Coupling biomass-derived substrate oxidation with the HER: hexagonal NiS for low-voltage, high-efficiency hydrogen production

Abstract

Developing a sustainable process for ultrapure hydrogen production is essential for achieving a carbon-free society. This study presents a coupling of cheap biomass-derived substrates' oxidation reaction and the hydrogen evolution reaction to produce hydrogen via water electrolysis utilizing hexagonal NiS, synthesized via a rapid solution combustion method, as an efficient and cost-effective electrocatalyst. The water electrolyzer, employing biomass-derived substrates such as glucose, glycerol, and ethylene glycol, showed superior performance, particularly with glucose. The glucose-based electrolyzer exhibited a low onset potential of 1.27 V vs. RHE and an overpotential of 1.34 V vs. RHE at 400 mA cm⁻² in a three-electrode configuration with a high turnover frequency (TOF) value of 0.076 s⁻¹ and mass activity of 1.5 A g⁻¹. The glucose-assisted electrolyzer (H-cell), composed of hexagonal NiS, required 1.57 V at 10 mA cm⁻² while the glucose-assisted single stack cell, composed of NiS, required a small potential of 1.33 V at 10 mA cm⁻². Further, the glucose-assisted stack cell exhibited 10.3% energy saving competence compared to the conventional electrolyzer composed of hexagonal NiS with more than 90% faradaic efficiency towards H₂ and formate formation. In agreement with the experimental findings, DFT calculations conducted on the NiS (110) surface to determine the adsorption mechanism showed that glucose has a high adsorption energy value of −9.286 eV, higher than those of glycerol (−8.794 eV) and ethylene glycol (0.557 eV). These findings highlight the potential of the glucose-assisted electrolyzer composed of hexagonal NiS for sustainable hydrogen production.

---

1. Introduction

Advancements across various global sectors, including modern lifestyles, rapid population growth, and the industrial revolution, have led to an unprecedented demand for energy.¹ Fossil fuels have been a primary source of energy for mankind for ages. Considering the fact that excess use of fossil fuels has not only brought them to the verge of depletion but also has left a negative mark on the ecosystem, with their major contribution to the growing greenhouse gases and global warming.² Hydrogen, the smallest molecule with high energy density and the least effect on the environment, is considered to be an upcoming energy source that will replace fossil fuels.³ Despite the potential for hydrogen production from methods like steam reforming and oxidation of hydrocarbons, which are in turn derived from fossil fuels, they continue to pose environmental issues. Consequently, prioritising the use of renewable sources for clean hydrogen production is essential.⁴ Water electrolysis (WE) is one of the simplest methods for producing hydrogen. Nevertheless, the widespread adoption of water electrolysis technology is severely hindered by the reliance on expensive electrocatalysts such as platinum (Pt), ruthenium (Ru), and iridium (Ir), which are renowned for their superior electrochemical activity.⁵ Furthermore, the oxygen evolution reaction (OER) demands more energy owing to the sluggish and multielectron transfer characteristics. The combination of the OER and HER may form explosive H₂/O₂ in conventional water electrolysis, posing a threat to public safety, which also needs to be tackled.

As is well known, the vital goal of the OER is to generate electrons, which are then accepted by the cathode to produce H₂. Since the efficiency, performance and energy consumption of water electrolysis are influenced by the kinetics of the OER, utilizing the oxidation of organic substrates in place of the OER at the anode can reduce the energy barrier by participating in the oxidation process, which in turn reduces the energy consumption in overall water electrolysis. Different oxidative species like alcohols,⁶ biomass,⁷ urea,⁸ and hydrazine having favourable kinetics for oxidation⁹ reactions, are reported in place of the OER to generate hydrogen, along with value-added products. However, the process generates gaseous byproducts such as CO, CO₂, and NO, which are released into the atmosphere.¹⁰ Nonetheless, using earth-abundant electrode materials and substrates derived from sustainable resources, such as biomass or their derived products, holds promise for fabricating energy-efficient water electrolyzers to produce both hydrogen and value-added chemicals simultaneously at much lower energy consumption. Glucose, glycerol, and ethylene glycol are of specific interest because of their high activity in electrooxidation, low cost, minimal toxicity and inflammability, making them easy to handle. Glucose, in particular, is abundant as a byproduct of various processes and can be directly derived from plant photosynthesis.^11–13 Similarly, glycerol is readily available from biodiesel production,¹⁴ while ethylene glycol can be obtained through the catalytic conversion of biomass-derived cellulose.¹⁵

