Impact of Ni particle size on CO₂ activation and CO formation during reforming process: A density functional theory study

Abstract

In recent years, the dry reforming of methane (DRM) reaction has gained widespread attention due to its effective utilization of two major greenhouse gases. Supported Ni-based catalysts for DRM exhibit a strong dependence on particle size, however, the reaction mechanisms involved remain unclear. In this work, the effect of metal particle size on CO₂ activation and CO formation was explored in the DRM reaction using the density functional theory. Nix/MgO (x = 13, 25, 37) was constructed to investigate the CO₂ activation and the formation of CO during the DRM reaction. It is found that CO₂ is more inclined to undergo chemisorption on Nix/MgO before activation. With the variation in particle size, the main activation pathway of CO₂ on the catalyst changes. On the smallest Ni13/MgO, CO₂ tends to directly dissociate, while on the larger Ni25/MgO and Ni37/MgO, the hydrogenation dissociation of CO₂ is more kinetically favorable. Compared to Ni13/MgO and Ni37/MgO, the oxidation of surface C atoms and the oxidation of CH occur more readily on Ni25/MgO. This indicates that C atoms are less likely to form on Ni25 particle and are more easily to be oxidized. To some extent, the results suggest that Ni25/MgO exhibits superior resistance to carbon formation.

Graphical abstract

Keywords

particle size / Ni/MgO catalyst / CO₂ activation / CO formation

1 Introduction

Global climate change represents one of the most severe environmental challenges the world is currently facing, with greenhouse gas emissions being the primary driver behind this phenomenon [1]. Carbon dioxide (CO₂) and methane (CH₄) are the two predominant greenhouse gases, accounting for a significant portion of total emissions. The rapid development of the global economy has led to a continuous increase in the emission of greenhouse gases, exerting substantial pressure on the environment and climate, thereby necessitating the urgent identification of effective emission reduction strategies. In response to the escalating environmental crisis, the international community has reached a consensus, with numerous countries signing the Paris Agreement, committing to various measures aimed at reducing greenhouse gas emissions in order to curb the trend of global warming [2,3]. In China, there has also been an active response to the challenges posed by global climate change, with goals set for “carbon peak” and “carbon neutrality” [4]. Efforts are being made to reduce CO₂ emissions through various measures, including enhancing energy efficiency, developing clean energy, and promoting low-carbon technologies [5–8]. Dry reforming of methane technology, as an effective method for converting CO₂ and CH₄ into useful chemicals (syngas), plays a significant role in reducing greenhouse gas emissions and achieving the circular utilization of carbon resources [9–13].

Precious metal catalysts exhibit commendable activity in the methane CO₂ reforming process [14–16]. However, considering the cost and production limitations, the use of precious metal catalysts is not feasible for large-scale industrial processes. Nickel, not only demonstrates catalytic activity comparable to that of precious metals but also has a significantly higher natural abundance and lower cost, increasingly attracting researchers’ attention [17]. Supported catalysts show superior catalytic performance compared to standalone active metal catalysts [18,19]. The support is another crucial component of supported catalysts. On the one hand, the support provides a large specific surface area for the dispersion of active metals, thus exposing more active sites and inhibiting metal agglomeration. On the other hand, the support can interact with the active metal, potentially altering the structure of catalyst and thereby affecting its catalytic performance [20,21]. Given that CO₂ is an acidic gas, catalysts supported on basic supports generally exhibit superior catalytic activity. Among common supports, MgO possesses a strong Lewis basicity, enhancing the chemical adsorption of CO₂. Therefore, researchers often choose MgO as the support for catalysts in their studies [22,23].

Coke is a primary cause of deactivation in Ni-based catalysts [24,25]. Researchers have identified that the sources of carbon deposition on catalysts stem from the deep cracking of methane and the disproportionation of CO [9,26,27]. However, the disproportionation reaction of CO is suppressed within the temperature range of the reaction, and thus most carbon deposition derives from the deep cracking of methane. The particle size of the active metal in Ni-based catalysts significantly influences the activity and stability of the catalyst. Extensive experimental studies have demonstrated that excellent activity and resistance to carbon deposition is exhibited by the catalysts with small active metal particles [28,29]. In their experiments, Lercher et al. [30] observed optimal performance of the catalyst against carbon deposition when the size of the active metal particles reached 2 nm on Ni/ZrO₂ catalysts. Zhang et al. [31] observed that achieving the best performance in Ni–Co/MgO catalysts required a Ni–Co size below 10 nm, thereby establishing 10 nm as the critical size in this system. Observations show that the catalyst consistently displays a superior resistance to coking when the active metal reaches a critical size. The critical size varies among different systems and reaction conditions but typically ranges from a few nanometers to around a dozen nanometers [32]. However, the size effect is widely observed in experiments, and its reaction mechanism remains unclear. Guo et al. [33] constructed catalyst models with Ni4, Ni8, and Ni12 active metal clusters supported on the MgO(100) surface and calculated the DRM reaction process on Ni particles of different sizes. The results indicated that on supported Ni catalysts, a smaller Ni particle is more efficient for the DRM reforming reaction. Researchers believe that the smaller the metal particle size, the higher the proportion of highly active Ni exposed on the catalyst, thereby enhancing the activity and stability of the catalyst. Although Guo et al. [33] conducted DRM-related studies on metal particles of different sizes, the models they constructed are focused on cluster particles, with no significant variation in size. Chen et al. [34] recently reported on the mechanism of CH₄ activation and C atom growth on the Nix/MgO surface, finding that particle size has a notable impact on CH₄ activation.

