Synthesis and Experimental Screening of Catalysts for H₂S to H₂ Decomposition Under Close-to-Industry Conditions

Abstract

The chemical engineering community has shown significant interest in investigating methods to decompose hydrogen sulfide into hydrogen and sulfur. However, there is still a lack of detailed experimental data enabling us to choose the optimal catalyst, reaction, and regeneration conditions, as well as the overall process design. The purpose of this work is to synthesize a series of catalysts and compare their catalytic activity under the same conditions, chosen on the basis of a possible large-scale H₂S conversion process. To achieve this, the obtained catalysts were characterized by BET, XRD, SEM, TEM, and XPS before and after the reaction. Decomposition was conducted in a laboratory fixed-bed reactor at a temperature of 500 °C, 10 vol% of H₂S in the feed, and a GHSV of 540–1000 h⁻¹. DFT calculations evaluated the H₂S bond cleavage on various catalyst surfaces. It was shown that the most promising catalyst was Ni₃S₂, offering an acceptable H₂S conversion of 40%. We also observed that Ni₃S₂ catalyst regeneration could be conducted at much milder conditions compared to those previously reported in the literature. These results highlight the viability of upscaling the process with the selected catalyst.

1. Introduction

Hydrogen sulfide gas is one of the byproducts of the oil industry, which can be present in natural gas ranging from trace amounts to tens of volume percent. H₂S is also produced by hydrotreating oil fractions up to 10 vol% in a circulating hydrogen stream and is concentrated in the acid gas during sour gas sweetening [1,2,3]. The Claus process has long been known and used as a method for the conversion of hydrogen sulfide to sulfur. It has been improved many times since its implementation to minimize harmful emissions and increase sulfur recovery, which nowadays can exceed 99% [4].

Alternative conversion methods enabling the conversion of H₂S into sulfur and valuable hydrogen also draw a lot of interest. Different approaches were suggested, such as plasma discharge [5,6], thermal and thermocatalytic conversion [7,8], or the use of microorganisms [9]. The latter is already commercially available technology [10]. Thermal decomposition is a straightforward and well-established method that involves the splitting of the H₂S molecule into hydrogen and sulfur through dissociation at elevated temperatures. However, even at temperatures as high as 1000 °C, a conversion of only 15% is possible due to the thermodynamic limitations of this reaction. By introducing catalysts, it is possible to lower the reaction temperature, hence lowering the energy consumption. To date, a considerable number of studies have been published concerning the decomposition of hydrogen sulfide in the presence of metal sulfides (Mo-W [11,12,13,14,15], Fe-Co-Ni [12,15,16,17,18,19,20,21], Cu [16], etc. [22]), metal oxides [23,24,25], and oxygen-containing structures such as hexaaluminates, perovskites, and hydrotalcites [26,27,28,29]. Using the suggested catalysts, high hydrogen yield has been achieved within the considerably lower temperature range of 500–900 ℃.

However, to proceed with the implementation of a thermocatalytic process for converting H₂S to H₂, there is a pressing need for organized experimental data on H₂S decomposition under close-to-industry reaction conditions. Today, transferring the reported results to the industry is still problematic. Numerous articles choose very low H₂S concentrations in the feed flow (around 1000 ppmv) or extremely high contact time. In practice, the H₂S concentration in sour gas streams can vary from 10 to 50 vol%, and the flow rate prevents the reaction from being conducted for hours. These results cannot be used to assess the technology. Furthermore, it is sometimes difficult to compare the results due to substantial variations in the experimental design and initial conditions. For example, some studies do not specify the parameters of the process, such as gas hourly space velocity (GHSV), contact time, or catalyst load, which could lead to severely different hydrogen yield in the presence of the same catalyst under seemingly the same conditions: temperature and H₂S initial concentration [11,12].

The purpose of this work is to synthesize the most promising (according to the literature) catalysts for hydrogen sulfide decomposition and compare their activity under the same conditions using one given reactor. Secondly, to generalize the routes of hydrogen sulfide conversion on these materials, the changes in catalyst composition were characterized after catalysis. Lastly, an effort was made to identify the main influencing factors affecting the process of upscaling. The activity of synthesized catalysts was assessed at a reaction temperature of 500 ℃, an initial H₂S concentration of 10 vol%, and a GHSV of 1000 h⁻¹. The temperature of 500 ℃ is the minimum in the studied range of thermocatalytic conversion of hydrogen sulfide, and its choice is determined by minimizing energy consumption during the possible implementation of the process; an initial concentration of hydrogen sulfide of 10 vol% was chosen for reasons related to modeling a realistically possible flow of acid gas containing H₂S; GHSV of 1000 h⁻¹ was selected from the range of typical values for catalytic plants involved in hydrogen sulfide processing [30,31,32].

