Promoter Effect on Carbon Nanosphere-Encapsulated Fe-Co Catalysts for Converting CO 2 to Light Olefins

: For this work, we investigated the promotor effect (M = Na + , K + . Ce 3+ , Zn 2+ , Mn 2+ ) on carbon nanosphere-encapsulated bimetallic Fe-Co core–shell catalysts for CO 2 hydrogenation, promising selectivity for converting CO 2 to light olefins. The fresh and spent catalysts were characterized using a combination of experimental techniques such as scanning electron microscopy (SEM), X-ray diffraction (XRD), thermogravimetric analysis and differential scanning calorimetry (TGA−DSC), and Raman spectroscopy, and our results reveal that the addition of the promotor M enhanced the formation of graphitic carbon and metal carbides in the promoted catalysts when compared with the unpromoted catalysts. The metal carbides were determined to be the active sites for the production of light olefins


Introduction
Abundant CO2 emissions have detrimental effects on the environment [1].Consequently, sea levels are rising, the number of hurricanes and wildfires is growing, and more dangerous heat waves and severe droughts are occurring in many areas.Therefore, there is a pressing need to regulate CO2 emissions to alleviate their negative impact on the environment.Transforming CO2 into value-added chemicals and fuels is highly desirable for CO2 emission mitigation [2,3].
Other catalytic processes for CO2 valorization, such as methanol synthesis (proposed by Cordero-Lanzac et al. [14]) and dimethyl ether (DME) synthesis (proposed by Ateka et al. [15,16]) have been widely studied, but they are not the focus of this work.
Different catalysts based on transition metals or metal carbides have been extensively studied for CO2 conversion [10,11,17,20,21].To improve catalyst performance, it is desirable to prepare stable small metal particles under reaction conditions.Confined nanocatalysts based on "nanoreactors" have received growing attention [22][23][24][25][26][27][28][29][30][31][32][33][34][35].One of the promising approaches is to construct core-shell catalysts wherein the core serves as the nanoreactor, the metal species within the core acts as the active sites, and the shell acts as the boundary to hold the metal species in the core to prevent from metal agglomeration.In particular, core-shell nanocatalysts have been studied extensively [22,23,26,[35][36][37][38][39][40][41][42][43][44][45][46][47].Gupta et al. designed a core-shell nanoreactor with partially graphitized carbon in the shell and Fe3O4 and Fe5C2 in the core for CO2 hydrogenation [38].Fe3O4 nanoparticles encapsulated within graphitic carbon shells are more efficient than conventional catalysts [43,44].Porous graphene-confined Fe-K and carbon-confined magnesium hydride nanolamellae are highly efficient catalysts for the direct hydrogenation of CO2 to light olefins [48,49].Fe3O4 was found to be responsible for RWGS, with the metallic Fe and iron carbides activating CO to generate hydrocarbons [50].Previous works in our lab have involved synthesizing carbon nanosphere-encapsulated Fe and Fe-Co core-shell catalysts and evaluating their utility for CO2 hydrogenation [51,52].The graphitic carbon with defects in the shell and a mixture of metal oxides, metallic metal, and metal carbides in the core was evidenced by SEM, TEM, XRD, and XPS [51].The Fe3O4@carbon nanostructured catalyst enjoys the benefit of the effective diffusion of carbon atoms from the shell into the core to form a Fe5C2 carbide phase that is active for hydrogenation [40,43].