The activation of CO₂ and the generation of CO were not considered in previous work. Therefore, in this work, the DFT theory are utilized to thoroughly investigate the relationship between the variation in metal particle size and both CO₂ activation and CO generation. By comparing the adsorption energies of various species and the energy barriers along the reaction pathways, the impact of changes in Ni particle size on the CO₂ activation pathways and the CO generation reaction have been explored. This work could provide theoretical guidance for the preparation of catalysts for highly efficient Ni reforming reaction with an optimal chemical performance.

2 Computational details and models

2.1 Computational method

All density functional theories are calculated using the Dmol3 module in the Materials Studio software [35,36]. The electron exchange and correlation approach adopts the generalized gradient approximation (GGA) alongside the Perdue-Burke-Enzerhof (PBE) exchange-correlation functionals [37–39]. To achieve an optimal balance between computational accuracy and efficiency, the DFT half-core pseudopotential (DSPP) method is applied. For the convergence of energy, the self-consistent field (SCF), the maximum displacement, and the maximum force, the thresholds are respectively established at 2.0 × 10⁻⁵ Ha, 1 × 10⁻⁵ Ha, 5.0 × 10⁻³ Å, and 4.0 × 10⁻³ Ha/Å. To identify the transition state of reaction, the complete linear/quadratic synchronous transit (LST/QST) method is utilized. Considering the magnetic nature of Ni, calculations incorporate electron spin polarization. The smearing is adjusted to 0.05 Ha and vacuum layer is set to 20 Å.

The adsorption energy (E_ads), energy barrier (E_b), and reaction energy (ΔH) are obtained from

(1)$E_{\mathrm{a}\mathrm{d}\mathrm{s}}=E_{\mathrm{a}\mathrm{d}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{t}\mathrm{e}/\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{f}}-E_{\mathrm{a}\mathrm{d}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{t}\mathrm{e}}-E_{\mathrm{s}\mathrm{u}\mathrm{r}\mathrm{f}},$

(2)$E_{\mathrm{b}}=E^{\mathrm{T}\mathrm{S}}-E^{\mathrm{I}\mathrm{S}},\mathrm{\Delta }H=E^{\mathrm{F}\mathrm{S}}-E^{\mathrm{I}\mathrm{S}},$

where E_{adsorbate/surf}, E_adsorbate, E_surf, E^IS, E^TS, and E^FS represent the electronic energy of the adsorption system, adsorbate alone, the clear metal surface, initial state (IS), transition state (TS), and final state (FS), respectively.

2.2 Structure model

Typically, the active metal particles of supported metal catalysts are adsorbed onto the support in a hemispherical shape [40]. Nix (x = 13, 25, 37) are constructed. A three-layer MgO(100) slab is used as the support [41]. The geometry optimization is calculated by set 3 × 3 × 1 k-point.

Research indicates that the interaction between the active metal and the support significantly alters the configuration of active metal clusters. Therefore, after loading the constructed active metal cluster particles onto the optimized MgO(100) surface, it is necessary to perform a comprehensive optimization of the entire catalyst model. During the optimization process of the catalyst model, the entire model is allowed to relax to ensure its rationality. In the following calculation process, MgO(100) is kept fixed. Fig.1 is the optimized catalyst model.

Fig.1 Models of Nix/MgO (x = 13, 25, 37) (Blue, Ni atoms; Red, O atoms; Green, Mg atoms) (adapted with permission from Chen et al. [34], copyright 2024, Royal Society of Chemistry).

Full size|PPT slide

3 Results and discussion

3.1 Adsorption of CO₂, CO and intermediates

During the DRM process, there are many important intermediates, such as COOH and CHO. COOH serves as a critical intermediate in the hydrogen-assisted dissociation pathway of CO₂, playing a vital role in the study of CO₂ activation pathways. CHO, an intermediate formed by the reaction of CH generated from the cracking of CH₄ with O atoms produced from the dissociation of CO₂, also contributes to the formation of products CO and H through its dissociation. Moreover, the formation of CHO competes with the cracking reaction of CH, potentially hindering the formation of carbon atoms on the catalyst surface, making CHO an extremely important intermediate as well. This section delves into the adsorption behavior of CO₂, CHO, COOH, O, and CO.

Fig.2 illustrates the adsorption configurations of CO₂, intermediates, and products on Nix/MgO. Consistent with previous literature reports [42], CO₂ is found to preferentially adsorb at the interface between the active metal and the support on Nix/MgO. Studies have shown that free CO₂ molecules possess a linear structure with a bond length of 1.16 Å [43]. Upon adsorption on the Nix/MgO surface, CO₂ molecules undergo significant deformation, exhibiting a bent structure with altered C−O bond lengths, which are not identical within the molecule. On Ni13/MgO, the C−O bond lengths on the side closer to and further from the support are 1.33 and 1.26 Å, respectively; on Ni25/MgO, they are 1.32 and 1.28 Å; and on Ni37/MgO, they are 1.31 and 1.26 Å. On all three surfaces, the C−O bond closer to the support is longer than the other. This suggests that the MgO support plays a role in the adsorption of CO₂. For the intermediate COOH, it adsorbs perpendicularly on the top site of the first layer of Ni atoms on Ni13/MgO, while on Ni25/MgO and Ni37/MgO, it adsorbs at the interface between the active molecule and the support. On Ni25/MgO, COOH adsorbs vertically, whereas on Ni37/MgO, its adsorption is offset. For the intermediate CHO, its optimal adsorption site is on the top site of the topmost Ni atoms on Ni13/MgO, in the bridge site between the middle and bottom Ni atoms on Ni25/MgO, and most stably at the interface between the active component and the support on Ni37/MgO. The adsorption configurations of O atoms and CO molecules are relatively similar across the three catalyst surfaces. On Nix/MgO, O atoms adsorb in the hollow sites of Ni atoms in different layers, while CO adsorbs at the interface between the active metal and the support. Research findings indicate that variations in the size of active metal particles have an impact on the optimal adsorption sites of adsorbates on the catalyst surface, with the adsorption of intermediates COOH and CHO being most significantly affected.

