WorksPhysics and Astronomy
Bifunctional Noble Metal-free Ternary Chalcogenide Electrocatalysts for Overall Water Splitting
Abstract
Hydrogen has been identified as a clean, zero carbon, sustainable, and promising energy source for the future, and electrochemical water splitting for hydrogen production is an emission-free, efficient energy conversion technology. A major limitation of this approach is the unavailability of efficient, abundant, inexpensive catalysts, which prompts the need for new catalytic materials. Here, we report the synthesis and electrocatalytic properties of a novel transition metal-based ternary chalcogenide family, LaMS (M = Mn, Fe, Co, Ni). Powder X-ray diffraction confirms the phase purity of these materials, while composition analysis using energy dispersive spectroscopy (EDS) confirms the presence of the stoichiometric ratio of elements in these compounds. X-ray photoelectron spectroscopy (XPS) and X-ray absorption spectroscopy (XAS) were used to study the chemical states on the surface and in bulk, respectively. These materials exhibit bifunctional catalytic activity towards the two half-reactions of the water-splitting process, with LaNiS being the most active material for both hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). The LaMS compounds show long-term stability with negligible change in the overpotential at a constant current density of 10 mA cm over 18 hours of measurements. As compared to the corresponding ternary oxides, the LaMS materials exhibit higher activity and significantly lower Tafel slopes. The ability to catalyze both half-reactions of water electrolysis makes these materials promising candidates for bifunctional catalysts and presents a new avenue to search for high-efficiency electrocatalysts for water splitting.
*Email: [email protected]
Introduction
Hydrogen is a promising energy carrier for future systems that can provide ondemand, emission-free energy at the point of use.[1, 2] Electrochemical water splitting is a clean hydrogen production technique when renewable energy is used as the input. Water splits into hydrogen () at the cathode by HER and oxygen () at the anode by OER under an ideal thermodynamic potential of 1.23 V,[3, 4] but overpotentials are necessary to overcome the barriers for each of these reactions, which increases the associated cost of hydrogen production. Out of these two reactions, OER is kinetically sluggish for being a four-electron transfer reaction compared to HER, which is a two-electron transfer reaction.[5, 6, 7] Hence, the development of catalysts for OER is often considered more challenging compared to HER. Highly efficient electrocatalysts that can reduce the overpotentials for both half-reactions OER and HER are noted as bifunctional catalysts, and they can simplify the design of the electrochemical system for water splitting. Currently, Pt and Ru/ Ir-based materials are the state-of-the-art catalysts that exhibit high catalytic activities for HER and OER, respectively.[3, 7, 8] Their scalable implementation is limited by the high costs, earth abundance, and the production capacity of the precious metals, especially for OER. This has led to increased efforts to find catalysts for water splitting that are free of noble metals.[9, 10, 11, 12, 13]
First-row late transition metal (Mn, Fe, Ni, Co) based compounds have been widely investigated as alternates to the noble metal-based catalysts for their earth abundance and multivalent nature of these elements arising from their -orbital character.[14, 15, 16] A class of materials that has been widely explored in this direction are transition metal oxides. Sustained research on optimizing the composition and control over the microstructure of these materials has led to improvements in their performance.[5, 17, 18, 19, 20] While these materials show good activity towards OER, they typically suffer from issues such as instability and low conductivity.[21, 22, 23] Binary sulfides, selenides, and phosphides have also appeared as candidates for water splitting.[24, 25, 26, 27, 28, 29] Although the performance of these materials is promising, the compositional space available to optimize the balance between activity and stability in these compounds remains limited. To address these limitations in the chemical design phase space, we explore ternary sulfides as candidates for electrocatalysts in water splitting. While ternary chalcogenides based on early transition metals, particularly Ti and Zr, have emerged as a family of electronic materials with potential applications in photovoltaics[30, 31, 32] and infrared optics,[33, 34, 35] our recent study on their electrocatalytic properties revealed large overpotentials and poor efficiency of water splitting.[36] As has been seen in oxides,[18] we presume that ternary chalcogenides based on late transition metals may have more attractive electrocatalytic properties. Among such ternary chalcogenides, (M= Mn, Fe, Co, Ni) powders have been reported to crystallize in the hexagonal space group at temperatures at or above 1100°C from the sulfurization of ternary oxide powders.[37, 38] These studies report electrical and magnetic properties of these materials with semiconducting or metallic properties, which show that such chalcogenides have higher conductivity than equivalent oxides and likely better electrocatalytic properties.
In this article, we report the synthesis of , where M = (Mn, Fe, Ni, Co), by sulfurization of the corresponding oxide powders in a stream of Argon and Carbon disulfide at 1100-1200°C and for the first time, their electrocatalytic properties for water splitting. We verified the phase purity and chemical composition of these materials using powder X-ray diffraction measurements and EDS. Bulk and surface chemical states were analyzed using XAS and XPS measurements, respectively. We made these powders into suspensions and measured their catalytic activity toward water-splitting reactions by preparing electrodes via drop casting.
Results and Discussions:
We synthesized polycrystalline powders of by annealing powders in a stream of Argon saturated with at temperatures ranging from 600 to 1200°C for annealing times ranging from 4 to 30 hours (SI, Figure S1). Based on our observations, powders annealed at temperatures below 1100°C were partially sulfurized with either unreacted oxide left in the product or lanthanum oxysulfide () as the major product. On the other hand, samples annealed at temperatures above 1100°C resulted in pure phases of in the final product. Diffraction patterns obtained from room temperature powder diffraction are shown in Figure 1(a). The diffraction patterns for all the materials are in agreement with the patterns reported in the past.[37] All compounds reported are isostructural and crystallize in a hexagonal symmetry (as shown in Figure 1(b)). The lattice parameters for these materials are reported in Table 1. Chemical compositions were determined for these materials using energy dispersive spectroscopy (EDS) to establish the atomic ratio between different elements. The result obtained was then calibrated using as a reference for the La: M ratio. The atomic ratios obtained after the corrections match well with the expected stoichiometric ratio of 1:1:3 for La: M: S (SI, Figure S2).