Despite being widely studied in the context of CO2 hydrogenation, Fe-based catalysts usually have low selectivity toward light olefins.The use of suitable metals/promotors to boost the yield of light olefins by adjusting electronic and structural properties has been extensively studied [53][54][55][56][57][58][59][60][61][62][63][64][65].Doping the catalyst with a second metal may improve the yield of light olefins by forming a highly active interface.Xu et al. studied the role of Fe-Co interaction in ternary ZnCo0.5Fe1.5O4catalysts and found that during the CO2 hydrogenation, Fe atoms in the Fe-Co alloy enhances the generation of active sites for production of light olefins [53].Alkali metals have often been used as promotors to modulate electronic properties [54][55][56][57][58][59][60], while metals such as Mn, Ce, and Ca have been utilized as structural promotors [61][62][63].Transition metals (e.g., Zn, Co, Cu, V, and Zr) have been used as both electronic and structural promotors [56].When modified with alkali promoters, Fe-based catalysts demonstrate higher olefin selectivity due to an increase in the adsorption of CO2 on the Fe phases and consequently suppress H2 chemisorption, which inhibits olefin re-adsorption to form alkanes, leading to higher olefin yields [60,64,65].Our previous study described the conversion of CO2 to light olefins with favorable catalytic activity through using a carbon nanosphere (CNS)-encapsulated bimetallic Fe-Co catalyst [52].
In this work, we introduce promotors M (M = Na, K, Ce, Mn, Ce, and Zn), which have been studied widely, to the core of CNS-encapsulated bimetallic Fe-Co catalysts to modify the metal species to further improve the catalytic activity of the catalysts for the hydrogenation of CO2 to light olefins.The promoted M-CNS-FeCo catalysts were synthesized using a diffusion impregnation method, and about 1 wt% of M was determined using ICP-MS analysis.The synthesized catalysts were evaluated for CO2 hydrogenation at atmospheric pressure, and testing results revealed that the addition of a promotor boosted the production of light olefins.SEM analysis, XRD analysis, TGA-DSC, and Raman spectroscopy were utilized to characterize the fresh and spent catalysts.Our results revealed that the addition of the promotor M enhanced the formation of graphitic carbon and metal carbides in the promoted catalysts when compared with the unpromoted catalysts.The metal carbides were determined to be the active sites for the production of light olefins.

Results and Discussion
2.1.Diffusion Process Used to Synthesize Promoted M-CNS-FeCo (M = Ce, Na, K, Mn, Zn) Several means were utilized to introduce the promotor to the catalysts [49,[55][56][57]59,[61][62][63][64][65].Among them, the most widely used approach is wetness impregnation method [57,59,61,62,64].However, this method was mainly used to decorate the surface of the catalysts.For the present work, we attempted to introduce the promotor M to the core of CNS to modify the bimetallic Fe-Co.The process used to prepare the promoted M-CNS-FeCo (M = Ce, Na, K, Mn, Zn) is illustrated in Scheme 1. Firstly, the prepared CNS-FeCo was mixed with saturated M salt solution and subjected to an ultrasound for 30 min to expediate the diffusion of the M salts to the core of the CNS.Secondly, we used rotovap to evaporate the solvent.During the rotary evaporation process, uniform rolling assisted the diffusion and simultaneously impregnated the promotor salts on the surface of the CNS; the last step was to remove the M salts on the surface of the CNS using a sufficient amount of Di-H2O.Metal contents for Fe, Co, and M wt% in M-CNS-FeCo were determined, and the results are shown in Table 1.Sufficient experiments were conducted to investigate the effect of the ultrasound duration, rotavap duration, water amount usage, and rinsing duration on promotor content and catalytic performance.Initially, approximately 5 wt% promoted catalysts were prepared after the series of the experiments carried out via in the above steps.The surface promotors could be washed off easily.Some of the promotors in the core could have been leached out too, but we were able to control the rinsing process to maintain approximately 1 wt% of M content in each individual promoted M-CNS-FeCo through using this synthetic approach.The consistency of the M metal content in the promoted samples indicated the validity of this synthetic process.