Fig.2 Configurations of CO2, CO, and intermediates adsorption on Nix/MgO.

Full size|PPT slide

The adsorption energy and charge transfer of various adsorbates on Nix/MgO are further investigated. Tab.1 records the adsorption energy and charge transfer for CO₂, intermediates, and products adsorbed on Nix/MgO. It demonstrates that adsorbates always gain charge from the catalyst surface for stable adsorption, and on the same catalyst surface, a greater amount of charge transfer correlates with a higher adsorption energy. Fig.3 further illustrates the impact of different sized Ni particles on Nix/MgO on the adsorption energy of adsorbates. It was found that the adsorption energies of CO₂ on Nix/MgO (x = 13, 25, 37) are −1.17, −1.37, and −1.22 eV. After adsorption on the Nix/MgO surface, CO₂ undergoes significant configurational changes, and the calculated adsorption energies are substantial, indicating that CO₂ adsorption on all three Nix/MgO surfaces is chemisorbed. CO₂ exhibits the strongest adsorption capability on Ni25/MgO, followed by Ni13/MgO, and the weakest on Ni37/MgO. For intermediates COOH and CHO, both exhibit the highest adsorption energy on Ni25/MgO, with the lowest on Ni13/MgO. For O atoms and CO molecules, their adsorption energies on the three different catalyst surfaces are similar. Notably, by comparing the adsorption configurations of O atoms and CO molecules on Nix/MgO surfaces, it was found that their adsorption configurations are also similar. This suggests that on Nix/MgO, the same adsorbate adsorbing at similar adsorption sites will have comparable adsorption energies. The different sizes of active metal particles on the catalyst primarily alter the optimal adsorption sites of adsorbates on the surface, thereby affecting the charge transfer and adsorption energy.

Tab.1 Adsorption energy and charge transfer of CO2, CO, and intermediates on Nix/MgO

	Ni13/MgO			Ni25/MgO			Ni37/MgO
	Eabs/eV	Charge/e		Eabs/eV	Charge/e		Eabs/eV	Charge/e
CO2	−1.17	0.16		−1.27	0.18		−1.01	0.14
COOH	−1.44	0.29		−1.89	0.22		−1.54	0.21
CHO	−1.25	0.20		−1.40	0.26		−1.39	0.23
O	−2.50	0.39		−2.50	0.35		−2.49	0.33
CO	−1.59	0.25		−1.60	0.25		−1.54	0.23

Fig.3 Adsorption energy of CO2, CO, and intermediates on Nix/MgO.

Full size|PPT slide

3.2 Activation pathway of CO₂

Generally speaking, during the DRM process, CO₂ can undergo two possible activation pathways [44–46]: (1) Direct dissociation of CO₂ (CO₂ → CO + O), O subsequently participating in the oxidation of surface C atoms or CH oxidation reactions; (2) H-assisted dissociation of CO₂, where CO₂ reacts with dissociatively generated H atoms from CH₄ to form COOH (CO₂ + H → COOH), which then dissociate to yield OH and the product CO (COOH → CO + OH). The mode of CO₂ dissociation varies across different catalyst systems. This section investigates, in detail, the two activation pathways of CO₂ on Nix/MgO, comparing and analyzing the impact of variations in the size of active particles on the CO₂ activation pathways.

Fig.4 shows the initial, transition, and FSs of CO₂ direct dissociation on the surface of Nix/MgO. The results indicate that the direct dissociation of CO₂ occurs at the interface between the active component of the catalyst and the support. On the Ni13/MgO surface, during the activation process of adsorbed CO₂, the C−O bond farthest from the support breaks, with the distance between the O and C atoms stretching to 2.40 Å in the TS. Ultimately, the O atom is stably adsorbed in the hollow site of Ni atoms on the active metal side, while CO is adsorbed at the interface, with the detached O atom being 3.47 Å away from the C atom. On Ni25/MgO and Ni37/MgO, the process of direct dissociation of CO₂ is similar to that on Ni13/MgO, but the distances between the dissociated O and C atoms in the transition and FSs differ. On Ni25/MgO, the distance between the O and C atoms is 1.92 Å in the TS and 3.14 Å in the FS; on Ni37/MgO, these distances are 1.85 Å and 3.77 Å in transition and FSs. Comparative analysis reveals that the direct dissociation process of CO₂ on Nix/MgO is not significantly different, but the distances of atomic migration in the TS vary, being greatest on Ni13/MgO and similar on Ni25/MgO and Ni37/MgO. This suggests that the energy barrier for the direct dissociation process is highest on Ni13/MgO.

Fig.4 Reaction process of direct dissociation of CO2 on Nix/MgO.