Figure 1. (a) Powder X-ray diffraction pattern with reference for LaMnS, LaFeS, LaCoS, and LaNiS. (b) Schematic crystal structure of LaNiS viewed along the -axis, showing hexagonal symmetry (top) and with -axis in plane, showing NiS chains parallel to the -axis (bottom).
| Material | Space Group | Lattice Parameters | |
| LaMnS | (Å) | (Å) | |
| LaFeS | 10.3152 | 5.7526 | |
| LaCoS | 10.3112 | 5.7566 | |
| LaNiS | 10.2988 | 5.7518 | |
| 10.2841 | 5.7573 |
Table 1. Lattice parameters for LaMS materials.
Figure S1. (a) A schematic representation of the CS₂ annealing setup used for the synthesis of LaMS₃ powders, and (b) the temperature profile used for synthesis.
Figure S2. EDS Spectrum of (a) , (b) , (c) and (d) powders. Elemental atomic ratios w.r.t the transition metal elements are shown in the insets.
To analyze the chemical states of and on the surface and in the bulk, we utilized X-ray photoelectron spectroscopy (XPS) and X-ray absorption near-edge structure (XANES), respectively (Figure 2). XPS analysis demonstrates Ni-S bond formation on the surface of the samples, with Ni 2 peaks observed at 870.3 eV (Ni 2) and 853.2 eV (Ni 2), and S 2p peaks at 162.9 eV (S 2) and 161.7 eV (S 2), are characteristic of the Ni-S bond.[39, 40] In addition to this, a low-intensity peak centered around 857 eV is observed, indicating the formation of sulfate () on the surface.[41, 42] XANES analysis further confirms these findings by the observation that the white line intensity of the Ni -edge spectrum for LaNiS falls between the NiS and NiS standards, as shown in Figure 2(c). This suggests the presence of a Ni-S coordination environment, which aligns with the results obtained from XPS and bulk-sensitive X-ray diffraction results. In contrast, LaCoS
Figure 2. XPS and XAS spectra for and . High resolution XPS spectra for (a) Ni and (b) S peaks, and (c) Ni- edge XANES spectra for ; and high resolution XPS spectra for (d) Co and (e) S peaks, and (f) Co- edge XANES spectra for . Spectra for relevant binary sulfides and ternary oxides are shown for comparison.
presents a distinct chemical state on its surface, as revealed by XPS analysis. The Co XPS spectrum identifies both Co–O and Co–S formation. For instance, the Co component is deconvolved into three peaks, including the Co–S peak at 778.4 eV, the Co–O peak at 779.6 eV, and a satellite peak at 781.8 eV. [43, 44, 45] The S spectrum, displaying binding energies at 161.7 and 162.9 eV, confirms the typical presence of , consistent with expectations for Co–S interactions. However, the XANES Co -edge spectrum of is nearly identical to that of CoS, showing a similar pre-edge intensity from the Co to transitions and a comparable white line intensity, which indicates the presence of Co–S as the dominant species. The comparison of the XPS and XANES results highlights the difference between the surface and bulk structures of ; while the surface of exhibits partial oxidation, its bulk retains the Co–S coordination environment.
Testing for the catalytic activity of LaMS, we first characterized the OER electrocatalytic performance of these materials. A standard three-electrode system with nickel foil counter electrode and 1 M KOH electrolyte was used to evaluate the OER electrocatalytic performance of the LaMS deposits on the electrodes achieved from suspensions, and the linear sweep voltammograms (LSVs) measured at 1 mV s and 2000 rpm are shown in Figure 3(a). The slow scan rate ensures steady state condition, and electrode rotation avoids mass transfer limitations and removes gas bubbles from the electrode surface. Based on these measurements, it is evident that the LaNiS and LaCoS powders show good catalytic activity towards OER.
Figure 3. OER activity tests in 1 M KOH solution. (a) Linear sweep voltammograms (LSVs), (b) Tafel plots in the linear region, and (c) stability test at 10 mA cm for 18 hours.
The LaNiS sample shows the highest activity with a low onset potential of 1.49 V, and 10 mA cm current density is achieved at an overpotential () of 373 mV (overpotential for OER = (vs. RHE) − 1.23). Similarly, the cobalt-containing sample (LaCoS) starts to evolve oxygen at around the same potential; however, 50 mV higher overpotential than LaNiS is required to reach 10 mA cm, suggesting a slightly lower catalytic activity. No significant gas evolution is observed for both LaFeS and LaMnS samples, even at 1.72 V. Nonetheless, LaFeS is found to be slightly more active than LaMnS. For instance, at 500 mV overpotential, LaFeS yields 1.9 mA cm current, whereas LaMnS shows only 0.5 mA cm. Thus, a periodic trend in the OER activity as LaNiS > LaCoS >> LaFeS > LaMnS could be noticed.
We also investigated whether the high activity of LaNiS and the activity trend originate from the intrinsic property of the materials or exclusively from the differences in their electrochemical surface areas (ECSAs). The ECSAs and roughness factors obtained from the scan rate dependence of cyclic voltammetry current in the non-Faradaic region (SI, Figure S4) are similar for all the catalysts around and 2.5, respectively (SI, table S2). Hence, the higher activity of LaNiS cannot be attributed to the enhanced number of surface-active sites.

Figure S4. (a) Current vs. Scan rate plots for all the sulfides, and (b) selected oxides and sulfides in the non-Faradaic region of the cyclic voltammograms in 1 M KOH solution.
*Note: 2.303 RT/F = 0.059 V
To further analyze the reaction kinetics and the OER mechanism, Tafel plots are derived from the polarization curves, as shown in Figure 3(b). The Tafel slope is a key parameter that mainly depends on the reaction mechanism and the nature of the rate-determining step (RDS).[46] A lower value of the Tafel slope indicates a faster electrochemical reaction. Here, LaNiS shows a Tafel slope value of 48 mV dec which is much lower than the values for LaCoS (66 mV dec), LaFeS (194 mV dec), and LaMnS (345 mV dec), indicating that LaNiS is the best-suited material among the series for OER, both in terms of low overpotential and small Tafel slope.
