Enhanced production of hydrogen via catalytic methane decomposition on a Pt₇-Ni (110) substrate: a reactive molecular dynamics investigation

Abstract

Numerous researchers in the energy field are engaged in a competitive race to advance hydrogen as a clean and environmentally friendly fuel. Studies have been conducted on the different aspects of hydrogen, including its production, storage, transportation and utilization. The catalytic methane decomposition technique for hydrogen production is an environmentally friendly process that avoids generating carbon dioxide gas, which contributes to the greenhouse effect. Catalysts play a crucial role in facilitating rapid, cost-effective and efficient production of hydrogen using this technique. In this study, reactive molecular dynamics simulations were employed to examine the impact of Pt₇ cluster decoration on the surface of a Ni (110) catalyst, referred to as Pt₇-Ni (110), on the rates of methane dissociation and molecular hydrogen production. The reactive force field was employed to model the atomic interactions that enabled the formation and dissociation of chemical bonds. Our reactive molecular dynamics simulations using the Pt₇-Ni (110) catalyst revealed a notable decrease in the number of methane molecules, specifically ~11.89 molecules per picosecond. The rate was approximately four times higher than that of the simulation system utilizing a Ni (110) catalyst and approximately six times higher than that of the pure methane, no-catalyst system. The number of hydrogen molecules generated during a simulation period of 150 000 fs was greater on the Pt₇-Ni (110) surface than in both the Ni (110) and pure methane systems. This was due to the presence of numerous dissociated hydrogen atoms on the Pt₇-Ni (110) surface.

Graphical Abstract

Open in new tabDownload slide

Introduction

Hydrogen is an energy carrier with significant potential for future applications. Currently, numerous researchers in the energy sector are dedicated to the advancement of various aspects of hydrogen technology, such as production [1–6], storage [1, 7–10], transportation [11] and utilization [1, 3]. In the industrial sector, hydrogen is widely produced using the steam methane reforming technique [12–14]; however, this technique emits carbon dioxide (CO₂) gas [12], exacerbating the greenhouse effect. Therefore, catalysed membranes have been employed to mitigate the release of CO₂ into the environment [15], with the captured carbon dioxide subsequently utilized in the dry reforming procedure to generate carbon monoxide and hydrogen [15].

In contrast, the catalytic methane decomposition (CMD) technique provides a means of directly converting methane into hydrogen using a catalyst [16–18]. In this scenario, the significance of the catalyst is paramount in terms of decreasing the activation energy and expediting hydrogen production [16–18]. The hydrogen formation process utilizing the CMD technique begins with dehydrogenation of methane molecules on the catalyst surface, followed by the dissociated hydrogen atoms bonding together to generate hydrogen molecules [19].

Nickel (Ni) is extensively utilized as a catalyst in the processes of methane decomposition [20–25] and hydrogen production [21, 22]. Several prior studies have documented findings regarding the utilization of Ni in methane decomposition [20, 24–27]. Shimamura et al. [25] and Arifin et al. [24] employed ab initio molecular dynamics (MD) simulations to examine the process of methane decomposition on Ni clusters and Ni (111) substrates, respectively. Research findings indicate that Ni catalysts are involved in lowering the activation energy required for the dissociation of the C–H bond in methane and its fragments [24, 25]. In a recent study, Wang et al. investigated the dissociation of methane on Ni (111), Pd (111) and Ni₃Pd (111) surfaces [26]. Based on the findings of experimental studies and density functional theory (DFT) calculations, it has been reported that the Ni (111) surface exhibits a slightly higher level of activity than the other two surfaces. The dissociated carbon atoms exhibited a high affinity for the Ni (111) and Ni₃Pd (111) surfaces, with the carbon adsorption energy being slightly lower on the Ni₃Pd (111) surface than on the Ni (111) [26]. Niu et al. also reported the use of Ni clusters as catalysts for methane decomposition, observing that larger cluster sizes corresponded to higher catalytic activity [27]. Miyazawa et al. [20] have conducted investigations on the dynamics of methane dissociation on Ni surfaces. The reactive force field (ReaxFF) was employed to simulate bond breaking and formation between atoms. Researchers have successfully elucidated the process of methane dissociation on both Ni (111) [20]. However, the utilization of Ni as a catalyst is constrained by the circumstances in which the reaction of methane decomposition becomes saturated owing to the accumulation of carbon (C) atoms on the surface of the Ni [25].