Full size|PPT slide

The energy barriers and reaction heats for the direct dissociation of CO₂ on Nix/MgO are presented in Tab.2. The results indicate that the energy barrier for direct CO₂ dissociation is highest on Ni13/MgO, with very similar values observed for Ni25/MgO and Ni37/MgO. This suggests, the direct dissociation of CO₂ is most difficult to occur from the kinetic aspect on Ni13/MgO, with the lowest catalytic activation for CO₂ dissociation observed. On catalysts with larger active metal particle sizes, the direct dissociation of CO₂ occurs more readily. This process is endothermic on both Ni13/MgO and Ni25/MgO surfaces, requiring more heat absorption on Ni25/MgO, while it is a mildly exothermic reaction on the Ni37/MgO surface.

Tab.2 Eb and ΔH for direct dissociation of CO2 on Nix/MgO surfaces

	Eb/eV	ΔH/eV
Ni13/MgO	1.40	0.08
Ni25/MgO	1.22	0.33
Ni37/MgO	1.24	−0.02

Fig.5 illustrates the initial, transition, and FSs of CO₂ undergoing hydrogenative dissociation on Nix/MgO surfaces. On Ni13/MgO, the specific process of CO₂ hydrogenative dissociation involves CO₂ adsorption at the interface between the active component and the support, with H atoms generated from CH₄ dissociation adsorbing at the bridge sites of the top-layer Ni atoms, forming a stable co-adsorption configuration. Initially, the distance between the H atom and the closest O atom in the CO₂ molecule is 4.27 Å. During activation, the O atom and the CO₂ molecule move upwards and gradually approach each other, with their distance reducing to 2.81 Å in TS1. Subsequently, the H atom bonds with CO₂ to form the intermediate COOH, where the O−H bond length is 1.00 Å. COOH is adsorbed perpendicularly to the support on the top site of the active metal Ni atom, with the C atom facing the metal particle and the O and H atoms facing the opposite side. Following this, COOH undergoes a dissociative activation, breaking the C−O bond. In TS2, the distance between the C atom in CO and the O atom in OH is 2.10 Å. In the FS, the produced CO is adsorbed at the interface between the active component and the support, and the generated OH is adsorbed on the top site of the top-layer Ni atom, forming a stable co-adsorption configuration with a C−O atom distance of 4.26 Å.

Fig.5 Reaction process of hydrogen-assisted dissociation of CO2 on Nix/MgO.

Full size|PPT slide

On Ni25/MgO, the stable co-adsorption configuration of CO₂ with H atoms involves CO₂ adsorbed at the interface, with H atoms adsorbed in the hollow sites of active metal Ni atoms. The distance between the H atom and the O atom in CO₂ closest to H is 2.84 Å. During the reaction, the H atom approaches the O atom in CO₂, with the distance reducing to 1.47 Å in TS1. Afterwards, the H atom bonds with the O atom in CO₂ to form COOH, with an O−H bond length of 1.00 Å. Unlike the adsorption configuration of COOH on Ni13/MgO, COOH is adsorbed at the interface between the active component and the support on Ni25/MgO, with the C atom facing the active metal Ni side, and the O and H atoms facing the opposite side. During the formation of CO from the intermediate COOH, the C−O bond breaks, with the distance between the C and O atoms being 2.80 Å in TS2. In the FS, this distance becomes 3.77 Å, with CO adsorbed at the interface between the bottom Ni atoms and the support, and OH adsorbed on the metal Ni particle.

The process of CO₂ hydrogenative dissociation on Ni37/MgO is similar to that on Ni25/MgO. Initially, the distance between the H atom and the O atom in CO₂ closest to H is 3.40 Å, reducing to 1.89 Å in TS1, and after forming COOH, the O−H bond is 1.01 Å. During COOH dissociation, the C−O bond breaks, with the C and O atom distance being 2.06 Å in TS2, and 3.31 Å in the FS.

Comparative analysis reveals that variations in metal particle size lead to changes in the active sites where reactions occur. On Ni13/MgO, both the formation of COOH from CO₂ and H atoms, and the dissociation of COOH, take place on the active metal particles, whereas on the other two catalyst surfaces (Ni25/MgO, Ni37/MgO), these processes occur at the interface between the active metal and the support.

Tab.3 presents the energy barriers and reaction heats for the hydrogenative dissociation of CO₂ on Nix/MgO surfaces. Among these, the process on the Ni13/MgO surface exhibits the highest energy barrier and reaction heat, significantly exceeding the corresponding values on the other two surfaces, with the lowest energy barrier observed on the Ni25/MgO surface. The study results demonstrate that kinetically and thermodynamically, this reaction is most challenging to occur on the Ni13/MgO surface, while it proceeds most readily on Ni25/MgO. The reaction is endothermic on all three surfaces. Therefore, the intermediate COOH is less likely to form on the Ni13/MgO surface, whereas it is one of the main intermediates on the Ni25/MgO and Ni37/MgO surfaces. The dissociation of the intermediate COOH occurs very easily on all three surfaces, with little difference in the energy barriers. This indicates that the intermediate COOH is unlikely to remain stable on the surface, and the formed COOH readily dissociates into CO molecules and OH. On the Ni13/MgO and Ni37/MgO surfaces, this reaction is exothermic, whereas it is endothermic on the Ni25/MgO surface.