In general, the OER mechanism is broadly classified into the conventional adsorbate evolution mechanism (AEM) and the lattice oxygen mechanism (LOM), depending on the origin of oxygen atoms.[47] In the case of AEM, all the oxygen atoms in the evolved oxygen come from the adsorbed oxygen species (OH/HO), whereas in LOM, all or partial oxygen atoms of the oxygen gas are from the lattice oxygen atoms of the catalyst material. Although LOM, particularly in perovskite oxides, can provide higher OER activity, the involvement of lattice oxygen could lead to inherent instability of the catalyst.[48] In the absence of oxygen atoms in the lattice, AEM is the most probable mechanism for the sulfide catalysts tested here. Based on early reports from Bockris et al. on the adsorbate-based OER mechanism on perovskites,[49] the Kobussen Pathway (Table 2) seems to be the most viable for LaNiS, with the RDS being step 2, since regardless of OH coverage, the predicted Tafel slope of 2.303*2RT/3F (40 mV dec) is close to the experimental value (48 mV dec). The results are also consistent with our previously published work on nickel-based OER catalysts in alkaline electrolyte.[50] The OER activity of the catalyst for this type of mechanism is highly dependent on the adsorption strength of the intermediates. However, the adsorption energies of the various intermediates are linearly correlated, which sets the lower limit of the OER overpotential at 370 mV.[51] In the case of LaNiS, the overpotential at 10 mA cm is 373 mV, which is close to the theoretical limit. Further improving the OER activity may require breaking such a scaling relationship, which provides the clue for the next steps in the design of better-performing catalysts.
| Rate Determining Step | |||
| Step | Low OH Coverage | High OH Coverage | |
| Step 1 | -- | ||
| Step 2 | |||
| Step 3 | |||
| Step 4 | |||
| Step 5 |
Table 2. The Kobussen OER Pathway
The long-term operation stability of the catalyst is crucial for their practical application. Under the highly oxidizing condition of OER, most of the catalysts will be corroded, leading to a significant increase in the overpotential. Hence, we tested the stability of the best-performing and catalysts with chronopotentiometry over an 18-hour duration at a constant current density of . As shown in Figure 3c, both and exhibit high stability with no appreciable changes in the required potential during continuous operation. shows slight enhancement in the activity at the end of 18 hours, probably due to improved surface area caused by vigorous gas evolution. The overall change in the overpotential was for both and during continuous oxygen evolution at for 18 hours.
HER is usually studied in an acidic medium due to the abundant supply of the reactant species ( or ) and relatively fast kinetics. Hence, we first examined the HER catalytic activity and stability of the materials in the series in . The linear sweep voltammograms measured at scan rate and 2000 rpm electrode rotation for all the sulfides are shown in Figure 4(a); the required overpotentials (overpotential for HER = ) for HER current densities () are 340, 375, 558, and 655 mV for , , , and , respectively. Although the onset potential of hydrogen evolution is similar around for both and catalysts, shows much better performance at a more negative potential () due to its lower Tafel slope compared to (as discussed below). Overall, a periodic change in the HER activity of with M-site cation could be observed (i.e., Ni Co Fe Mn), which is similar to that in OER and is maintained even after the normalization to ECSAs. Thus, shows the highest OER and HER intrinsic activity among all the catalysts tested here. According to the Sabatier principle, the binding energy of the catalyst with the reactant ( in this case) should be optimum for high catalytic activity.[52] In perovskite oxides of the general form , the transition metal ions located at the M-sites are often considered the active sites for the HER.[53] Therefore, the periodic change in the activity could be attributed to a gradual variation in the binding strength of the ion with transition metal cation at the M-site. This is also consistent with the Tafel slope analysis discussed later, which suggests weak adsorption of as the rate-limiting process. It is therefore anticipated that further enhancement in the activity could be achieved by tuning the binding strength using partial doping at the A-site or M-site, straining the surface, or creating more anion vacancies. In addition to the highest activity, exhibits the least Tafel slope of 79 mV and high stability during 18 hours of electrolysis at 10 mA with only 3 mV increase in the overpotential (Figure 4(b) and Figure 4(c)). Other catalysts are also stable, with a potential change in the range of 1-5 mV during continuous operation at 10 mA for 18 hours. The Tafel slope values are 147, 116, and 117 mV for , , and , respectively.
Figure 4. HER activity tests in 0.5 M and 1 M KOH. (a) and (d) Linear sweep voltammograms, (b) and (e) Tafel plots in the linear region, and (c) and (f) stability test at 10 mA cm for 18 hours.
In general, HER in acidic medium follows,
Volmer:
Tafel:
Heyrovsky:
Here, “M” refers to an empty surface site of the catalyst, and the “M-H” refers to the absorbed hydrogen atom. Theoretically (at 25°C and charge transfer coefficient, = 0.5), if the Volmer step (electrosorption of ) is the RDS, a Tafel slope of 120 mV is expected, whereas 30 and 40 mV are predicted for Tafel and Heyrovsky steps as the RDSs. The measured Tafel slopes in all the catalysts except for are close to 120 mV , suggesting the Volmer step as the RDS.
Similar values have been reported for the weakly proton binding transition metals such as Au or Ag, which shows the Volmer step as the RDS, followed by either electrochemical Heyrovsky or chemical Tafel steps.[54] The Tafel step occurs only when the distance between the surface adsorbed hydrogen is shorter than the Van der Waals radius of the two adsorbed hydrogen atoms. Conversely, the Volmer-Heyrovsky mechanism dominates when the active sites are diluted, and the distance between two adsorbed hydrogen is larger than the Van der Waals radius, which is more likely to occur in the present case.[55] Nevertheless, the Tafel slope obtained in the case of LaNiS doesn’t match with any of the predicted values, but it is comparable with the values reported for Ni-Co-S,[56] Co−S,[57] Ni−S,[40] and MoS.[58]
We also tested the HER activity in alkaline electrolytes. Compared to acidic electrolytes, water splitting in alkaline media can operate with non-noble metals-based catalysts and provide better stability. However, the activity of HER in alkaline media, even for noble metals, is usually 2-3 orders of magnitude lower than in acids.