In contrast, platinum (Pt) [28] exhibits greater catalytic activity in the process of methane decomposition than Ni [29] and methane dissociation on various Pt surfaces has been the subject of prior research [19, 30, 31]. Torio and Busnengo assessed the impact of Pt surface sites on the stability of methane fragments by employing DFT [30]. Roy et al. subsequently demonstrated, also through DFT investigations, that using a Pt–Ni combination as a catalyst can effectively reduce the energy barrier associated with methane dissociation compared with pure Pt and Ni catalysts [31]. Furthermore, the use of Pt is expected to successfully hinder the deactivation of the catalyst surface produced by the infiltration of C atoms on the Ni surface [32] owing to the limited solubility of carbon in Pt [33]. By incorporating Pt clusters, the surface area of Ni (110) exposed to C atoms can be reduced. Furthermore, Arifin et al. recently elucidated the mechanism of methane dehydrogenation, followed by hydrogen production, on the Pt (100) surface by employing reactive MD simulations [19]. The cost of Pt [34] significantly exceeds that of Ni [35], however. Hence, it is imperative to optimize the design of Pt catalysts for mass hydrogen production to achieve cost-effective manufacturing. The goal of this research is the development of a catalyst for hydrogen production using the CMD technique. This can be achieved by combining Ni and Pt clusters as the catalysts. A small Pt_n cluster with n = 7 was selected to minimize the amount of Pt metal used. We aim to demonstrate an enhanced rate of hydrogen production by employing a Ni (110) catalyst decorated with Pt₇ clusters, based on reactive MD simulations. In addition, we aim to elucidate the underlying reaction mechanism.

1 Numerical methods

We used the ReaxFF program 2023.1 [36–38], integrated into the Amsterdam Modeling Suite, to conduct MD simulations. The ReaxFF force field was employed to model the interactions among Ni, Pt, C and H atoms, utilizing the parameters specified by Fantauzzi et al. [39].

In this study, we investigated the effect of employing a catalyst in the methane decomposition process. We modelled three systems to simulate hydrogen production: (i) 700 methane molecules at elevated temperatures with no catalyst, (ii) a Ni (110) surface as the catalyst and (iii) a Ni (110) surface decorated with Pt₇ clusters as the catalyst, as depicted in Fig. 1a. The density of methane in all the three simulation systems was 0.27 g/cm³. Simulation systems (ii) and (iii) utilized a Ni (110) substrate consisting of 20 layers. In simulation system (iii), 18 Pt₇ clusters were deposited onto both the upper and lower surfaces of the substrate, serving as active surfaces and facilitating the reaction with methane molecules. To the best of our knowledge, the influence of Pt cluster size on the dissociation of methane molecules has not been the subject of any prior research. Based on the observation of numerous active sites in the methane dissociation process, specifically at the cluster edges close to the Ni atoms (as illustrated in Fig. 1a), the Pt₇ cluster was chosen for this investigation. A minor region of the Ni (110) surface that was not covered by Pt₇ clusters provided an opportunity for the deposition of dissociated C atoms because of the arrangement of these clusters on the Ni surface.

The methane adsorption energy, $Eads$⁠, on the substrate surface was calculated using Equation (1), as follows:

The energy of the complete system, which includes both the substrate and molecules, is denoted as $Es+mol$⁠. $Es$ represents the energy of the substrate and $Emol$ represents the energy of the isolated molecule. The computation of the adsorption energy of methane on the surface of the catalyst utilizes a simulation system that is comparatively smaller than that employed in MD simulations (see Fig. 1b).

Reactive MD simulations were used to investigate the process of H₂ production from methane molecules, including both catalysed and non-catalysed scenarios. The temperature in the simulations was controlled using the Nosé–Hoover chain technique [40]. To guarantee observation of the hydrogen production process in all models, the methane molecules were subjected to a very high temperature of 3000 K. In contrast, the substrate temperature in models (ii) and (iii) was maintained at 800 K to ensure that it remained in the solid state. The simulation was conducted for 600 000 steps, with each step representing a duration of 0.25 fs. Thus, the total simulation time was 150 000 fs (150 ps).