Tab.3 Eb and ΔH for hydrogen-assisted dissociation of CO2 on Nix/MgO

	CO2+H→COOH			COOH→CO+OH
	Eb/eV	ΔH/eV		Eb/eV	ΔH/eV
Ni13/MgO	1.72	0.61		0.73	−0.32
Ni25/MgO	0.93	0.20		0.72	0.07
Ni37/MgO	1.18	0.13		0.75	−0.34

Fig.6 illustrates the energy barrier for different CO₂ activation pathways on Nix/MgO surfaces. The study reveals that on the Ni13/MgO surface, the energy barrier for the reaction of CO₂ with H atoms to form COOH is significantly higher than that for direct dissociation. Although the subsequent dissociation barrier of COOH is lower, the overall pathway of CO₂ hydrogenative dissociation remains more challenging compared to direct dissociation. Thus, direct dissociation emerges as the predominant reaction pathway on Ni13/MgO. On the Ni25/MgO surface, in contrast to the direct dissociation of CO₂, the energy barrier for hydrogenative dissociation is lower, indicating that CO₂ hydrogenative dissociation occurs more readily on Ni25/MgO. Similarly, on Ni37/MgO, the energy barrier for CO₂ hydrogenative dissociation is also slightly lower than that for direct dissociation, making hydrogenative dissociation relatively easier to occur. In summary, the activation pathway of CO₂ on Ni13/MgO is primarily direct dissociation, while on Ni25/MgO and Ni37/MgO, CO₂ tends to undergo hydrogenative dissociation, making it the main activation pathway. The variation in the size of active metal Ni particles on the MgO support significantly influences the activation pathway of CO₂, leading to a change in the predominant mode of activation.

Fig.6 Energy curves of different activation paths of CO2 on Nix/MgO surfaces.

Full size|PPT slide

3.3 Formation of CO

There are several sources of CO formation in DRM reactions, such as activation dissociation of CO₂, oxidation of surface C atoms, and dissociation of CHO species. Activation dissociation of CO₂ has been discussed in detail in Section 3.2, and this section focuses on exploring the other two reactions.

Fig.7 displays the structures of the IS, TS, and FS for the oxidation of surface carbon atoms. On Ni13/MgO, the carbon atom is adsorbed in the hollow site of the topmost Ni atoms, and the hydrogen atom is adsorbed at the hollow site of the side Ni atoms. In the IS, the distance between the carbon and oxygen atoms is 3.69 Å. During the reaction, the carbon and oxygen atoms move closer to each other, reaching a distance of 2.05 Å in the TS. The reaction proceeds to form a bond between the carbon and oxygen atoms, with a bond length of 1.21 Å in the FS. The CO molecule is adsorbed at the bridge site of the topmost Ni atoms, with the carbon atom closer to the Ni particle surface and the oxygen atom further from the surface. On Ni25/MgO, the most stable co-adsorption configuration has both the carbon and oxygen atoms adsorbed at the hollow sites on the side of the Ni particles, with an interatomic distance of 4.38 Å. During the reaction, the atoms move downwards and closer to each other, with the distance between them becoming 2.01 Å in the TS. In the FS, the formed CO molecule is adsorbed at the interface between the active metal and the support, with the carbon atom closer to the particle surface. The formed C−O bond length is 1.23 Å. On Ni37/MgO, in the IS, the carbon atom is adsorbed at the hollow site on the side of the Ni particles, and the oxygen atom is adsorbed at the bridge site, forming a stable co-adsorption configuration with an interatomic distance of 4.55 Å. During the reaction, the carbon atom moves downwards and the oxygen atom moves closer to the carbon atom, with the distance between the two atoms being 2.00 Å in the TS. In the FS, the formed CO molecule is also adsorbed at the interface between the carrier and the metal particles, with a bond length of 1.23 Å.

Fig.7 Reaction process of oxidation of C on Nix/MgO.

Full size|PPT slide

Tab.4 presents the energy barriers and reaction heats for the oxidation of carbon atoms on Nix/MgO. The results indicate that, kinetically, the reaction is most favorable on Ni25/MgO, followed by Ni13/MgO, and is least favorable on Ni37/MgO. Thermodynamically, the reaction is endothermic on Ni13/MgO, while it is exothermic on Ni25/MgO and Ni37/MgO. Comparative analysis reveals that the oxidation of carbon atoms on Ni25/MgO is the most favorable both kinetically and thermodynamically, making it the most likely to occur.

Tab.4 Eb and ΔH for oxidation of C atoms on Nix/MgO surfaces

	Eb/eV	ΔH/eV
Ni13/MgO	1.52	0.09
Ni25/MgO	1.10	−1.31
Ni37/MgO	1.63	−0.83

CHO specie is important intermediates in DRM reactions, and the origin is generated by the reaction of CH generated from CH₄ dissociation with CO dissociation to generate O atoms. It is worth mentioning that this reaction is in competition with the reaction of CH depth dissociation to generate surface C atoms. Therefore, the study of CHO species in dry reforming is important.

Fig.8 illustrates the initial, transition, and FSs of the formation and dissociation reactions of CHO on Nix/MgO surfaces. On Ni13/MgO, CH and O are co-adsorbed in the hollow sites on the side of the metal particles, forming a stable co-adsorption configuration, with the oxygen atom 3.61 Å away from the carbon atom in CH. During the reaction, CH moves upwards while the oxygen atom moves toward CH, reaching a distance of 2.00 Å in the TS. In the FS, a C−O bond is formed with a bond length of 1.25 Å, and the resulting CHO is adsorbed at the bridge site of the topmost Ni atoms, with the carbon atom closer to the Ni particle surface and the hydrogen and oxygen atoms further from the surface. During the CHO dissociation process, the C−H bond breaks, with the distance between C and H increasing from 1.13 Å in the IS to 1.58 Å in the TS. The hydrogen moves to the top site of the topmost Ni atoms while CH moves downwards, resulting in a distance of 3.16 Å between C and H in the FS. The generated hydrogen atom and CO molecule are finally adsorbed at different bridge sites between the topmost Ni atoms.