2. Results and Discussion

2.1. Catalysts

In this work, several catalysts for the decomposition of hydrogen sulfide into hydrogen and sulfur were synthesized, namely the following:

• Series of FeS doped with Mo at a concentration of 1–3 at%;
• Series of FeS doped with W at a concentration of 1–3 at%;
• A sample of pyrrhotite-like structure doped with Mo: (Fe_0.98Mo_0.02)_0.87S;
• A sample of pyrrhotite-like structure doped with W: (Fe_0.98W_0.02)_0.87S;
• A sample of Ni₃S_2;
• Series of hexaaluminate catalysts: LaFe_xAl_12−xO₁₉, where x = 2, 4, or 6.

The methods for preparing these catalysts are provided separately in Section 3, and more detailed data are provided in the Supplementary Materials.

The structure and compositions of catalysts were analyzed and confirmed using TEM and XRD. According to micrographs (Figure 1), most samples consisted of particles with a size of 100–500 nm. Particles of the Fe_0.98Mo_0.02S sample were somewhat larger in the range of 1–3 μm. The XRD spectra of sulfide catalysts are provided in Figure 2 and Figure 3, while synthesized aluminate XRD spectra are summarized in Figure 4. Target phases were detected in all obtained samples, and the data were consistent with the literature [21,26] as all diffraction peaks were observed, which were attributed to (1) FeS with a distorted lattice due to doping; (2) rhombohedral Ni₃S₂; and (3) a magnetoplumbite phase, which is characteristic for reported hexaaluminate catalysts. A sole exception was the identified additional phase of MoS₂/WS₂ in some samples, the impact of which on the catalytic activity will be discussed below. According to STEM elemental mapping, the distribution of metals over the surface of the particles is even (Figure 5).

According to the adsorption isotherms obtained by BET analysis, the catalyst powder samples can be classified as non-porous. The estimated specific surface area of such samples lies in the range of 1–10 m²/g. On the other hand, the synthesized supported Ni₃S₂/AAO is characterized by a high specific surface area of 251 m²/g, with an initial characteristic of 291 m²/g. The decrease in specific surface area is most likely due to the blocking of pores by active phase particles deposited on the surface.

2.2. Catalyst Activity Tests

The performance of doped troilite samples is illustrated in Figure 6. The observed impact of the dopant aligns with the existing body of knowledge: the addition of Mo as a dopant leads to an enhancement in hydrogen production, resulting in an increase in hydrogen yield from 8% to 10%. The optimal concentration of dopant is 2 atomic percent (at%). It should be noted that the W-doped FeS sample with the W concentration of 2 at% exhibits comparable activity in the decomposition of hydrogen sulfide. Based on DFT calculations, Jangam and co-workers came to the conclusion that the role of the dopant is mainly to facilitate the diffusion of hydrogen rather than affect the observed reaction energy barrier [21]. Apparently, the same effect can be achieved by introducing W atoms into the FeS catalyst. The relatively low hydrogen yield in the case of Fe_0.99W_0.01S against the Fe_0.99Mo_0.01S catalyst (6% hydrogen yield compared to 8%, respectively) is supposedly due to the phase separation for the W-doped sample. This was suspected after the XRD results demonstrated strong reflexes corresponding to the WS₂ phase. As mentioned before, an additional phase of MoS₂/WS₂ was identified in some troilite samples. However, in the case of Fe_0.99W_0.01S, such phase separation was stronger and possibly hindered its activity.

