibbs free energy have shown that the gasification of coke by water is a thermodynamically favorable reaction (C + H 2 O → CO + H 2 ).Overall, the reverse water gas shift reaction and gasification reaction were also present along with the DRM reaction, resulting in an increase in CO concentration or CO yield [37].The 2.5Ni2.5Ce/60Mg40Alcatalyst showed a moderate activity towards DRM and H 2 yield, and the CO yield remained between 51-49% and 60-59% during 24 h TOS, respectively.The activity of the 2.5Ni2.5Fe/60Mg40Aland 2.5Ni2.5Sr/60Mg40Alcatalysts was found to be inferior, and their catalytic H 2 yield remained about 45% during 24 h TOS.

Catalytic Activity Result and Discussion
The catalytic activity of the pure NiO and support-60Mg40Al towards dry reforming verse water gas shift reaction and gasification reaction were also present along with the DRM reaction, resulting in an increase in CO concentration or CO yield [37].The 2.5Ni2.5Ce/60Mg40Alcatalyst showed a moderate activity towards DRM and H2 yield, and the CO yield remained between 51-49% and 60-59% during 24 h TOS, respectively.The activity of the 2.5Ni2.5Fe/60Mg40Aland 2.5Ni2.5Sr/60Mg40Alcatalysts was found to be inferior, and their catalytic H2 yield remained about 45% during 24 h TOS.Here, we discuss the catalytic routes and coke deposition routes over different catalyst surfaces.The possible coke deposition routes are shown in Figure 7.The catalyst may bear the sites for CH 4 decomposition or CO 2 decomposition [38][39][40].Further, the decomposed species (reaction intermediate) were quite active and could interact with another species or the same species instantly.In the current catalyst system, metallic Ni sites were available for CH 4 decomposition.It is well accepted that CH 4 is decomposed into CH 4−x and H 2 over metallic Ni (Step S1; Figure 7).After that, CH 4−x may be oxidized by CO 2 (under DRM) into syngas (CO + H 2 ) (Step S2, Figure 7).However, the delay in the oxidation of CH 4−x allowed for the polymerization of CH 4−x into coke (Step S2 ; Figure 7).In a recent study, Zhang et al. carried out DFT studies focused on methane cracking over a catalyst surface and the subsequent growth of cracked carbon from the C1 carbon unit to the C6 carbon unit (C1-C6) [41].From the C1-C4 carbon growth, "chain to chain" growth path (lengthening of carbon chain) was the main growth path.Further, at >C4, the growth path shifted from "chain to chain" to "chain to ring".The growth energy of the "chain to ring" path was lower (1.5 eV) than the "chain to chain" path from C4-C5, and it was further lowered to 0.55 eV from C5-C6.Clearly, the C6 unit was formed majorly from chain to ring growth pathways.It is basic unit of complex structural carbon species such as graphene and carbon nanotube.
The 2.5Ni2.5Co/60Mg40Alcatalyst has massive coke deposition (weight loss 46%); these cokes are sp 2 hybridized defective carbon (disordered type) and ordered carbon (graphitic types).However, the TEM image of the spent-2.5Ni2.5Co/60Mg40Alcatalyst shows very few carbon nanotubes, but the impression of dark marks is frequently visible.The catalyst has the highest and constant catalytic activity (54% H 2 yield from 14 h to 24 h reaction time) among the rest of the catalysts.The presence of few carbon nanotubes, the impression of frequent dark marks, massive weight loss, and the high catalytic activity of the 2.5Ni2.5Co/60Mg40Alcatalyst indicate that carbon deposits are mostly accumulated as dark impressions over the catalyst, without affecting the catalytic active sites.If the rate of coke formation (at active sites) is firmly matched to the rate of coke diffusion (far from catalytic active sites) (Route 1; R 1 ; Figure 7), then active sites remain exposed and coke is accumulated far from the active sites [42].This means that activity is not affected, even with a high coke deposition.