2 Results and discussion

2.1 The adsorption of methane and its fragment on the Ni (110) and Pt₇-Ni (110) surfaces

The initial step of the hydrogen formation process using the CMD technique involved the adsorption of methane onto the catalyst surface. This study investigated the adsorption of methane molecules on the surfaces of Ni (110) and Pt₇-Ni (110). Hence, we computed the adsorption energy (E_ads) values for methane and its fragments on both Ni (110) and Pt₇-Ni (110) surfaces, as presented in Table 1. The adsorption energy was determined using Equation (1) at a temperature of 0 K, where lattice vibrations are absent.

Table 1 displays the adsorption energy values, revealing a weak interaction between methane molecules and Ni atoms on the Ni (110) surface. This is evident from the smaller negative adsorption energy values compared with those of the other fragments. To clarify, methane molecules were physically adsorbed [41] onto the Ni (110) surface. The adsorption of methane molecules onto the Pt₇-Ni (110) surface exhibited a comparable interaction. Table 1 shows that a decrease in the quantity of hydrogen atoms in the methane fragments leads to an increase in their interaction with the nickel atoms on the nickel (110) surface. This is indicated by the increasingly negative adsorption energy values. Furthermore, it has been verified that the adsorption energy of hydrogen atoms in this study exhibits a mere deviation of 0.05 eV compared with the calculation results obtained previously using the DFT method [42].

However, when methane and its fragments were adsorbed onto the Pt₇-Ni (110) surface, distinct results were observed. Methyl, methylene, methylidyne and C fragments were found to have comparable adsorption energies. Table 1 shows that the methylene, methylidyne and C fragments exhibit a higher affinity for the Ni (110) surface than for those situated above the Pt₇ cluster. In contrast, the adsorption energy of hydrogen atoms on the Pt₇-Ni (110) surface was more negative than that on the Ni (110) surface, demonstrating the affinity of hydrogen atoms for forming bonds with the Pt atoms on the Pt₇-Ni (110) surface. These results exhibit a discrepancy when compared with the findings for the pure Pt surface [43]. As the number of hydrogen atoms in the methane fragment decreased, the interaction between the fragments and the pure Pt surface became stronger [43]. Based on DFT calculations, it has been documented that the adsorption energies of methane, methyl, methylene, methylidyne, C atoms and hydrogen atoms on the Pt (110) surface are reported as 0.46, 3.34, 5.81, 7.64, 9.05 and 4.42 eV [43], respectively.

Interestingly, when the distance between the methane molecule and the Pt₇ cluster on the substrate surface was very small (i.e. 1.41 Å in our calculation), a hydrogen atom promptly dissociated and established bonds with the Pt atoms (see Fig. 2). This was not observed during methane adsorption on the Ni (110) surface. It can be confirmed that the lower adsorption energy of hydrogen atoms on the Pt₇-Ni (110) surface affects the ease with which hydrogen atoms dissociate from methane molecules.

2.2 Mechanism of methane decomposition on the Ni (110) and Pt₇-Ni (110) surfaces

The first step in hydrogen molecule synthesis via CMD involves the removal of hydrogen atoms from methane on the surface of the catalyst. This study included the decomposition of methane using Ni (110) and Pt₇-Ni (110) as catalysts. The mobility of methane was enhanced by increasing the simulation temperature to 3000 K, resulting in a rapid dissociation process. Conversely, the substrate is kept in the solid phase by maintaining a temperature of 800 K.

The dissociation of a single hydrogen atom from methane on the Ni (110) surface is illustrated in Fig. 3. At 2950 fs, methane molecules approached the Ni (110) surface. C–H bond elongation then took place in the methane molecule at 3000 fs, where one of the hydrogen atoms was in close proximity to the Ni (110) surface. At 3050 fs, the hydrogen atom in the methane molecule detached from the carbon atom and underwent surface hopping on Ni (110). The occurrence of hydrogen atom hopping across surfaces was further validated by prior simulation outcomes of methane dissociation on the Ni (111) surface using the DFT–MD technique [44]. Furthermore, it was observed that methyl fragments exhibited mobility on the Ni (110) surface, transitioning between different sites. This transitioning may occur due to the weak adsorption of the methyl fragment and hydrogen atoms on the Ni (110) surface, as indicated by their adsorption energy values (see Table 1).