A comparison of hydrogen yield obtained using pyrrhotite-like catalysts (Mo_0.02Fe_0.98)_0.87S and (W_0.02Fe_0.98)_0.87S with the corresponding troilite-like samples is shown in Figure 7. It can be concluded that sulfides with excess sulfur have low activity in the conversion of hydrogen sulfide to hydrogen. This result is consistent with the sulfur looping concept, according to which metal sulfides with a metal/sulfur ratio of less than 1 are considered deactivated and require the removal of sulfur. The decline in catalyst activity can also be attributed to a reduction in the reaction rate caused by the accumulation of sulfur on the catalyst’s surface. This conclusion is indirectly confirmed by the results of the work of Kwok and others [13]. In their study, the researchers successfully obtained a hydrogen yield of 10% using a flow reactor and a core–shell catalyst consisting of a MoS₂ core encapsulated in a SiO₂ shell. The creation of that specific structure facilitated the diffusion of hydrogen sulfide and the elimination of reaction byproducts, resulting in a beneficial effect on the hydrogen yield. The difference in hydrogen yield for (Mo_0.02Fe_0.98)_0.87S and (W_0.02Fe_0.98)_0.87S was not considered in this work as a significant result since it is most likely associated with a large error in determining the hydrogen yield at such low conversion values.

The obtained hydrogen yield during the decomposition of H₂S in the presence of hexaaluminates is shown in Figure 8. The studied systems have low catalytic activity in the decomposition of hydrogen sulfide under the studied conditions (500 °C, 1000 h⁻¹). The obtained hydrogen yield fluctuates around 1%, and its change for studied hexaaluminates is more likely due to experimental error than due to the influence of the Fe/Al ratio. In the literature, these materials were proposed in the work of Jiang and co-workers [26], where it was demonstrated that it was possible to achieve high conversion of hydrogen sulfide (above 50%) at a temperature of 800 °C. The authors proposed a path for direct dehydrogenation of hydrogen sulfide catalyzed by Fe³⁺ centers in the aluminate structure, which was also supported by DFT calculations.

After catalysis, the XRD spectra of hexaaluminate catalysts are essentially identical to those of the initially synthesized ones. All sulfide catalysts demonstrate the emergence of new phases: Fe₇S₈ for pyrrhotite-like and troilite-like catalysts and NiS for the N₃S₂ catalyst (Figures S1–S4 in the Supplementary Materials). This means that no sulfur was included in the structure of hexaaluminate catalysts during contact with H₂S. However, according to STEM elemental mapping, an accumulation of sulfur is observed on the surface of the aluminates (Figure 9), which apparently represents elemental sulfur formed during the decomposition of hydrogen sulfide during interaction with iron atoms in the structure of the aluminate, as suggested in the literature.

2.3. Catalyst Regeneration

The highest catalytic activity among the studied materials under the selected H₂S decomposition conditions (10 vol% H₂S, 500 °C, GHSV 1000 h⁻¹) is demonstrated by the Ni₃S₂ sample exhibiting a hydrogen yield of 23%. The activity of such sulfide is due to the sulfidation process yielding hydrogen and NiS phase. This was verified through X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and scanning transmission electron microscopy (STEM) elemental mapping analysis. Please refer to Figures S4–S7 in the Supplementary Materials for further details. The Ni2p spectrum was fitted with a sum of three components: Ni³⁺ (BE = 856 eV), Ni-S satellite (BE = 861.6 eV), and Ni²⁺ (BE = 853 eV). The obtained XPS spectrum correlates with the published data, and a minor contribution of Ni²⁺ is often observed for Ni₃S₂ [33]. The S2p peak was fitted with three components, namely NiS (BE = 162.5 eV), NiS₂ (BE = 164 eV) and SO₄²⁻ (BE = 169.6 eV). The presence of sulfates might be related to surface oxidation also observed before [33]. In the “sulfur looping” process, the catalyst is heated in the presence of an inert gas. This results in the sulfur that builds up during the decomposition of hydrogen sulfide to be released from the crystal structure of the sulfide and evaporate from the surface. In the meantime, the sulfide phase responsible for breaking down hydrogen sulfide is regenerated. Moreover, within this study, it was found that the regeneration of the catalyst can be carried out at a temperature lower than proposed in the literature [17], which is confirmed by TGA data for the deactivated nickel catalyst powder. Indeed, the beginning of the transition of NiS to Ni₃S₂ with a mass loss close to 11.8% is observed already at a temperature of 600 °C. The catalytic activity of the nickel catalyst after the thermal treatment at this temperature for an hour is restored almost completely (Figure 10). The difference in hydrogen yield for fresh and regenerated catalysts is negligible (23 vs. 22%) given the relative error in the analysis of the gaseous products (10% relative as stated in the Section 3).