In the case of HER in acidic solution, as discussed above, the proton adsorption binding energy plays a key role. However, in alkaline conditions, the activity is controlled by the balance of three major criteria: (i) water dissociation to produce free in solution, (ii) the adsorption of on the active sites, and (iii) poisoning of active sites by adsorption.[59] HER activity is more sluggish in alkaline medium primarily due to additional energy needed to break the strong covalent bond in the Volmer step rather than the weak dative bond of ions in acids.
Volmer step (alkaline medium):
Hence, the electrocatalysts that can dissociate water efficiently and have optimum binding strength with hydrogen will exhibit high HER activity in alkaline condition.
Interestingly, the HER activity of in alkaline medium is comparable to that in acid with an overpotential requirement of 362 mV at 10 mA , whereas , , and require 450, 478, and 640 mV, respectively at the same current density (Figure 4(d)). The Tafel slope values are 110, 130, 104, and 124 mV for , , , and , consistent with the Volmer step as the RDS (Figure 4(e)). Since water dissociation and subsequent formation of adsorbed hydrogen is the RDS in alkaline medium, the excellent HER activity of indicates its ability to dissociate water efficiently. This is also supported by the reports from Markovic et al. on improving the HER activity of metals in alkaline electrolyte by modifying the surface with water dissociation catalyst.[60] In addition, exhibited remarkable stability with only 0.2 mV change in the overpotential during 18 hours of continuous operation at a current density of 10 mA (Figure 4(f)). and also show a negligible increase in overpotential of 0.35 mV and 1.7 mV . Interestingly, shows some activation, leading to a decrease in the required overpotential at 10 mA by 0.08 V in 18 hours.
To further illustrate the advantages of sulfurization, we compared the OER performance of LaCoS and LaNiS with their parent oxides (i.e., LaCoO and LaNiO) under identical conditions in 1 M KOH. Figure 5(a) suggests that sulfides are more active towards OER than oxides with LaCoO and LaNiO requiring 53 and 15 mV higher overpotentials at 10 mA cm than LaCoS and LaNiS, respectively. In addition, LaCoO and LaNiO show relatively higher Tafel slopes of 90 mV dec and 79 mV dec, respectively. These results clearly suggest that sulfides are better OER catalysts than their corresponding oxides, and among all the catalysts tested here, LaNiS requires the least overpotential at 10 mA cm current density and exhibits the smallest Tafel slope and excellent durability in 1 M KOH. The OER activity of LaNiS is also comparable to those of many reported oxides and sulfides.[9, 10]
Figure 5. Linear sweep voltammograms (LSVs) comparing selected sulfides with corresponding oxides. (a) LSVs comparing OER performance in 1 M KOH, and (b) and (c) LSVs comparing HER performance in 0.5 M HSO and 1 M KOH, respectively.
Similarly, we also compared the HER performance of sulfides with oxides in both alkali and acidic media. The oxide catalysts, LaCoO and LaNiO, require more than 250 mV higher overpotential than their corresponding sulfides, LaCoS and LaNiS at 10 mA cm, both in 0.5 M HSO and 1 M KOH solutions (Figure 5(b) and Figure 5(c)), which indicates the advantageous role of sulfurization towards HER in a wide range of pH. It is also observed that the difference in activity for the same catalyst between acidic and alkaline media is higher in the case of oxides than sulfides. For instance, exhibits overpotential values of 340 and 362 mV at 10 mA in acidic and alkaline media, implying only a 22 mV increase in the overpotential on switching the electrolytes. On the other hand, shows a substantial increase in overpotential of around 170 mV at 10 mA when the electrolyte is changed from acidic to alkaline. This again can be attributed to the higher efficiency of the sulfides to dissociate water in the Volmer step of alkaline HER than oxides. Furthermore, sulfides are electronically more conducting than oxides.[37] Since carbon is thermodynamically stable under the hydrogen evolution potential, HER testing of catalysts was done by adding 20 wt% SuperP carbon. Although the high surface area of carbon ensures electrical connection between individual particles, carbon plays a relatively less significant role in the case of sulfides compared to oxides. For example, when HER testing was performed without carbon, the required overpotential for 10 mA is increased by 145 mV for , whereas the increase is only 48 mV for (SI, Figure S6). This suggests that, unlike the case of oxides, the addition of carbon may not be crucial for the sulfides, especially when tested as nanoparticles or thin films.

Figure S6. Linear sweep voltammograms of and , with and without SuperP carbon addition in 0.5 M solution at 1 mV and 2000 rpm. All potential values are iR corrected.
Conclusion
In conclusion, we have introduced a new family of materials, complex ternary transition metal chalcogenides, as catalysts for water splitting. We synthesized a series in this family of complex ternary transition metal chalcogenides (LaMS) using CS sulfurization of corresponding ternary oxides at high temperatures. These materials, when tested as catalysts for the two half-reactions of water splitting, show good activity towards OER in alkaline medium and HER in both alkaline and acidic media. Among all the materials in the series, LaNiS exhibits the best activity, with low overpotentials and low Tafel slope values, while exhibiting excellent long-term stability for both HER and OER. When compared to corresponding oxides, these materials demonstrate better activity in terms of lower overpotentials and smaller Tafel slopes. This study opens up an opportunity to explore the wider compositional space offered by complex chalcogenides for electrocatalysts and shows a key step towards achieving optimum activity for water splitting using noble metal-free sulfide catalysts.