Fig.8 Reaction process of CHO formation and dissociation on Nix/MgO.

Full size|PPT slide

On Ni25/MgO, CH and the oxygen atom are adsorbed on the same side of the metal particle surface, with C and O 4.06 Å apart. During the reaction, the oxygen atom gradually approaches CH, reaching a distance of 1.90 Å in the TS. Subsequently, C and O form a bond with a bond length of 1.27 Å, and the resulting CHO is adsorbed at the interface between the active metal particle and the carrier. During the dissociation process of CHO, the C−H bond breaks, and the hydrogen atom moves toward another surface of the metal particle. The distance between the two atoms changes from 1.13 initially to 1.59 Å in the TS. In the FS, the generated CO molecule is adsorbed at the bridge site of Ni atoms on one side of the metal particle, while the hydrogen atom is adsorbed at the hollow site on the other side, ultimately 3.38 Å apart.

On Ni37/MgO, the specific process of CHO formation and dissociation involves CH being adsorbed at the hollow site on the side of Ni particles, with the oxygen atom adsorbed at a hollow site on an adjacent side. In the co-adsorbed configuration, C and O are 3.09 Å apart. During the reaction, CH moves downwards, and the oxygen atom moves toward CH, with the distance between C and O atoms being 1.35 Å in the TS. Subsequently, a C−O bond is formed, and the resulting CHO is adsorbed at the interface between the metal particle and the carrier, with a C−O bond length of 1.35 Å and a C−H bond length of 1.12 Å. During the CHO dissociation reaction, the C−H bond breaks, with the distance between C and H being 1.63 Å in the TS. The hydrogen atom moves upwards, and in the FS, a stable co-adsorption configuration is formed with the CO molecule adsorbed at the interface, the carbon atom closer to the Ni particle surface, the oxygen atom further from the particle surface, and the hydrogen adsorbed at the bridge site between the bottom and top layer Ni atoms, with a distance of 2.84 Å between them.

Tab.5 documents the energy barriers and reaction heats for the formation and dissociation processes of CHO. The research results indicate that on Ni13/MgO, the energy barrier for the formation of CHO is the highest, while on the other two surfaces, this reaction barrier is significantly lower, with the lowest observed on Ni25/MgO. This suggests that kinetically, the reaction is most favorable on Ni25/MgO and most challenging on Ni13/MgO. Furthermore, the reaction is exothermic on three catalyst surfaces, indicating it is thermodynamically feasible. The results demonstrate that CHO becomes one of the main intermediates on the surface of Ni37/MgO due to its easier formation and more challenging dissociation.

Tab.5 Eb and ΔH for CHO formation and dissociation on Nix/MgO surfaces

	CH+O→CHO			CHO→CO+H
	Eb/eV	ΔH/eV		Eb/eV	ΔH/eV
Ni13/MgO	1.37	−0.35		0.57	−0.07
Ni25/MgO	0.61	−0.39		0.59	−0.01
Ni37/MgO	0.78	−0.40		0.88	0.18

Fig.9 depicts the energy variation for the aforementioned processes. The research findings indicate that the optimal reaction pathway for C+O occurs on Ni25/MgO, suggesting that this reaction is more readily facilitated on its surface, and even a minimal generation of C atoms on the Ni particle surface can be quickly oxidized by O atoms. The optimal pathway for the CH + O reaction is also found on Ni25/MgO, demonstrating that the intermediate CHO is more likely to form on its surface. This reaction competes with the CH dissociation studied in previous work [35], with the energy barrier for CHO formation being lower than that for CH dissociation. CH on Ni25/MgO is more inclined to bond with O atoms other than dissociating into C atoms, which to some extent indicates that this catalyst has strong resistance to carbon buildup. Regarding the CHO dissociation reaction, the energy barriers on all three surfaces are below 1.0 eV, making this reaction relatively easy to occur. The dissociation of CHO also serves as a principal source of the product CO.

Fig.9 Energy variation. (a) C oxidation; (b) CHO production and dissociation.

Full size|PPT slide

4 Conclusions

In this work, the mechanisms of CO₂ activation and the CO formation on Nix/MgO surfaces are explored. The impact of Ni particle size is examined from the adsorption of intermediates and products, and the formation pathways of reactants. The following is a summary of the key findings:

(1) The adsorption sites for CO₂, CHO and CO on different Nix/MgO surfaces are essentially the same, with similar corresponding adsorption energies. These species are chemisorbed on the Nix/MgO surface. The adsorption of COOH is significantly influenced by changes in the catalyst surface particle size, with its adsorption site transitioning from the metal particles to the interface between the particles and the support as the particle size increases.

(2) The primary activation pathway of CO₂ on the catalyst surface changes with the variation in Ni particle size. On Ni13/MgO, the direct dissociation of CO₂ into CO molecules and O atoms occurs more readily, with direct dissociation being the main pathway of activation. However, as the particle size increases, on Ni25/MgO and Ni37/MgO, the hydrogenation dissociation of CO₂ exhibits lower reaction barriers, leading to a shift in the main activation pathway toward hydrogenation dissociation.