2.4. Supported Catalyst Tests

Figure 11 presents a comparison of the activity of powder and supported catalysts for nickel-based systems. An increase in the hydrogen yield from 23 to 27% was noted, which can be explained by an increase in the dispersion of active phase particles and their stabilization by the Al₂O₃ matrix. Here, we also demonstrated that the distribution of Ni₃S₂ particles on the matrix has a beneficial effect on the catalyst, lowering its deactivation rate (Figure 12). The sudden loss of activity is attributed to near complete Ni₃S₂ sulfidation to NiS, which is supported by XRD analysis (Figure S4 in the Supplementary Materials). The latter phase is not active in H₂S conversion, and small quantities of Ni₃S₂ left are either inaccessible or cannot support adequate conversion due to the very high actual GHSV of H₂S per active catalyst left.

The effect of GHSV on the yield of hydrogen during the decomposition of hydrogen sulfide in the presence of Ni₃S₂ is presented in Figure 13. With the decrease of GHSV to 540 h⁻¹, an increase in the hydrogen yield is observed by almost 1.5 times. The obtained result can be explained by the fact that a decrease in the GHSV means an increase in the contact time with the surface of Ni₃S₂, providing conditions for the diffusion of H₂S molecules to unreacted centers, their decomposition, and desorption of H₂. A further decrease in the GHSV did not lead to an additional increase in hydrogen yield, which apparently means that, in this case, it is already limited by the availability of fresh active sites, which decreases with the accumulation of sulfur on the catalyst surface.

The comparison of the obtained results with the state of the art reported in the literature is presented in

Table 1. Achieved hydrogen yield under different conditions of H₂S decomposition. Comparison of the results with recent publications.

Ref.	Catalyst	Temp., °C	GHSV, h−1	H2S Content, vol%	H2 Yield, %
Metal sulfides
[12]	FeS	500	150 *	100	3
[13]	MoS2 core SiO2 shell	500	n/a	0.25	10
This work	Fe0.98Mo0.02S Fe0.98W0.02S	500	1000	10	10
Supported nickel sulfide
[16]	Supported Ni	550	13,500	10	2.5
[17]	Supported Ni3S2	700	5000	0.9	80
This work	Supported Ni3S2	500	540	10	40
Aluminates
[26]	LaFexAl(12−x)O19	800	24,000	0.1	50
This work	LaFexAl(12−x)O19	500	1000	10	1

* Estimated by authors.

. As can be seen from the data, metal sulfides themselves possess quite low activity in H₂S decomposition. According to the reported DFT calculations, the reason for this is diffusion limitations [21]. Kwok et al. proposed a solution to this by synthesizing core-shell catalyst particles. However, in this work, we verified that doping sulfide is also a promising approach. The obtained doped catalyst gives a similar yield of hydrogen at a much higher initial H₂S concentration (10% versus 2500 ppm).

New data on H₂S decomposition to hydrogen in the presence of Ni₃S₂ were obtained in this work. We demonstrated that this material is highly active in hydrogen production from H₂S even at much lower temperatures and higher initial hydrogen sulfide concentrations than was reported before. These conditions simulate the more adequately possible conversion of H₂S in industry and could be used for evaluating hydrogen production on a larger scale, e.g., while designing a pilot plant.

In this work, for aluminates (described by LaFe_xAl_(12−x)O₁₉ formula), there is practically no catalytic activity that has been observed. We suppose that the main reason for such a low H₂ yield is the selected reaction temperature. Specifically, high H₂ yield was reported in the literature but at temperatures as high as 800 °C (Table 1, Ref. [26]). In this work, it was demonstrated that at 500 °C, this catalyst is practically inactive in H₂S decomposition. However, further research is needed to study the temperature dependence of H₂ yield for this catalyst.

2.5. DFT Study

Jangam and co-authors [17] examined the effect of the surface area of the catalyst on the conversion of hydrogen sulfide, which is an important parameter in the general theory of heterogeneous catalysis. Unfortunately, in this article, the observed surface of Ni₃S₂ was limited to only S terminated (111) surface, while the ideal equilibrium morphology for Ni₂S₃ nanoparticles can contain both (111) and (11−1)-type faces, based on theoretical calculations [34]. In our investigation, we take into consideration other possible surfaces of the Ni₃S₂ crystal: (001), (11−1), (110), (1−10), (−111), (−210), and (111). The analysis of the surface energy (see SM DFT Section 1) reveals the favorable ones: (111)_S, (001) with unequal sides A and B, (11−1)_S, (110)_3, (−111), (1-10), and (−210); see Figure 14.