Catalytic Performance of Promoted M-CNS-FeCo (M = Ce, Na, K, Mn, Zn)
The above-mentioned synthesized catalysts were evaluated for CO2 hydrogenation under identical testing conditions to facilitate a direct comparison with the unpromoted CNS-FeCo.Figure 1a-e depict catalytic performance (hydrocarbon distribution, CO2 conversion, CO selectivity) as a function of temperature over promoted catalysts (Figure 1ae: Ce, K, Mn, Na, Zn, respectively) to facilitate a comparison with the unpromoted catalyst (Figure 1f).It was demonstrated that the addition of promotor M (M = Ce, K, Mn, Na, Zn) to CNS-FeCo enhanced the catalytic performance in terms of light olefins in the hydrocarbon distribution for CO2 hydrogenation at different temperatures when compared with the unpromoted catalysts.As the temperature increased from 275 to 400 °C, the CO2 conversion generally increased for all catalysts; the CO selectivity increased from 275 to 350 °C but dropped at 400 °C.Regarding the hydrocarbon distribution, the hydrocarbon contained CH4, light olefins (C2-C4 = ), light alkanes (C2-C4( 0 ), and C5+ (C5+ = , C5+ 0 , C6+ = , C6+ 0 , C7+ = , C7+ 0 ); the main product in all catalysts (promoted and unpromoted) was CH4, and it decreased as temperature increased from 275 to 350 °C and then backed up for most catalysts, while it continued to drop for the Mn-promoted catalysts.For the unprompted catalyst, almost no light olefins were produced at temperatures lower than 275 °C, but the addition of a promotor enabled light olefin production even at 275 °C.As the temperature increased from 275 to 350 °C, the amount of light olefins produced increased and peaked at 350 °C before starting to drop for most catalysts, except for the Mn-promoted catalyst, which continued to rise at 400 °C.Lower temperatures (275~325 °C) were conducive to light alkane production, especially for the Na-and K-promoted catalysts.Almost no long chains of C5+ hydrocarbons were produced for all catalysts.
The catalytic performance data of hydrocarbon distribution, CO2 conversion, and CO selectivity at 350 °C for various promoted catalysts are summarized in Figure 1g; the promotor effect could easily be perceived for the production of light olefins among the promoted and unpromoted catalysts, with Na being the most promising promoter, enhancing the production of light olefins at 350 °C.
As shown in Figure 1b,d, the catalytic performance trends for the Na-and K-promoted catalysts were similar.Alkali metals such as K and Na, as electron donors, seemingly adjusted the electronic properties of the Fe/Co sites in the core of the CNS, which facilitated the production of light olefines and inhibited secondary hydrogenation to light alkanes at 350 °C.The Mn, Ce metals, as structural promotors, decorated the structural properties of the Fe/Co sites in the core of CNS, which promoted the production of light olefins at higher temperatures (>350 °C) (Figure 1a,c).The transition metal Zn, as both an electronic and structural promotor, behaved slightly differently from the other promoted catalysts in terms of the trends for CO selectivity as the temperature changed.
A previous work showed that the CO2 conversion can be significantly enhanced with the introduction of Co into the Fe catalyst [52].The intimate contact between the Fe and Co sites favored the production of C2-C4 = .Xu et al. investigated the roles of Fe-Co interactions over ternary ZnCo0.5Fe1.5O4catalysts and unveiled that during CO2 hydrogenation, the formation of electron-rich Fe atoms in the Fe-Co alloy phase significantly boosted the generation of active metal carbides phases [53].Similarly, Chaipraditgul et al. found that the addition of the transition metal Co to Fe-based catalysts remarkably altered the interaction between the adsorptive CO2 and H2 and the surfaces of the metal catalysts [64].