Methods
Synthesis of LaMS powders: We synthesized the polycrystalline powders of LaMS using the following approach. First, precursor polycrystalline powders of LaMO were synthesized using the standard synthesis methods reported for the corresponding ternary oxides. We used high-temperature solid-state reaction method to synthesize LaCoO and LaFeO. First, LaO was heated to 600°C for 10 hours before use to get rid of any hydroxide impurities. Then, binary oxides LaO (Alfa Aesar, 99.99%) and CoO (Sigma Aldrich, 99.99%), and FeO (Alfa Aesar, 99.9%) in corresponding molar ratios were mixed together in an agate mortar and pestle till a uniform powder mixture was attained. After that, the powder mixtures were packed in an alumina crucible and heated to 1300°C with a ramp rate of 100°C h, held for 48 hours, and allowed to cool down naturally by turning off the furnace. In case of LaNiO and LaMnO, citrate-nitrate sol gel synthesis was used where La(NO).6HO (Alfa Aesar, 99.99%), Ni(NO).6HO (Alfa Aesar, 99.9985%), and Mn(NO).xHO (ProChem Inc., 99.999%) were used as the precursors. Nitrates of relevant precursors in appropriate molar ratios were dissolved in deionized water, along with citric acid monohydrate (CHO.HO) (Sigma Aldrich, 99.0%) such that the molar ratio of metal nitrates to citric acid is 1:2. The solution thus obtained was dried at 120°C for 24 hours on a hot plate, followed by annealing at 700°C for 4 hours, with a heating/cooling rate of 300°C h. Next, LaMS powders were synthesized by annealing the LaMO powders in a carbon disulfide annealing furnace. LaMO powders were loosely packed in an alumina crucible and loaded in a tube furnace of 1” dia. Samples were heated to 1100-1200°C, with a ramp rate of 10 °C min, and held for 4-30 hours, and Argon was bubbled through CS (Alfa Aesar, 99.9%), with the flow rate being 10-15 sccm. After the hold time, the furnace was turned off, and the samples were allowed to cool down naturally.
Powder Characterizations: Structural characterization using powder diffraction was carried out using a Bruker D8 Advance diffractometer using Cu K wavelength at room temperature. Energy dispersive X-ray spectroscopy measurements were done in a Nova NanoSEM 450 Field Emission Scanning Electron Microscope, equipped with an OXFORD Ultim Max detector using an acceleration voltage of 20kV. The measurements were made using 6mm diameter pellets made from LaMS powders using YLJ-100E 100T Electric Hydraulic Press from MTI. X-ray photoelectron spectroscopy (XPS) measurements were performed using a Kratos Axis Ultra DLD - X-ray Photoelectron Spectrometer, equipped with a magnetic immersion lens and charge neutralization system with the spherical mirror and concentric hemispherical analyzers. Powder samples were pressed into 6mm pellets for measurement. The monochromatic X-ray source used was the Al K line at 1486.6 eV, and a pressure of Torr was maintained. Survey scans were carried out to collect the low-resolution spectra in the energy range of 0-1200 eV, with a step size of 1 eV, and a pass energy of 160 eV. High-resolution spectra were collected with a step size of 0.1 eV in the specific energy region of interest, with a pass energy of 40 eV. Data analysis was done using CasaXPS software. X-ray absorption near edge spectroscopy (XANES) measurements were done at the Stanford Synchrotron Radiation Lightsource (SSRL) at the 4-1 beamline. Samples were prepared by sandwiching the powders of relevant materials between two layers of Kapton tape. Measurements were done at liquid nitrogen temperature at the transition metal -edge in fluorescence mode using a Lytle-type fluorescence-yield ion chamber detector. Energy for each material was calibrated using the respective standard metallic foil sample. Data processing was done using the Athena software of the Demeter package.
Electrochemical Characterizations: All electrochemical measurements were performed using a VersaSTAT multi-channel potentiostat (AMETEK Scientific Instruments) in a three-electrode configuration with 1 M KOH or 0.5 M HSO aqueous solution as the electrolyte. High-purity water (18.2 MOhm cm) was used to prepare the electrolyte solutions. The catalysts loaded gold RDE (5 mm diameter) and glassy carbon RDE (5 mm diameter) were used as the working electrodes for the OER and HER, respectively. A graphite rod counter electrode and Hg/HgSO (MSE in 0.5 M HSO) reference electrode were used in acidic medium, whereas nickel foil counter and Hg/HgO (MMO in 1 M KOH) reference electrodes were used for the studies in alkaline medium. Prior to the catalyst ink preparation, all the materials were ground gently using a mortar and pestle. Then, 10 mg of the catalyst powder was added to 1 mL of DMSO solvent containing 10 L of 5 wt % Nafion binder. In the case of HER studies, 20 wt % SuperP carbon (Alfa Aesar) was also added unless otherwise noted. The mixture was vortex mixed followed by sonication for 2 hours at 30 min intervals to get a homogeneous catalyst ink. 20/ 30 L of the ink was drop-cast onto the surface of a polished RDE and dried at 80°C for HER/OER. The resulting catalyst loading was roughly 1 and 1.5 mg cm for HER and OER respectively. The HER and OER activity were measured using linear sweep voltammetry (LSV) at 1 mV s scan rate and electrode rotation of 2000 rpm. Stability measurements were carried out at 10 mA cm for 18 hours at 2000 rpm. Electrochemical surface area (ECSA) measurements were done using cyclic voltammetry in the non-Faradaic region at 25, 50, 100, 200, and 400 mV s scan rates. Electrochemical impedance spectroscopy (EIS) measurements were performed at the desired dc potential with a sinusoidal excitation voltage of 5-10 mV in the frequency range of 100 kHz to 0.1 Hz. Prior to this measurement, the electrode was held at the potential for 15 min. The resulting impedance spectra were analyzed with ZSimpWin software using suitable equivalent circuit model. All potential values are corrected to account for the uncompensated solution resistance between the working and reference electrodes and converted to the RHE scale.
Acknowledgements
This work was supported by the University of Southern California (USC) Provost New Strategic Directions for Research Award and The Ershaghi Center for Energy Transition, Viterbi School of Engineering at USC. The authors gratefully acknowledge the use of facilities at the Core Center for Excellence in Nano Imaging (CNI) at USC for the results reported in this manuscript. Use of the Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515. We acknowledge Tom Czyszczon-Burton at USC for helping with XPS measurements. We acknowledge helpful discussions with Prof. Shaama Sharada and Prof. Brent Melot of USC.
Conflict of Interest
The authors declare no conflict of interest.
Carbon disulfide Annealing Setup
The schematic for the annealing setup used for the sulfurization of powders to synthesize powders is shown in Figure S1. On the inlet size, a three-way valve is used to switch between Argon and + Argon gas lines. The + Argon line has a glass bubbler filled with liquid, which, when used, bubbles Argon gas through the liquid column and carries saturated Argon gas through the quartz tube over the sample. On the exhaust side, the gas is passed through a gas washing bottle filled with 2M aqueous solution of that neutralizes the excess vapors.