The barriers of individual stages of the hydrogen cleavage reaction from the hydrogen sulfide molecule on selected Ni₃S₂ crystal surfaces were calculated using the NEB method in order to reveal the catalytic properties of these surfaces. The calculated values of the barriers are presented in

Table 2. Barrier heights of individual stages of H₂S decomposition on Ni₃S₂ surfaces.

Surface	Barrier, (eV)
	TS1	TS2
(111)_S	0.39	0.81
(001) side A	1.16	1.86
(001) side B	1.20	1.44
(11-1)_S	0.34	1.14
(110)_3	0.33	1.17
(−111)	1.01	0.97
(1-10)	1.24	0.57
(−210)	0.89	1.77

.

The transition state TS1 corresponds to reaction H₂S^* → HS^* + H^* and TS2 to HS^* → S^* + H^* reaction. The reaction paths were calculated between the most energetically favorable sorption positions of the H₂S molecule and its individual fragments (HS, S, H), which were determined before NEB calculations. Among all considered surfaces of Ni₃S_2, only three of them, (110)_3, (11-1)_S, and (111)_S, demonstrate a barrier less than 0.4 eV; at the same time, only (111) S surfaces show the lowest barrier for second hydrogen cleavage among these three with 0.81 eV barrier. Despite the 0.57 eV barrier for TS2 at (1-10) surface of Ni₃S_2, the reaction could be limited by the first stage with a 1.24 eV barrier.

3. Materials and Methods

For the synthesis of catalysts, FeS, MoS₂, WS₂ powders and Ni (II), La (III), Fe (III), Al (III) nitrates (all reagents ≥ 98 wt% pure, provided by LLC “ChemExpress”), and an (NH₄)₂S solution (Sigma-Aldrich, 20 wt% aq. Solution, St. Louis, MO, USA) were used without further purification.

The synthesis of Fe sulfides doped with Mo and W was carried out using sintering of homogenized powders at a temperature of 850 °C for 48 h. The choice of FeS is justified by the fact that it is one of the potentially cheap and easily accessible compounds found in nature. Moreover, a theory has been proposed for the natural formation of hydrogen from H₂O molecules in the presence of Fe (II)-containing minerals, including FeS [35,36]. The crystallization of materials with the structure of both troilite (FeS) and pyrrhotite (Fe_0.87S) is possible in the Fe-S system [37]. The method of synthesis is described in detail elsewhere [38] and in the Supplementary Materials.

Nickel sulfide (Ni₃S₂) was synthesized according to the method proposed by Bezverkhyy and co-authors [39] by treating NiS with the flow of H₂-H₂S mixture at a temperature of 300 °C. This material was chosen as one of the promising catalysts for the “sulfur looping scheme” [40]. To evaluate the effect of the support, Ni₃S₂ supported on Al₂O₃ was synthesized. To obtain a supported catalyst, nickel was first distributed on Al₂O₃ by wet impregnation with Ni(NO₃)₂ and then converted to NiS by treating the catalyst with (NH₄)₂S. For more information, please refer to Supplementary Materials.

A series of hexaaluminate catalysts, LaFe_xAl_12−xO₁₉ (x = 2, 4, 6), were prepared using the carbonate route [26] via precipitating La, Fe, and Al nitrates solutions. A more detailed description of synthesis is given in the Supplementary Materials.

XRD analysis was carried out using a Rigaku X-ray spectrometer Rotaflex RU-200 using copper Kα radiation (λ = 0.154 nm) in 50 kV/160 mA mode in the angular range 2θ = 1–145° with a step of 0.04° and a shooting speed of 2 deg/min. Powdered samples (with a typical size of less than 10 µm) were used for analysis. Phase identification was carried out using the ICDD PDF-2 database.

SEM and TEM analysis were carried out using a FEI Tecnai G2 F20 S-TWIN TMP instrument with an EDAX EMF attachment and a holder for inertia-free heating DENS Solutions A-SH30-4M-FS, accelerating voltage 200 kV. Powder samples were pre-dispersed in distilled water in an ultrasonic bath for 2 min, settled for 2–3 min, and then powder particles were deposited onto a copper mesh coated with a thin layer of amorphous carbon.

XPS spectra were obtained using a PHI VersaProbe II 5500 X-ray photoelectron spectrometer (primary X-ray photon energy 1100 eV, analysis depth up to 10 nm). Samples in powder form were placed on a thin layer of indium and pressed through glass to form a flat surface. To determine the elemental composition of the samples, the obtained XPS spectra were analyzed using specialized MultiPak 9.9.2 software.