(3) The oxidation reaction of surface C atoms exhibits the lowest reaction barrier (1.10 eV) on Ni25/MgO, with higher energy barriers observed on both Ni13/MgO and Ni37/MgO surfaces, making the oxidation of C atoms more challenging on the surfaces of Ni13/MgO and Ni37/MgO. The oxidation of CH occurs more readily on the Ni25/MgO surface, where CH is more inclined to combine with O atom rather than dissociating into C atom.

References

[1] Soeder D J. Greenhouse gas sources and mitigation strategies from a geosciences perspective. Advances in Geo-Energy Research, 2021, 5(3): 274–285

[2] Keohane R O, Victor D G. Cooperation and discord in global climate policy. Nature Climate Change, 2016, 6(6): 570–575

[3] Barrett S. Coordination vs voluntarism and enforcement in sustaining international environmental cooperation. Proceedings of the National Academy of Sciences of the United States of America, 2016, 113(51): 14515–14522

[4] Liu Z, Deng Z, He G. . Challenges and opportunities for carbon neutrality in China. Nature Reviews. Earth & Environment, 2021, 3(2): 141–155

[5] Zhang C, Li Y, Chu Z. . Thermodynamic analysis of integrated sorption-enhanced staged-gasification of biomass and in-situ CO₂ utilization by methane reforming process based on calcium looping. Energy Conversion and Management, 2023, 278: 116710

[6] Niu J, Zhang C, Liu H. . Enhanced performance of oxygen vacancies on CO₂ adsorption and activation over different phases of ZrO₂. Frontiers in Energy, 2023, 17(4): 545–554

[7] Lin B, Li Z. Towards world’s low carbon development: The role of clean energy. Applied Energy, 2022, 307: 118160

[8] Wang X, Pan C, Romero C E. . Thermo-economic analysis of a direct supercritical CO₂ electric power generation system using geothermal heat. Frontiers in Energy, 2022, 16(2): 246–262

[9] Alioui O, Badawi M, Erto A. . Contribution of DFT to the optimization of Ni-based catalysts for dry reforming of methane: A review. Catalysis Reviews, 2022, 65(4): 1468–1520

[10] Bradford M C J, Vannice M A. CO₂ reforming of CH₄. Catalysis Reviews. Science and Engineering, 1999, 41(1): 1–42

[11] Fan M S, Abdullah A Z, Bhatia S. Catalytic technology for carbon dioxide reforming of methane to synthesis gas. ChemCatChem, 2009, 1(2): 192–208

[12] Qian L, Cai W, Zhang L. . The promotion effect of hydrogen spillover on CH₄ reforming with CO₂ over Rh/MCF catalysts. Applied Catalysis B: Environmental, 2015, 164: 168–175

[13] Wang L, Ao X, Wang S. Catalysts for carbon dioxide catalytic reforming of methane to synthesis gas. Progress in Chemistry, 2012, 2012(9): 1696–1706

[14] Gaur S, Pakhare D, Wu H. . CO₂ reforming of CH₄ over Ru-substituted pyrochlore catalysts: Effects of temperature and reactant feed ratio. Energy & Fuels, 2012, 26(4): 1989–1998

[15] Ghelamallah M, Granger P. Impact of barium and lanthanum incorporation to supported Pt and Rh on α-Al₂O₃ in the dry reforming of methane. Fuel, 2012, 97: 269–276

[16] Silva F A, Hori C E, da Silva A M. . Hydrogen production through CO₂ reforming of CH₄ over Pt/CeZrO₂/Al₂O₃ catalysts using a Pd–Ag membrane reactor. Catalysis Today, 2012, 193(1): 64–73

[17] Stolze B, Titus J, Schunk S A. . Stability of Ni/SiO₂-ZrO₂ catalysts towards steaming and coking in the dry reforming of methane with carbon dioxide. Frontiers of Chemical Science and Engineering, 2016, 10(2): 281–293

[18] Noronha F B, Fendley E C, Soares R R. . Correlation between catalytic activity and support reducibility in the CO₂ reforming of methane over Pt/Ce_xZr_1−xO₂ catalysts. Chemical Engineering Journal, 2001, 82(1-3): 21–31

[19] Aghamohammadi S, Haghighi M, Maleki M. . Sequential impregnation vs. sol-gel synthesized Ni/Al₂O₃-CeO₂ nanocatalyst for dry reforming of methane: Effect of synthesis method and support promotion. Molecular Catalysis, 2017, 431: 39–48

[20] Ibrahim A A, Al-Fatesh A S, Kumar N S. . Dry reforming of methane using Ce-modified Ni supported on 8% PO₄+ZrO₂ catalysts. Catalysts, 2020, 10(2): 242

[21] Qin Z, Zhou Y, Jiang Y. . Recent advances in heterogeneous catalytic hydrogenation of CO₂ to methane. In: Ravanchi M T, ed. New Advances in Hydrogenation Processes—Fundamentals and Applications. InTech, 2017, 57–82

[22] Li Y, Wei Z, Wang Y. Ni/MgO catalyst prepared via dielectric-barrier discharge plasma with improved catalytic performance for carbon dioxide reforming of methane. Frontiers of Chemical Science and Engineering, 2014, 8(2): 133–140

[23] Niu J, Liu H, Jin Y. . A density functional theory study of methane activation on MgO supported Ni₉M₁ cluster: Role of M on CH activation. Frontiers of Chemical Science and Engineering, 2022, 16(10): 1485–1492

[24] Niu J, Zhang C, Li K. . The mechanism insight into methane activation on Cu–Ni/ZrO₂ surface and formation-removal of surface carbon. International Journal of Hydrogen Energy, 2024, 59: 1399–1408