Chemical Composition Analysis by SEM-EDS
Energy dispersive spectroscopy (EDS) was used to determine the stoichiometric ratio of elements to determine chemical composition. Measurements were made using 6mm pellets of powders prepared using an MTI YLJ-100E 100T electric hydraulic press and a 900 MPa CIP (cold isostatic press) die set at ~830 MPa for 5 min. The La: M ratios were calibrated using measurements. All the measurements give ~1:1:3 ratios for La: M: S.
XPS and XANES Measurements
XPS survey scans were done for LaCoS and LaNiS, and the obtained spectra, indicating the relevant element peaks, are shown in Figure S3 and (b), for LaCoS and LaNiS, respectively. The high-resolution XPS spectrum for the La peak in the case of LaCoS is shown in Figure S3. The fitted peak positions obtained are 835.2 and 838.7 eV for La , and 852 and 855.5 eV for La . These values are very close to the peak position values obtained by fitting the La + Ni spectra for LaNiS, where it is 835.2 and 838.6 eV, and 852 and 855.4 eV for La and respectively.

Figure S3. Survey spectra for (a) LaCoS, and (b) LaNiS; and (c) high resolution XPS spectrum for La peak for LaCoS.
In XANES data processing, the edge energy () was determined by taking the energy of the first large peak from the first derivative of the absorption coefficient vs energy spectra. The values for all the materials reported in Figure 2(c) and (Figure 2f) are tabulated below.
| Material | E (eV) (Ni-K edge) | Material | E (eV) (Co-K edge) |
| NiS | 8338.4 | CoS | 7714.2 |
| NiS | 8338 | LaCoS | 7714.1 |
| LaNiS | 8338.3 | LaCoO | 7724.6 |
| LaNiO | 8349.1 |
Table S1. XAS Edge Energy (E) values
Electrochemical Surface Area (ECSA) Measurements
ECSA measurements confirm that the trend observed in the activity of LaMS materials for OER in 1 M KOH is due to the intrinsic property of the catalysts. The very high surface area of LaNiO could be attributed to its nanoparticle morphology. A specific capacitance value of 40 F cm was used for the ECSA calculation in accordance with electrocatalysis literature in alkaline medium.1
| Catalyst | Capacitance (F) | ECSA = Capacitance / 40 (cm) | RF = ECSA / Geometrical area |
| LaNiS | 19.2 | 0.48 | 2.45 |
| LaCoS | 20.5 | 0.51 | 2.60 |
| LaFeS | 19.6 | 0.49 | 2.50 |
| LaMnS | 20.3 | 0.50 | 2.55 |
| LaNiO | 363 | 9.1 | 46.5 |
| LaCoO | 18.3 | 0.46 | 2.34 |
Table S2. Capacitance, Electrochemical Surface Area (ECSA), and Roughness Factor (RF) of all the catalysts in 1 M KOH electrolyte.
LaMS Activity towards OER in Acidic Medium

Figure S5. OER tests in 0.5 M HSO solution. (a) Linear sweep voltammograms of all the sulfides at 1 mV s and 2000 rpm, and (b) corresponding Tafel plots in the linear region. All potential values are iR corrected.
The electrocatalytic OER performance of LaMS compounds was tested in 0.5 M HSO solution. Unlike the alkaline medium, no significant activity is observed for any of the catalysts in the acidic medium. The polarization curves also show a broad, intense current peak prior to the onset of OER, indicating the oxidation of the catalyst. The Tafel slopes are notably higher for LaNiS and LaCoS in acidic electrolytes compared to the alkaline medium, with values of 129 mV dec for LaNiS and 141 mV dec for LaCoS.
Effect of Carbon Addition on HER Performance
The effect of carbon addition on the HER performance of both oxides and sulfides was studied in 0.5 M . A reduction of activity without carbon in terms of increased overpotential is observed for both and , but the increase is more significant in the case of compared to .
Electrochemical Impedance Spectroscopy (EIS) Measurements
Nyquist plots for the LaNiS and LaCoS catalysts during OER at 346 mV overpotential indicate a semicircle (Figure S7). An equivalent circuit consisting of connected in series with a parallel combination of and (inset in Figure S7) was used to model the spectrum. Here is solution resistance, is constant phase element for the double layer capacitance and is charge transfer resistance for the OER.[2] The LaFeS and LaMnS catalysts exhibit an additional semicircle at high frequency region (Figure S7). An equivalent circuit with two parallel R-C segments connected in series with the solution resistance was used to model the spectrum (inset in Figure S7). Here, the additional R-C circuit at high frequencies, denoted () and (), shows minimal dependance on potential, as shown in Figure S7, and may be due to anodic surface layer.[3] Such EIS behavior is reported for Fe and Mn based electrolytes in alkaline media.[4, 5, 6] EIS on Fe and Mn based electrodes are sensitive to an anodic surface film formation, which is not detected on Ni or Co based electrodes. The LaNiS catalyst shows the least charge transfer resistance value of 97 whereas it is 137, 244 and 341 for LaCoS, LaFeS and LaMnS, respectively. Since the charge transfer resistance is inversely related to rate constant, EIS data suggests highest OER activity for the LaNiS, consistent with the catalytic activity trend observed in LSV measurements.

Figure S7. EIS measurements for LaMS materials. (a) EIS during OER in 1 M KOH at 346 mV overpotential. Inset shows the equivalent circuits for LaNiS and LaCoS (top), and LaFeS and LaMnS (bottom). (b) Variation in R, R, and R with overpotential for LaFeS electrode during EIS for OER in 1 M KOH. (c) EIS during HER in 0.5 M HSO at 350 mV overpotential. Inset shows the Nyquist plot of LaNiS highlighting the two semicircles. (d) EIS during HER in 1 M KOH at 375 mV overpotential. Inset shows the equivalent circuit used for HER in both HSO and KOH. (e) Variation in R, R and R with overpotential during EIS for LaNiS during HER in 0.5 M HSO.