Fig. 4 shows the dissociation of the methyl groups into methylene groups. At 3450 fs, one hydrogen atom dissociated from the methyl fragment without an initial observation of the bond stretch, in contrast to the dissociation of one hydrogen atom from a methane molecule. As depicted in Fig. 5, one hydrogen atom subsequently dissociated from the methylene fragment to form a methylidyne fragment on the Ni (110) surface at 7775 fs. The methylene fragment forms chemical bonds with two Ni atoms on the substrate surface. Upon transformation into a methylidyne fragment, bonds are formed with three atoms on the Ni (110) surface. Fig. 5 shows that a hydrogen atom becomes dissociated from the carbon atom within the methylidyne fragment at 35 750 fs. The carbon atom within the methylidyne fragment, initially positioned slightly above the Ni (110) surface, commenced subsurface penetration after dissociation. These simulation findings validate ongoing debates in the field of graphene development on Ni surfaces, in which carbon atoms permeate into the subsurface of nickel [45–48]. At the subsurface level, neighbouring C atoms chemically link to create hexagonal structures, carbon chains and several other formations [48]. Moreover, newly created C structures emerge from the surface by disintegrating the Ni structure located above them [48].

Next, the time interval between one H atom dissociation event and the subsequent dissociation was determined. The time intervals for methane to convert to methyl, methyl to methylene, methylene to methylidyne and methylidyne to C atoms are 3050, 400, 4325 and 27 975 fs, respectively. The lifetime of the methyl fragments on the Ni (110) surface was very short. In contrast, the longest lifetime on the Ni (110) surface was found for the methylidyne fragment. These results indicate that methyl is the most unstable methane-molecule fragment on the Ni (110) surface, whereas methylidyne is the most stable.

Figs 6 and 7 illustrate the methane breakdown process on the Pt₇-Ni (110) surface. Fig. 6 shows the dissociation of a hydrogen atom from methane, which occurred at a simulation duration of 500 fs. Initially, the methane molecule moved towards the Pt₇ cluster, where a hydrogen atom engaged with a corresponding Pt atom. Following dissociation, the H and C atoms in the methyl fragment exhibited a preference for remaining on the Pt₇ cluster rather than migrating to the Ni (110) surface. These results corroborate the calculation findings shown in Table 1, indicating that the adsorption energy of methyl and hydrogen fragments on the Pt₇-Ni (110) surface is comparatively lower than the corresponding values for Ni (110). The hydrogen atom separated from the methyl fragment at 725 fs to produce a methylene fragment, followed by the C atoms on the methylene fragment forming bonds with Ni atoms that were not covered by the Pt₇ cluster. This arose because of the significantly lower adsorption energy of methylene fragments on the Ni (110) surface than on Pt₇-Ni (110).

Fig. 7 illustrates the dissociation of a single hydrogen atom on the Pt₇-Ni (110) catalyst at 825 fs. Dissociation occurred at the surface locations of Ni, specifically between the two Pt₇ clusters. In addition to associating with Ni atoms, C atoms also form bonds with Pt atoms. After a certain period, the methylidyne fragment eventually released a single hydrogen atom onto the Pt₇-Ni (110) surface at 1775 fs. Methane decomposition was significantly accelerated on the Pt₇-Ni (110) catalyst surface compared with that on the Ni (110) surface. The lifetimes of methane, methyl, methylene and methylidyne were 500, 225, 100 and 950 fs, respectively. Based on these results, we determined that the decomposition of a single methane molecule on the Pt₇-Ni (110) catalyst surface is ~20 times faster than that on the Ni (110) surface.