If the precursor of the carbon nanotube/C6 unit is not diffused out in a timely manner (far from the active site), it rapidly settles into the cell gap inside the Ni particle (or cell gap of active sites) (Route 2; R 2 ; Figure 7).From here, there are two sub-routes.The first is the diffusion of coke inside the Ni particle, followed by the precipitation of coke at the site near the support.Slowly, the precipitated cokes are accumulated, the carbon nanotube grows, and, finally, catalytic active Ni is uplifted (R 2a ).In this way, metallic Ni is always mounted above the nanotube and remains exposed to the reactant.So, the activity of the catalyst is not suppressed.The second sub-route is coke deposition mounting in the cell gap and slowly encapsulating the active sites (metallic Ni) (R 2b ).Encapsulation of the active sites cuts the contact between the reactant and the catalyst.If step S2 (CH 4−x oxidation) is delayed over step S2 (CH 4−x polymerization), encapsulation of the active sites is faster, and the catalytic activity is retarded.The 2.5Ni2.5Fe/60Mg40Aland 2.5Ni2.5Sr/60Mg40Alcatalysts have a comparable activity (45-46% H 2 -yield) but inferior to the 2.5Ni2.5Ce/60Mg40Aland 2.5Ni2.5Co/60Mg40Alcatalysts.The 2.5Ni2.5Fe/60Mg40Alcatalyst has the same CO 2 interaction profile as the 2.5Ni2.5Sr/60Mg40Alcatalyst, but the bandgap energy (for electron transfer from the valance band to the conduction band) of 2.5Ni2.5Fe/60Mg40Al is about half that of 2.5Ni2.5Sr/60Mg40Al.The 2.5Ni2.5Fe/60Mg40Alcatalyst is additionally populated with reducible iron oxide, which is known to trigger CH 4 decomposition [42], but it has a huge coke deposition (~54% weight loss).The CH 4 -TPSR experiment also verifies the presence of the highest density of CH 4 decomposition sites being in the 2.5Ni2.5Fe/60Mg40Alcatalyst.The low catalytic activity (45% H 2 -yield) of 2.5Ni2.5Fe/60Mg40Alindicates that many of the active sites are arrested/encapsulated by bulk coke.The 2.5Ni2.5Sr/60Mg40Alcatalyst has less surface area (and pore volume), which carries a diverse array of carbon deposits from the active to inert range (verified by Raman of spent catalyst).The shading of catalytic active sites by inert carbon over 2.5Ni2.5Sr/60Mg40Alcauses the least catalytic activity (~45% H 2 -yield).The activity of the 2.5Ni2.5Ce/60Mg40Alcatalyst is more progressed than the 2.5Ni2.5Fe/60Mg40Aland 2.5Ni2.5Sr/60Mg40Alcatalysts due to its capability to release lattice oxygen instantly (prior than CO 2 ) to oxidize carbon deposits.This results in the lowest non-graphitic carbon deposit and 49% H 2 yield during 24 h TOS.Here, we discuss the catalytic routes and coke deposition routes over different catalyst surfaces.The possible coke deposition routes are shown in Figure 7.The catalyst may bear the sites for CH4 decomposition or CO2 decomposition [38][39][40].Further, the decomposed species (reaction intermediate) were quite active and could interact with another species or the same species instantly.In the current catalyst system, metallic Ni sites were available for CH4 decomposition.It is well accepted that CH4 is decomposed into CH4−x and H2 over metallic Ni (Step S1; Figure 7).After that, CH4−x may be oxidized by CO2 (under DRM) into syngas (CO + H2) (Step S2, Figure 7).However, the delay in the oxidation of CH4−x allowed for the polymerization of CH4−x into coke (Step S2′; Figure 7).In a recent study, Zhang et al. carried out DFT studies focused on methane cracking over a catalyst surface and the subsequent growth of cracked