In the case of EIS during HER in both acidic and alkaline electrolytes, two semicircles are observed for all the catalysts (Figure S7). The porosity of the electrode surface is responsible for the high-frequency semicircle, which is independent of the applied potential (Figure S7).[2] On the contrary, the low-frequency semicircle shows a strong dependence on the potential, suggesting a charge transfer process at the electrode/ electrolyte interface. An equivalent circuit with two parallel R-C segments connected in series with another resistance is used (inset in Figure S7). At 350 mV overpotential in 0.5 M HSO, the charge transfer resistance values for the HER are 72, 560, 1380, and 1953 for LaNiS, LaCoS, LaFeS, and LaMnS, respectively. Similarly, the charge transfer resistance values of 50, 576, 970, and 4400 are obtained for LaNiS, LaCoS, LaFeS, and LaMnS, respectively at 375 mV overpotential in 1 M KOH. The EIS results suggest the highest HER activity for LaNiS, followed by LaCoS, LaFeS, and LaMnS in the order of decreasing activity.
SI References
[1] C. C. L. McCrory, S. Jung, J. C. Peters, T. F. Jaramillo, J. Am. Chem. Soc. 2013, 135, 16977.
[2] A. Irshad, R. Elizalde-Segovia, B. S. Jayathilake, B. Zayat, S. R. Narayanan, J. Electrochem. Soc. 2022, 169, 110528.
[3] M. E. G. Lyons, M. P. Brandon, Journal of Electroanalytical Chemistry 2009, 631, 62.
[4] T. Kessler, J. R. Vilche, M. Ebert, K. Jüttner, W. J. Lorenz, Chemical Engineering & Technology 1991, 14, 263.
[5] B. Messaoudi, S. Joiret, M. Keddam, H. Takenouti, Electrochimica Acta 2001, 46, 2487.
[6] X. Zou, Y. Wu, Y. Liu, D. Liu, W. Li, L. Gu, H. Liu, P. Wang, L. Sun, Y. Zhang, Chem 2018, 4, 1139.
References
- [1]F. Dawood, M. Anda, G. M. Shafiullah, International Journal of Hydrogen Energy 2020, 45, 3847.
- [2]K. Mazloomi, C. Gomes, Renewable and Sustainable Energy Reviews 2012, 16, 3024.
- [3]A. Ursua, L. M. Gandia, P. Sanchis, Proc. IEEE 2012, 100, 410.
- [4]M. Wang, Z. Wang, X. Gong, Z. Guo, Renewable and Sustainable Energy Reviews 2014, 29, 573.
- [5]J. Suntivich, K. J. May, H. A. Gasteiger, J. B. Goodenough, Y. Shao-Horn, Science 2011, 334, 1383.
- [6]J. Zhang, Q. Zhang, X. Feng, Adv. Mater. 2019, 31, 1808167.
- [7]S. Wang, A. Lu, C.-J. Zhong, Nano Convergence 2021, 8, 4.
- [8]J. Chi, H. Yu, Chinese Journal of Catalysis 2018, 39, 390.
- [9]L. Du, L. Xing, G. Zhang, M. Dubois, S. Sun, Small Methods 2020, 4, 2000016.
- [10]J. S. Kim, B. Kim, H. Kim, K. Kang, Adv. Energy Mater. 2018, 8, 1702774.
- [11]N. Dubouis, A. Grimaud, Chem. Sci. 2019, 10, 9165.
- [12]D. Strmcnik, P. P. Lopes, B. Genorio, V. R. Stamenkovic, N. M. Markovic, Nano Energy 2016, 29, 29.
- [13]G. Zhao, K. Rui, S. X. Dou, W. Sun, Adv. Funct. Mater. 2018, 28, 1803291.
- [14]S. Li, E. Li, X. An, X. Hao, Z. Jiang, G. Guan, Nanoscale 2021, 13, 12788.
- [15]J. Wang, X. Yue, Y. Yang, S. Sirisomboonchai, P. Wang, X. Ma, A. Abdula, G. Guan, Journal of Alloys and Compounds 2020, 819, 153346.
- [16]M. Ju, X. Wang, X. Long, S. Yang, CrystEngComm 2020, 22, 1531.
- [17]W. T. Hong, M. Risch, K. A. Stoerzinger, A. Grimaud, J. Suntivich, Y. Shao-Horn, Energy Environ. Sci. 2015, 8, 1404.
- [18]D. Liu, P. Zhou, H. Bai, H. Ai, X. Du, M. Chen, D. Liu, W. F. Ip, K. H. Lo, C. T. Kwok, S. Chen, S. Wang, G. Xing, X. Wang, H. Pan, Small 2021, 17, 2101605.
- [19]Q. Ji, L. Bi, J. Zhang, H. Cao, X. S. Zhao, Energy Environ. Sci. 2020, 13, 1408.
- [20]H. Wang, M. Zhou, P. Choudhury, H. Luo, Applied Materials Today 2019, 16, 56.
- [21]L. C. Seitz, C. F. Dickens, K. Nishio, Y. Hikita, J. Montoya, A. Doyle, C. Kirk, A. Vojvodic, H. Y. Hwang, J. K. Norskov, T. F. Jaramillo, Science 2016, 353, 1011.
- [22]K. J. May, C. E. Carlton, K. A. Stoerzinger, M. Risch, J. Suntivich, Y.-L. Lee, A. Grimaud, Y. Shao-Horn, J. Phys. Chem. Lett. 2012, 3, 3264.
- [23]Y. Guo, Y. Tong, P. Chen, K. Xu, J. Zhao, Y. Lin, W. Chu, Z. Peng, C. Wu, Y. Xie, Adv. Mater. 2015, 27, 5989.
- [24]J. Yang, H. S. Shin, J. Mater. Chem. A 2014, 2, 5979.
- [25]C. Tsai, K. Chan, J. K. Nørskov, F. Abild-Pedersen, Surface Science 2015, 640, 133.
- [26]X. Xia, L. Wang, N. Sui, V. L. Colvin, W. W. Yu, Nanoscale 2020, 12, 12249.
- [27]X. Peng, Y. Yan, X. Jin, C. Huang, W. Jin, B. Gao, P. K. Chu, Nano Energy 2020, 78, 105234.
- [28]J. Kibsgaard, C. Tsai, K. Chan, J. D. Benck, J. K. Nørskov, F. Abild-Pedersen, T. F. Jaramillo, Energy Environ. Sci. 2015, 8, 3022.