With the addition of a promotor, the surface of the metal species became carbon-rich and hydrogen-poor, which facilitated C−C coupling and therefore the formation of light olefins and the suppression of CH4, as illustrated in Figure 1.Among the promotors investigated, it was found that the Na and K performed similarly with respect to enhancing the catalytic performance in terms of light olefin selectivity and CO2 conversion.Adding alkali metals (i.e., Na, K) could enhance CO2 adsorption on the more electron-rich Fe/Co phases, facilitate the generation of active sites χ-(Fe1-xCox)5C2, and suppress H2 chemisorption, which inhibits olefin re-adsorption for secondary hydrogenation to alkanes [56][57][58]63,66].As evidenced in Figure 1, the Mn-promoted catalysts performed slightly different from other promoted catalysts.The addition of Mn to the Fe-Co species in the core of the CNS might have allowed for an increase in the catalyst's basicity, which prevented the secondary hydrogenation of olefins into paraffins.Furthermore, the synergistic effects between Fe-Co and Mn could promote the production of Fe carbides and improve CO2 dissociation and the conversion of CO2 to hydrocarbons.Additionally, the addition of Mn could reduce the amount of weakly adsorbed H atoms and consequently lessen the hydrogenation aptitude, which could explain why the Mn-promoted CNS-FeCo catalyst had a higher level of light olefin production even at a higher temperature (400 °C) when compared with the other promoted and unpromoted CNS-FeCo catalysts.Similar effects were observed by Liang et al. and Jiang et al. [61,62].Ce, as a structural promoter, can restrain the growth of an Fe2O3 crystallite, weaken the interaction between Fe and cobalt, and enable and ease the reduction of iron oxide and cobalt oxides, as confirmed by the studies conducted by Zhang et al. [63].Conversely, the addition of Zn can adjust the basicity of the Fe-Co surface, increase the number of active sites for the adsorption of CO2 and H2, and ease the reducibility of metal oxides to a certain degree, as confirmed by Witoon et al.'s work [67].

Physicochemical Properties of Promoted M-CNS-FeCo (M = Ce, Na, K, Mn, Zn)
The Na-promoted catalysts were selected to investigate the morphology of the catalysts before and after CO2 hydrogenation. Figure 2a,b display the SEM images of the fresh Na-promoted catalyst before CO2 hydrogenation, and Figure 2c,d exhibit the SEM images of the spent Na-promoted catalysts after CO2 hydrogenation.More agglomerations were observed for the spent catalysts, which confirmed the deactivation of the catalysts over the time of the stream.
Figure 3 shows the XRD patterns of the fresh and spent M-CNS-FeCo catalysts with different promoters (i.e., Na, K, Ce, Mn, and Zn) at 2θ = 10.0−70.0°.The patterns could be attributed to a mixture of carbon, metal oxides, metal carbides, and metallic metals, as denoted in our XRD analysis.The broad peaks in the 10.0−20.0°range are characteristic of amorphous carbon, indicating the presence of amorphous carbon (defects) on the CNS.The intensity of these broad peaks were increased for the Na-and K-promoted catalysts, indicating a relative higher amount of amorphous carbon when compared to the Ce-, Mnand Zn-promoted catalysts.The peak at 2θ = 26.0°was indexed to the (002) plane of graphite (PDF 00-41-1487) [68].The intensity of this graphite carbon peak enhanced after the addition of the M promotors (M = Ce, K, Mn, Na, Zn), indicating that the M promotors  311), (511), and (440) facets of Fe3O4-Co3O4, respectively (PDF 00-065-0731).The peak at 2θ = 42.7°could be attributed to the (110) plane of metallic Fe-Co (PDF 00-006-0696), while the 45.5 ° peak could be attributed to the (112) plane of Fe5C2-Co5C2 (PDF 00-051-0997).In the spent M-CNS-FeCo, the broad amorphous carbon peak in the range of 10.0−20.0°increased for the Zn-and Ce-promoted catalysts, suggesting more carbon deposition on the catalysts.Interestingly, the peaks for Fe2O3-Co2O3, Fe3O4-Co3O4, and the metallic metals Fe-Co almost disappeared for the spent catalysts.On the contrary, the peak for the iron-cobalt carbide Fe5C2-Co5C2 at 45.5° drastically increased, which indicated that the metal carbides were the active sites for CO2 hydrogenation.The carbonaceous intermediates possibly further carbonized bimetallic Fe-Co, forming the Fe-Co metal carbide phase, as evidenced by the nearly complete disappearance of metal oxides.Our XRD data confirmed that the iron-cobalt species (mainly iron-cobalt carbides) encapsulated in CNS are in a more reduced state.The reduction of metal oxides and the dominant form of metal carbides in the spent catalyst resembled a favorable environment