[25] Niu J, Li K, Zhang C. . Mechanism study on carbon atom growth on different Ni facets in CO₂ reforming reaction. International Journal of Hydrogen Energy, 2024, 58: 1332–1344

[26] Ochoa A, Bilbao J, Gayubo A G. . Coke formation and deactivation during catalytic reforming of biomass and waste pyrolysis products: A review. Renewable & Sustainable Energy Reviews, 2020, 119: 109600

[27] Argyle M D, Bartholomew C H. Heterogeneous catalyst deactivation and regeneration: A review. Catalysts, 2015, 5(1): 145–269

[28] Chen D, Christensen K O, Ochoa-Fernández E. . Synthesis of carbon nanofibers:Effects of Ni crystal size during methane decomposition. Journal of Catalysis, 2005, 229(1): 82–96

[29] Kim J H, Suh D J, Park T J. . Effect of metal particle size on coking during CO₂ reforming of CH₄ over Ni–alumina aerogel catalysts. Applied Catalysis A, General, 2000, 197(2): 191–200

[30] Lercher J A, Bitter J H, Hally W. . Design of stable catalysts for methane-carbon dioxide reforming. Studies in Surface Science and Catalysis, 1996, 101: 463–472

[31] Zhang J, Wang H, Dalai A K. Effects of metal content on activity and stability of Ni–Co bimetallic catalysts for CO₂ reforming of CH₄. Applied Catalysis A, General, 2008, 339(2): 121–129

[32] Kim J H, Suh D J, Park T J. . Effect of metal particle size on coking during CO₂ reforming of CH₄ over Ni–alumina aerogel catalysts. Applied Catalysis A, General, 2000, 197(2): 191–200

[33] Guo Y, Feng J, Li W. Effect of the Ni size on CH₄/CO₂ reforming over Ni/MgO catalyst: A DFT study. Chinese Journal of Chemical Engineering, 2017, 25(10): 1442–1448

[34] Chen S, Niu J, Zheng X. . Unraveling the effect of particle size of active metals in Ni/MgO on methane activation and carbon growth mechanism. Physical Chemistry Chemical Physics, 2024, 26(2): 1255–1266

[35] Delley B. From molecules to solids with the DMol₃ approach. Journal of Chemical Physics, 2000, 113(18): 7756–7764

[36] Delley B. Fast calculation of electrostatics in crystals and large molecules. Journal of Physical Chemistry, 1996, 100(15): 6107–6110

[37] White J A, Bird D M. Implementation of gradient-corrected exchange-correlation potentials in Car-Parrinello total-energy calculations. Physical Review B: Condensed Matter, 1994, 50(7): 4954–4957

[38] Perdew J P, Burke K, Ernzerhof M. Generalized gradient approximation made simple. Physical Review Letters, 1996, 77(18): 3865–3868

[39] Bai M, Song Z, Yang Z. . Catalytic conversion mechanism of guaiacol as the intermediate of lignin catalytic pyrolysis on MgO surface: Density functional theory calculation. Journal of Molecular Liquids, 2023, 369: 120920

[40] Sao-Joao S, Giorgio S, Mottet C. . Interface structure of Ni nanoparticles on MgO(100): A combined HRTEM and molecular dynamic study. Surface Science, 2006, 600(7): L86–L90

[41] Speziale S, Zha C S, Duffy T S. . Quasi-hydrostatic compression of magnesium oxide to 52 GPa: Implications for the pressure-volume-temperature equation of state. Journal of Geophysical Research, 2001, 106(B1): 515–528

[42] Guo J, Lou H, Mo L. . The reactivity of surface active carbonaceous species with CO₂ and its role on hydrocarbon conversion reactions. Journal of Molecular Catalysis A Chemical, 2010, 316(1–2): 1–7

[43] Srivastava A K. CO₂-activation and enhanced capture by C₆Li₆: A density functional approach. International Journal of Quantum Chemistry, 2019, 119(20): e25904

[44] Bradford M C J, Vannice M A. Catalytic reforming of methane with carbon dioxide over nickel catalysts II. Reaction kinetics. Applied Catalysis A, General, 1996, 142(1): 97–122

[45] Luo J Z, Yu Z L, Ng C F. . CO₂/CH₄ reforming over Ni–La₂O₃/5A: An investigation on carbon deposition and reaction steps. Journal of Catalysis, 2000, 194(2): 198–210

[46] Wei J, Iglesia E. Isotopic and kinetic assessment of the mechanism of reactions of CH₄ with CO₂ or H₂O to form synthesis gas and carbon on nickel catalysts. Journal of Catalysis, 2004, 224(2): 370–383

Acknowledgements

This work was supported by Open fund of Key Laboratory of Low-grade Energy Utilization Technologies and Systems (Chongqing University) of the Ministry of Education, China (Grant No. LLEUTS-202308), the National Natural Science Foundation of China (Grant No. 52106179), the Fundamental Research Program of Shanxi Province, China (Grant No. 20210302124017), the Fund Program for the Scientific Activities of Selected Returned Overseas Professionals in Shanxi Province (Grant No. 20230012), the Shanxi Scholarship Council of China (Grant No. 2023-065), and the Graduate Education Practical Innovation of Shanxi Province, China (Grant No. 2023SJ056).

Competing Interests

The authors declare that they have no competing interests.

RIGHTS & PERMISSIONS

2024 Higher Education Press 2024