- [29]H. Du, R.-M. Kong, X. Guo, F. Qu, J. Li, Nanoscale 2018, 10, 21617.
- [30]Y.-Y. Sun, M. L. Agiorgousis, P. Zhang, S. Zhang, Nano Lett. 2015, 15, 581.
- [31]S. Niu, H. Huyan, Y. Liu, M. Yeung, K. Ye, L. Blankemeier, T. Orvis, D. Sarkar, D. J. Singh, R. Kapadia, J. Ravichandran, Adv. Mater. 2017, 29, 1604733.
- [32]S. Niu, D. Sarkar, K. Williams, Y. Zhou, Y. Li, E. Bianco, H. Huyan, S. B. Cronin, M. E. McConney, R. Haiges, R. Jaramillo, D. J. Singh, W. A. Tisdale, R. Kapadia, J. Ravichandran, Chem. Mater. 2018, 30, 4882.
- [33]J. Wu, X. Cong, S. Niu, F. Liu, H. Zhao, Z. Du, J. Ravichandran, P. Tan, H. Wang, Adv. Mater. 2019, 31, 1902118.
- [34]S. Niu, G. Joe, H. Zhao, Y. Zhou, T. Orvis, H. Huyan, J. Salman, K. Mahalingam, B. Urwin, J. Wu, Y. Liu, T. E. Tiwald, S. B. Cronin, B. M. Howe, M. Mecklenburg, R. Haiges, D. J. Singh, H. Wang, M. A. Kats, J. Ravichandran, Nat. Photon. 2018, 12, 392.
- [35]B. Zhao, M. S. B. Hoque, G. Y. Jung, H. Mei, S. Singh, G. Ren, M. Milich, Q. Zhao, N. Wang, H. Chen, S. Niu, S.-J. Lee, C.-T. Kuo, J.-S. Lee, J. A. Tomko, H. Wang, M. A. Kats, R. Mishra, P. E. Hopkins, J. Ravichandran, Chem. Mater. 2022, 34, 5680.
- [36]N. Humphrey, A. Tsung, S. Singh, A. Irshad, B. Zhao, S. Narayan, J. Ravichandran, S. Mallikarjun Sharada, ChemPhysChem n.d., n/a, e202300953.
- [37]T. Murugesan, S. Ramesh, J. Gopalakrishnan, C. N. R. Rao, Journal of Solid State Chemistry 1981, 38, 165.
- [38]T. Takahashi, T. Oka, O. Yamada, K. Ametani, Materials Research Bulletin 1971, 6, 173.
- [39]R. Karthikeyan, D. Thangaraju, N. Prakash, Y. Hayakawa, CrystEngComm 2015, 17, 5431.
- [40]N. Jiang, L. Bogoev, M. Popova, S. Gul, J. Yano, Y. Sun, J. Mater. Chem. A 2014, 2, 19407.
- [41]H. W. Nesbitt, D. Legrand, G. M. Bancroft, Phys Chem Min 2000, 27, 357.
- [42]S. Mehmood, F. U. Khan, M. N. Shah, J. Ma, Y. Yang, G. Li, W. Xu, X. Zhao, W. He, X. Pan, Front. Chem. 2022, 10, 971859.
- [43]Y. Zhou, D. Yan, H. Xu, J. Feng, X. Jiang, J. Yue, J. Yang, Y. Qian, Nano Energy 2015, 12, 528.
- [44]A.-M. Cao, J.-S. Hu, H.-P. Liang, W.-G. Song, L.-J. Wan, X.-L. He, X.-G. Gao, S.-H. Xia, J. Phys. Chem. B 2006, 110, 15858.
- [45]J. F. W. Mosselmans, R. A. D. Pattrick, G. van der Laan, J. M. Charnock, D. J. Vaughan, C. M. B. Henderson, C. D. Garner, Phys Chem Minerals 1995, 22, 311.
- [46]T. Shinagawa, A. T. Garcia-Esparza, K. Takanabe, Sci Rep 2015, 5, 13801.
- [47]X. Xie, L. Du, L. Yan, S. Park, Y. Qiu, J. Sokolowski, W. Wang, Y. Shao, Adv. Funct. Mat. 2022, 32, 2110036.
- [48]H. Chen, C. Lim, M. Zhou, Z. He, X. Sun, X. Li, Y. Ye, T. Tan, H. Zhang, C. Yang, J. W. Han, Y. Chen, Advanced Science 2021, 8, 2102713.
- [49]J. O. Bockris, T. Otagawa, J. Phys. Chem. 1983, 87, 2960.
- [50]B. Zayat, D. Mitra, S. R. Narayanan, J. Electrochem. Soc. 2020, 167, 114513.
- [51]M. J. Craig, G. Coulter, E. Dolan, J. Soriano-López, E. Mates-Torres, W. Schmitt, M. García-Melchor, Nat Commun 2019, 10, 4993.
- [52]H. Ooka, J. Huang, K. S. Exner, Front. Energy Res. 2021, 9, 654460.
- [53]Md. S. Alom, C. C. W. Kananke-Gamage, F. Ramezanipour, ACS Omega 2022, 7, 7444.
- [54]H. Prats, K. Chan, Phys. Chem. Chem. Phys. 2021, 23, 27150.
- [55]A. Raveendran, M. Chandran, R. Dhanusuraman, RSC Adv. 2023, 13, 3843.
- [56]A. Irshad, N. Munichandraiah, ACS Appl. Mater. Interfaces 2017, 9, 19746.
- [57]Y. Sun, C. Liu, D. C. Grauer, J. Yano, J. R. Long, P. Yang, C. J. Chang, J. Am. Chem. Soc. 2013, 135, 17699.
- [58]D. Merki, H. Vrubel, L. Rovelli, S. Fierro, X. Hu, Chem. Sci. 2012, 3, 2515.
- [59]N. Mahmood, Y. Yao, J. Zhang, L. Pan, X. Zhang, J. Zou, Advanced Science 2018, 5, 1700464.
- [60]N. Danilovic, R. Subbaraman, D. Strmcnik, K.-C. Chang, A. P. Paulikas, V. R. Stamenkovic, N. M. Markovic, Angew. Chem. 2012, 124, 12663.