The specific surface area (BET) of the catalysts was measured using a Belsorp instrument miniX with preliminary thermal degassing of the samples at a temperature of 300 °C and a pressure of 10 Pa for 10 h.

DTG analysis was carried out using Mettler Toledo Thermal Analysis System TGA/DSC 3+ while heating the sample at a rate of 10 °C/min in the flow of argon up to 1000 °C.

Catalytic tests were carried out in a tubular quartz reactor with a stationary catalyst bed at a temperature of 500 °C. 600 mg of catalyst was distributed in a layer of inert quartz beads to preserve a constant volume of the catalyst bed of 7 mL throughout the tests. The supported catalyst was loaded by weight without quartz beads so that the mass of the active phase was 600 mg. A hydrogen sulfide flow with an initial concentration of 10 vol% (balanced by N₂) was fed at a rate of 1000 h⁻¹ (for this experimental setup, it corresponds to a feed flow rate of 3.6 l/h, with the estimated total volume of voids of 3.5 mL or 50% of the total volume).

Gaseous reaction products were collected for analysis by redirecting the gas flow at the outlet of the reactor into the chromatograph loop. Analysis of gaseous products was carried out using an LLC Meta-Chrome “Kristallux 4000M” chromatograph equipped with a capillary column HP-PLOT Q (15 m × 0.32 mm, stationary phase thickness 20 µm) and a thermal conductivity detector (TCD). Nitrogen was used as a carrier gas. The chromatograph was first calibrated using mixtures of H₂S-N₂ and H₂-N₂ mixtures. The average relative error in determining the hydrogen content is 10%. Hydrogen yield was calculated by the following equation:

$$\mathrm{Y}={\displaystyle \frac{\mathrm{C}\left({\mathrm{H}}_2\right)}{\mathrm{C}\left({\mathrm{H}}_2\mathrm{S}\right)}}\times 100\%,$$

where C(H₂) is the concentration of H₂ (vol%), and C(H₂S) is the initial concentration of hydrogen sulfide (in this work 10 vol%).

The quantum–chemical calculations were carried out within the density functional theory (DFT) [41,42], implemented in the VASP package [43,44,45]. The exchange-correlation function was calculated in the generalized gradient approximation (GGA) in the Perdue–Burke–Ernzerhof (PBE) parameterization [46]. To describe the ion–electron interaction, the projector augmented wave method (PAW) [47] was used. Integration in the first Brillouin zone was performed using the Monkhorst–Pack method [48]. Relaxation of the atomic structure was carried out until the energy difference between the two steps of the ion optimization became less than 10⁻⁴ eV, and the energy difference between the two steps of the electron optimization cycle became less than 10⁻⁵ eV. In the calculations, Grimm dispersion corrections were applied to describe the van der Waals interaction (DFT-D3 method) [49]. The diffusion barriers were calculated using the nudged elastic band method (NEB) [50]. The VESTA [51] programs were used to visualize the atomic structure.

4. Conclusions

In this work, a number of catalysts were synthesized, and their catalytic activity was compared in the decomposition of H₂S to sulfur and hydrogen under the same conditions, chosen on the basis of modeling possible large-scale H₂S conversion process: temperature of 500 °C, H₂S initial concentration of 10 vol% and GHSV of 540–1000 h⁻¹. The obtained results were compared with the state of the art reported in the literature. It was demonstrated that under such conditions, the most promising catalyst is Ni₃S₂. This catalyst can provide an acceptable H₂S conversion of 40% at concentrations as high as 10 vol% of H₂S in the feed. We also observed that Ni₃S₂ catalyst regeneration can be conducted at much milder conditions compared to previously reported in the literature. This contributes to the energy-effectiveness of the potential process. Obtained results contribute to the progress towards upscaling and implementing H₂S decomposition to H₂ and highlight the viable approach–thermocatalytic conversion in the presence of Ni₃S₂.

The DFT calculations reveal three surfaces of Ni₃S₂, (110)_3, (11−1)_S, and (111)_S, with the barrier of hydrogen cleavage from H₂S less than 0.4 eV at the first stage, which makes them most catalytic active, but only (111) S surfaces show the lowest barrier for hydrogen cleavage from HS among these three with an 0.81 eV barrier.