for CO2 hydrogenation, as evidenced by the catalytic performance.Compared to the unpromoted catalysts, it seemed that the M promotors played a pivotal role in promoting metal carbide formation within the core of CNS during CO2 hydrogenation.The exception to this are the Zn-promoted catalysts, which retained metal oxide species.The presence of oxide phases weakened the conversion of CO2, as confirmed by the catalytic performance (Figure 1e).Raman spectra of the fresh and spent promoted M-CNS-FeCo catalysts and the unpromoted catalyst CNS-FeCo are shown in Figure 4.The peaks at 1358 cm −1 correspond to the D-band associated with defects and edge planes.The presence of a strong D band indicates a high density of defects and porosity that could be important for reactant molecules to access the active sites in the core [51,52].The peaks at 1585 cm −1 (G-band) were ascribed to the in-plane vibrations of the E2g graphene sheet zone-center mode [69], while the peaks at 2700 cm −1 (G'-band) were attributed to the stacking order of graphene layers [70].Comparing the fresh promoted and unpromoted catalysts (Figure 4a), the fresh promoted catalysts contained higher amounts of graphitic carbon in the G band and G' band relative to the amount of amorphous carbon in the D band, indicating that the promotor M facilitated the formation of graphitic carbon, consistent with the XRD data (Figure 3a).Comparing the fresh and spent promoted catalysts, as shown in Figure 4b, there was a significant increase in the intensity of the amorphous carbon (D band) relative to the graphitic carbon (G band) for the spent catalysts, indicating that carbon deposition occurred after CO2 hydrogenation, and this was also confirmed by the XRD data (Figure 3b).The high content of graphitic carbon was consistent with the graphitic shell in the CNS, as reported in an earlier work [51].Figure 5 shows the TGA-DSC results of both the fresh and spent M-promoted catalysts.Figure 5a delineates the degradation profiles of the fresh M-promoted catalysts.Both samples displayed minimal mass variation below 650 °C, possibly attributable to thermal stable graphitic carbon, with a meager 2.5% change for all fresh catalysts.An exothermic peak was observed at around 530~560 °C for all spent catalysts, indicating that a chemical reaction involving the amorphous carbon burning occurred, with an overall weight loss of 7%.The higher amount of amorphous carbon for the spent catalysts compared to the fresh catalysts was due to the carbon deposition during CO2 hydrogenation, which is consistent with the XRD and Raman data, and reflected the deactivation of the catalyst due to the carbon deposition over the time of the stream.Overall, the TGA data demonstrated the thermal-stable nature of the CNS-encapsulated Fe-Co catalysts.

Mechanistic Insights of Promoted M-CNS-FeCo for CO2 Hydrogenation
For comparison, CNS with an empty core (metal species were completely removed from the core), a promotor diffused into the CNS with an empty core, and a promotor impregnated onto the surface of the CNS with empty core were evaluated and showed no catalytic activity for CO2 hydrogenation, as illustrated in Scheme 2. The catalytic activity for CO2 hydrogenation was observed only when a CNS-encapsulated metal species, such as Fe or Fe-Co, was added [51,52].In addition, other controlled catalysts were prepared using the incipient wetness method to introduce the promotors on the surface (shell) of CNS-FeCo; the resulting catalysts performed much worse than the unpromoted CNS-FeCo catalyst for the hydrogenation of CO2 to light olefins.Therefore, the effect of the presence of the promoter component in the carbon shell or the residue promoter in the carbon shell on the catalytic performance of promoted CNS-FeCo catalysts in the hydrogenation of CO2 could be neglected.The catalytic performance was boosted only after the promotor ions were diffused into the core of CNS to make contact with Fe-Co.These controlled experiments indicated that the metal in the core is essential and that the diffusion of the promotor into the metal core is critical to adjust the electrical and structural properties of the metals.Therefore, it is important to have three components in the catalytic composition for CO2 hydrogenation: CNS in the shell, encapsulated Fe-Co in the core, and a promotor.CNS acted as the carbon source and confined reactor, Fe-Co served as the active sites in the core, and the promotor functioned as a booster to improve the production of light olefins (Scheme 2).As evidenced by the XRD analysis, Raman spectroscopy, and TGA results, the M promotor facilitated the formation of graphitic carbon and metal carbides.