2.3 Mechanism of hydrogen production via direct methane decompositions

This discussion focuses on elucidating the mechanism of hydrogen molecule creation using three different simulation system models. In the first model, hydrogen molecules were generated by the dissociation of methane at elevated temperatures in the absence of a catalyst. At 3000 K, the methane molecules exhibited significant vibrations in their C–H bonds. In addition, methane molecules moved extremely actively at such a temperature in both translation and rotation. A hydrogen molecule is created when two hydrogen atoms that are apart from the C atom in the methane molecule are in close proximity to each other. This mechanism is illustrated in Fig. 8, which shows snapshots of the formation of the first hydrogen molecule in the system simulation with no catalyst. The formation of a hydrogen molecule involves the bonding of two hydrogen atoms, which may originate either from within the same methane molecule or from two different ones. The successful production of pure hydrogen directly from methane without the use of a catalyst was recently reported by Yousefi and Donne [49]. The findings indicated that the initial reaction rate within the reactor was significantly higher and the temperature sensitivity during the initial phases of the reaction was notably higher [49]. Nevertheless, once a specific concentration of hydrogen was attained at each temperature, the impact of temperature on the reaction rate diminished and became nearly constant [49].

Fig. 9 illustrates the mechanism underlying the formation of the first and third hydrogen molecules on the Ni (110) surface. The first hydrogen molecule is generated through the combination of two dissociated hydrogen atoms, which subsequently undergo hopping motion on the Ni (110) surface. At ~7100 fs, the two atoms approached each other sufficiently closely to form a chemical bond and yield a hydrogen molecule. The formation of a third hydrogen molecule occurs as the methane molecule approaches the Ni (110) surface. At 8200 fs, one of the hydrogen atoms within the methane molecule was in close proximity to the hydrogen atom located on the Ni (110) surface. The two neighbouring hydrogen atoms then bonded, forming a hydrogen molecule at a simulation time of 8225 fs. Hydrogen molecules were promptly released from the Ni (110) surface upon formation. In this instance, the second hydrogen molecule was produced via the direct dissociation of hydrogen from two methane molecules, facilitated by subjecting methane molecules to elevated temperatures.

Fig. 10 shows the formation mechanism of the first and second hydrogen molecules on the Pt₇-Ni (110) surface. The first mechanism shows that two hydrogen atoms located on top of two different Pt atoms approached each other. The two atoms then bonded to form a hydrogen molecule. In the second mechanism, hydrogen molecules are formed from two hydrogen atoms located on the same Pt atom. It should be noted that the formation of the first hydrogen molecule on the Pt₇-Ni (110) surface occurred much faster than on the Ni (110) surface. The enhanced rate of methane decomposition on the Pt₇-Ni (110) surface compared with that on the Ni (110) surface resulted in a higher availability of hydrogen atoms on the Pt₇-Ni (110) surface.

2.4 The role of Pt₇ clusters on the Ni (110) surface to the enhancement of hydrogen production

In this section, we present a comparative analysis of the reaction rates of methane dissociation and hydrogen production in the three models. These models included a system without a catalyst, a system with a Ni (110) catalyst and a system with a Pt₇-Ni (110) catalyst. The comparison was based on reactive MD simulations. In this study, the rates of methane dissociation and hydrogen production were calculated by analysing the slope of the fitting line on the curve depicting changes in the number of molecules over time. The slope, denoted by m, represents the rate of change in the number of molecules. Negative values of m indicate a decrease in the number of molecules, whereas positive values indicate an increase.

Based on the data presented in Fig. 11, it is evident that there is a notable decline in the number of methane molecules and a corresponding increase in the number of hydrogen molecules over a period of 150 000 fs of reactive MD. In Fig. 11a, it can be observed that, at 3000 K, the rate of decrease in the number of methane molecules and that of the formation of hydrogen molecules are comparable. Specifically, the rate of decrease for methane was 2.02 molecules per 1000 fs (or 2.02 molecules/ps). The rate of formation for hydrogen was 2.07 molecules/ps. The observed trend for decreasing methane molecules and increasing hydrogen molecules followed a linear pattern. The comparable rate of decline in methane quantity and the corresponding increase in hydrogen quantity indicates that the majority of hydrogen molecules are generated through the dissociation of methane by the mechanism described earlier (see Fig. 8).