carbon from the C1 carbon unit to the C6 carbon unit (C1-C6) [41].From the C1-C4 carbon growth, "chain to chain" growth path (lengthening of carbon chain) was the main growth path.Further, at >C4, the growth path shifted from "chain to chain" to "chain to ring".The growth energy of the "chain to ring" path was lower (1.5 eV) than the "chain to chain" path from C4-C5, and it was further lowered to 0.55 eV from C5-C6.Clearly, the C6 unit was formed majorly from chain to ring growth pathways.It is basic unit of complex structural carbon species such as graphene and carbon nanotube.The 2.5Ni2.5Co/60Mg40Alcatalyst has massive coke deposition (weight loss 46%); these cokes are sp 2 hybridized defective carbon (disordered type) and ordered carbon (graphitic types).However, the TEM image of the spent-2.5Ni2.5Co/60Mg40Alcatalyst shows very few carbon nanotubes, but the impression of dark marks is frequently visible.The catalyst has the highest and constant catalytic activity (54% H2 yield from 14 h to 24 h reaction time) among the rest of the catalysts.The presence of few carbon nanotubes, the

Conclusions
The formation of mixed oxide phases (orthorhombic MgAl 2 O 4 phase) by MgO-Al 2 O 3 indicates a more homogeneous distribution of atoms (Mg, O, and Al) in the support material.The support serves as a carrier for NiO-MO x (M = Ce, Co, Fe, and Sr) metal oxides, which catalyze the DRM reaction.The CO 2 interaction profiles of both the 2.5Ni2.5Sr/60Mg40Aland 2.5Ni2.5Fe/60Mg40Alcatalysts are similar, but the latter has less bandgap energy for electron transfer.Th relativelylow surface area and shading of active sites by inert carbon species make the 2.5Ni2.5Sr/60Mg40Alcatalyst inferior (45% H 2 -yield).The presence of surface-active reducible iron oxide over 2.5Ni2.5Fe/60Mg40Alcatalyst triggers pronounced CH 4 decomposition, resulting in a huge carbon (inert and active) deposit (54% weight loss), which results in an inferior catalytic activity (47% H 2 -yield).Again, the CO 2 interaction profiles of the 2.5Ni2.5Ce/60Mg40Aland 2.5Ni2.5Co/60Mg40Alcatalysts are similar.The sudden interaction of lattice oxygen to the 2.5Ni2.5Ce/60Mg40Alcatalyst causes the potential oxidation of coke.During 24 h on stream, the 2.5Ni2.5Ce/60Mg40Alcatalyst achieves 49% H 2 yield with minimal deposit of non-graphitic carbon.The 2.5Ni2.5Co/60Mg40Alcatalyst system is characterized by the lowest bandgap for electron transfer and non-shading of the catalytic active sites, even with high coke deposition (weight loss of 46%).This is accomplished through the diffusion of coke far away from the active sites or by mounting the active sites over the carbon nanotubes.These surface modifications favor DRM, and the 2.5Ni2.5Ce/60Mg40Alcatalyst shows a 54% H 2 yield for up to 24 h in the stream test.

Figure 7 .
Figure 7. Different routes of coke deposition.Route 1 (R 1 ): Decomposition of CH 4 at catalytic active sites (into CH 4−x ), polymerization of CH 4−x (into coke), and diffusion of coke far from the active sites.Route 2 (R 2 ): decomposition of CH 4 at catalytic active sites (into CH 4−x ), polymerization of CH 4−x (into coke), and coke deposition into the cell gap of active sites (as Ni-metal).Sub-route R 2a : Coke deposition inside Ni-M, precipitation at metal−support junction, and mounting of Ni over coke.Sub-route R 2b : Coke deposition inside Ni-M, encapsulation of active sites (Ni-M).M = Fe, Co, Ce, Ni; S 1 : CH 4 decomposition into CH 4−x , S 2 : Oxidation of carbon deposit, S 2 .polymerization of carbon deposit into coke.