Based on the product distribution (CO, CH4, C2−C4 (=) , C2−C4 (0) , C5+), the reaction pathway over the catalyst M-CNS-FeCo follows the CO2-FTS route, which consists of two successive processes: the reverse water-gas shift (RWGS) reaction (CO2 + H2 → CO + H2O) and subsequent Fischer-Tropsch synthesis (FTS) (nCO + 2nH2 → (CH2)n + nH2O).As suggested previously [51], the mechanisms of light olefin synthesis by CO2 hydrogenation involved C−O bond cleavage and C−C bond formation [36,51].Through the CO2−FTS process over the promoted M-CNS-Fe1Co2 catalysts, the Fe2O3/Co2O3 phase was reduced by hydrogen to Fe3O4-Co3O4 as the active phase for the RWGS reaction, which was then reduced to Fe-Co in metallic states and further carbonized to form metal carbides, as confirmed by our XRD analysis.The reactant gases CO2 and H2 diffused through the porous carbon shell into the core and reacted with Fe3O4-Co3O4 to form carboxylate species (*CO2, *representing the adsorbed state) after CO2 was absorbed and stimulated on the active phases of Fe3O4-Co3O4 for RWGS.Some earlier works shed light on how the promoters affected the behavior of the ironbased catalysts [55,56].Yang et al. uncovered that, in their study, the addition of promoters altered the equilibrium between the iron oxides and iron carbides in the catalyst, which affected the CO2 conversion, facilitated the production of C2+ olefins, and inhibited the competitive methanation reaction [56].Similarly, a previous study found that the addition of a Na promoter into Fe-based catalysts boosted the adsorption of CO2, enhanced the stability of the active Fe5C2 phase, and suppressed the secondary hydrogenation of alkenes [55].Catalyst regenerability and the removal of coke comprise elements of our ongoing research and will be discussed in future work.

Synthesis of Promoted M-CNS-FeCo (M = Ce, Na, K, Mn, Zn)
The synthesis of promoted M-CNS-FeCo (M = Ce, Na, K, Mn, Zn) catalysts mainly involved two steps: the first step was to synthesize the CNS−encapsulated Fe−Co catalyst (CNS-FeCo), and the second step was to use a diffusion procedure to impregnate promotors M (Ce, Na, K, Mn, Zn) CNS-FeCo.
(1) Synthesis of CNS-FeCo A CNS-encapsulated bimetallic Fe−Co catalyst (CNS-FeCo) was synthesized following a modified previously reported method [36,51,52,[71][72][73][74][75].Specifically, the synthesis of CNS-FeCo involved three steps: The first step was the synthesis of iron polymeric complex solution (0.20 M) and cobalt polymeric complex solution (0.10 M).A total of 0.20 M of iron polymeric complex solution was prepared by adding 19.21 g of citric acid, 5.58 g of iron powder, and 100 mL of de-ionized water to a 500 mL beaker.The above-mentioned mixture was forcefully agitated in air for 48 h, resulting in a clear greenish solution.A total of 0.10 M of cobalt polymeric complex solution was prepared in a similar manner.The second step was polymerization.For this step, 12.20 g of resorcinol, 18.0 g of formaldehyde (37% in water), and 100 mL of 0.20 M of the earlier-prepared iron solution and 40 mL of 0.10 M of the earlier-prepared cobalt solutions were added to a 500 mL round-bottom three-neck flask.The resulting mixture was agitated until resorcinol was dissolved, followed by adding ammonium hydroxide (28% in water) dropwise.The final pH of the mixture reached 10.The resulting slurry was then aged at 85 °C for three hours, with the resulting solid being collected via vacuum filtration and dried at 65 °C in a vacuum oven overnight.The third step was carbonization.For this step, the collected polymer was heated at 85 °C for two hours to remove moisture and then 850 °C for 5 h.The resulting sample is referred to as CNS-FeCo in this study.