Fig. 11b shows an observable increase in the level of methane decomposition in the presence of the Ni (110) catalyst. From the initial stage of the simulation to ~50 000 fs, the rate of methane depletion was observed to be ~4.32 molecules/ps. Subsequently, the deposition of numerous methane fragments on the Ni (110) surface resulted in a decrease in surface reactivity. Consequently, the rate of methane reduction decreased to ~2.07 molecules/ps. The rate of degradation in this case was nearly equivalent to that observed for the catalyst-free system. The number of hydrogen molecules increased linearly at a rate of 4.36 molecules/ps. While the rate of methane-molecule decrease slowed, the rate of hydrogen increase remained high because the dissociated hydrogen atoms on the Ni (110) surface served as the source for hydrogen molecule formation.

The methane concentration in the simulation system utilizing a Pt₇-Ni (110) catalyst decreased at a rate of 11.89 molecules/ps from the time the simulation began until 30 000 fs. As dissociated carbon atoms and methane fragments began to populate the Pt₇-Ni (110) surface, the methane dissociation rate fell to 3.38 molecules/ps (between 30 000 and 90 000 fs of the simulation time). The reduction rate then decreased to a mere 0.84 molecules/ps, as the catalyst surface became saturated with methane fragments and only a few methane molecules remained. From the conclusion of the simulation period, ~93% of methane molecules underwent decomposition.

In a catalyst-free simulation system, ≤21.6% of the hydrogen atoms in methane were transformed into hydrogen molecules during the simulation run at 150 000 fs at 3000 K. Using Ni (110) and Pt₇-Ni (110) catalysts, the simulated system generated hydrogen molecules by employing ~42.8% and ~47.1% of the total hydrogen atomic sources, respectively. Our results show that the Pt₇-Ni (110) surface exhibits excellent catalytic performance in the methane breakdown process, as shown by the simulation results for a duration of 150 000 fs, which indicates that only 7.4% of the methane molecules remain. This percentage is much lower than the remainders in the system when using a Ni (110) catalyst and in the system without any catalyst, which were 36.7% and 57.9%, respectively.

This study has effectively elucidated the impact of Pt₇ clusters on the Ni (110) surface, leading to enhanced methane dissociation and hydrogen molecule formation. However, additional investigations are required to explore the effect of cluster size on these reactions, employing more precise calculation methods, such as DFT or similar approaches.

3 Conclusion

Based on the findings obtained from our reactive methane decomposition simulations conducted on the Pt₇-Ni (110) catalyst, we observed that there was a significant initial reduction in the number of methane molecules, of ~11.89 molecules/ps. This rate is approximately four times higher than that of the simulation system using a Ni (110) catalyst and approximately six times higher than that of the pure methane system. A high rate of reduction in the amount of methane on the Pt₇-Ni (110) surface occurred because the H atoms easily dissociated from the methane molecules on the substrate. It is important to mention that, in this simulation, the assigned temperatures for methane and the substrate were 3000 and 800 K, respectively. Similarly, when there is an abundance of dissociated hydrogen atoms on the Pt₇-Ni (110) surface, the number of hydrogen molecules produced during a simulation period of 150 000 fs surpasses that achieved in both Ni (110) and pure methane systems. These findings suggest that employing a Pt₇-Ni (110) catalyst may enhance the rate of hydrogen production and expedite the dissociation of methane, in comparison with either a Ni (110)-catalyst or a methane-only system.

Author contributions

Rizal Arifin: Conceptualization, Methodology, Software, Investigation, Writing—Original Draft, Finding acquisition. Zulkarnain: Formal Analysis, Writing—Review & Editing. Abdurrouf: Validation, Writing—Review & Editing. Yoyok Winardi: Project Administration, Writing—Review & Editing. Didik Riyanto: Visualization, Writing—Review & Editing. Darminto: Methodology, Writing—Review & Editing.

Conflict of interest statement

There are no conflicts to declare.

Funding

This research was funded by a PFR 2023 research grant from the Ministry of Education, Culture, Research, and Technology of the Republic of Indonesia (contract number 183/E5/PG/02.00.PL/2023).

Data Availability

Raw data were generated using ReaxFF software under Academic license. Derived data supporting the findings of this study are available from the corresponding author on request.

References