(2) Synthesis of M-CNS-FeCo (M = Ce, Na, K, Mn, Zn) Diffusion was utilized to impregnate promotors to CNS−FeCo.For this, in a 100 mL round-bottom flask, 1.0 g of the earlier-carbonized product was added to metal salts 4.0 g of KNO3, 3.4 g of NaNO3, 11.9 g of Zn(NO3)2•6H2O, 17.4 g of Ce(NO3)3•6H2O, and 10.0 g of Mn(NO3)2•4H2O in 20 mL of water, respectively.The resulting mixture was subjected to an ultrasound for 30 min to allow the metal salts to diffuse into the core of the CNS.The vessel was then loaded onto the Buchi ® Rotavapor ® R-3 evaporator (New Castle, DE, USA) to evaporate at 60 °C until visibly dry.The product was washed with an ample amount of deionized water to remove the metal salts attached to the surface of the CNS.The resulting solid was dried and calcined at 350 °C for five hours.The calcined samples were then grounded into a powder and pelleted to a 40-60 mesh size prior to catalyst testing.

Catalyst Evaluation
The synthesized catalysts were tested using a standard procedure reported in a previously published work [48].Further details on the testing procedure and calculation equations are provided in the Supplementary Materials.

Catalyst Characterization
Metal analyses for Fe, Co were carried out in a manner that has been described in detail previously [51].Metal analyses for K, Na, Mn, Ce, and Zn were carried out in a

Figure 2 .
Figure 2. SEM images of fresh Na-CNS-FeCo catalyst before CO2 hydrogenation (a,b) and spent Na-CNS-FeCo catalyst after CO2 hydrogenation (c,d).might have facilitated the conversion of the amorphous carbon into graphitic carbon.The diffraction peaks at 2θ = 18.5, 30.4,35.8, and 53.8° correspond to the (111), (220), (311), and (422) crystal facets of Fe2O3-Co2O3, respectively (PDF 00-039-1346).The peaks at 2θ = 30.2,35.5, 57.3, and 63.0° were assigned to the (220), (311), (511), and (440) facets of Fe3O4-Co3O4, respectively (PDF 00-065-0731).The peak at 2θ = 42.7°could be attributed to the (110) plane of metallic Fe-Co (PDF 00-006-0696), while the 45.5 ° peak could be attributed to the (112) plane of Fe5C2-Co5C2 (PDF 00-051-0997).In the spent M-CNS-FeCo, the broad amorphous carbon peak in the range of 10.0−20.0°increased for the Zn-and Ce-promoted catalysts, suggesting more carbon deposition on the catalysts.Interestingly, the peaks for Fe2O3-Co2O3, Fe3O4-Co3O4, and the metallic metals Fe-Co almost disappeared for the spent catalysts.On the contrary, the peak for the iron-cobalt carbide Fe5C2-Co5C2 at 45.5° drastically increased, which indicated that the metal carbides were the active sites for CO2 hydrogenation.The carbonaceous intermediates possibly further carbonized bimetallic Fe-Co, forming the Fe-Co metal carbide phase, as evidenced by the nearly complete disappearance of metal oxides.Our XRD data confirmed that the iron-cobalt species (mainly iron-cobalt carbides) encapsulated in CNS are in a more reduced state.The reduction of metal oxides and the dominant form of metal carbides in the spent catalyst resembled a favorable environment for CO2 hydrogenation, as evidenced by the catalytic performance.Compared to the unpromoted catalysts, it seemed that the M promotors played a pivotal role in promoting metal carbide formation within the core of CNS during CO2 hydrogenation.The exception to this are the Zn-promoted catalysts, which retained metal oxide species.The presence of oxide phases weakened the conversion of CO2, as confirmed by the catalytic performance (Figure1e).
The adsorbed H then hydrogenated the resulting *CO2 to form an *HOCO intermediate that then dissociated into *CO and *OH, with *CO subsequently dissociating into *O and *C.The produced *C could diffuse into the metal Fe-Co to form Fe-Co carbides, the active phase for the Fisher-Tropsh synthesis reaction as confirmed by the XRD studies performed for this work.The C* on the surface of the metal carbides was then hydrogenated to CHx* species.CHx* + CHx* and C* + CHx* are the most likely reaction pathways for C−C coupling [5,71].