The cheap, sustainable, earth-abundant transition metals as electrocatalysts show the same or even better electrocatalytic activity than that of noble metals. In particular, metal sulfides, a class of semiconducting compounds, have gained significant attention in recent years due to their adjustable physical and optical properties, rich redox active sites, tunable bandgaps, and excellent stability.¹⁶ Leveraging these attributes, nickel sulfides have been widely employed as electrocatalysts in conventional water electrolysis processes due to their rich valence states, inexpensiveness, unique d-electron configuration, and low-toxicity constituents.¹⁷

Nickel sulfide (NiS) is a promising metal chalcogenide due to its low overpotential and enhanced stability, yet its practical application in large-scale water electrolysis remains limited.¹⁸ Various synthesis methods, including hydrothermal,¹⁹ solvothermal,²⁰ electrodeposition²¹ and chemical vapour deposition,²² are employed to synthesize NiS. These reported methods are time-consuming, quite complex, environmentally not benign and have limited physicochemical properties to realize enhanced water electrolysis. Meanwhile, the solution combustion method offers the advantages of self-driven energy sustenance, rapid formation of crystalline materials in a short reaction time (typically in minutes) and capability of producing a wide variety of materials.²³ The synthesis of NiS using energy-efficient protocols and its use as an efficient electrode for the fabrication of water electrolyzers utilizing the oxidation of biomass-derived substrates and the hydrogen evolution reaction has garnered significant research interest.

Thus, this work presents a simple approach to synthesizing highly electroactive hexagonal nickel sulfide (NiS) using a self-driven energy-sustained solution combustion method. This method involving the exothermic reaction between nickel nitrate and thioacetamide results in electrochemically active crystalline NiS nanoparticles. The electrochemical performance of combustion-processed hexagonal NiS in alkaline electrolytes (KOH) is investigated both in the presence and absence of glucose, glycerol, and ethylene glycol for the first time. Additionally, the first-principles density functional theory is employed to analyze the adsorption energies of glucose, glycerol, and ethylene glycol molecules onto the NiS (110) surface, considering both chemisorption and physisorption mechanisms. Notably, a stacked single cell composed of hexagonal NiS exhibits superior electrochemical performance in the presence of glucose compared to conventional water electrolyzers.

2. Experimental

2.1 Synthesis of the hexagonal nickel sulfide electrocatalyst

Analytical grade nickel nitrate hexahydrate (Ni(NO₃)₂·6H₂O), thioacetamide (C₂H₅NS), potassium hydroxide (KOH), glucose (C₆H₁₂O₆), glycerol and ethylene glycol were purchased from SD Fine-Chem Limited. All the chemicals are used without further purification.

Briefly, 0.86 mmol of nickel nitrate hexahydrate was dissolved in a 100 mL beaker containing 6 mL of deionized water. To this, 15 mmol of thioacetamide was added and stirred for 10 min. The resulting homogeneous solution was kept in a preheated furnace at 500 °C for 10 min. The beaker was later removed from the furnace and cooled to room temperature naturally. The resultant product was washed with 100 mL of deionized water and extracted via centrifugation. Finally, the centrifuged product was dried at 50 °C and used for further studies.

2.2 Characterization of the NiS electrocatalyst

The crystal structure and purity of the synthesized powder sample are evaluated using an X'Pert PRO X-ray diffractometer. Using Al Kα radiation (1486.6 eV) and a Kratos AXIS Ultra DLD X-ray photoelectron spectrometer (XPS), the elemental composition is determined. An Apreo 2 S Thermo fisher (FEI) scanning electron microscope and Talos F200S G2 transmission electron microscope are used to examine the microstructure. The pore size distribution and surface area are determined by Barrett–Joyner–Halenda (BJH) and Brunauer–Emmett–Teller (BET) (Micrometrics, 3-FLEX, Norcross, GA, USA). A Bruker Avance 300 (300 mHz) NMR spectrometer is used to assess the organic substrate oxidized product, where dimethyl sulfoxide is used as an internal standard.

2.3 Electrode fabrication and electrochemical performance assessment

The electrocatalytic ink was prepared by dispersing 3 mg of hexagonal NiS in a blend of 0.3 mL water and 0.1 mL ethanol. The resulting mixture was sonicated for 15 min to ensure the formation of evenly distributed hexagonal NiS. The obtained hexagonal NiS ink was coated on a nickel foam (NiS/NF) with a geometric surface area of 1 × 1 cm² and was subsequently dried at room temperature. The dried electrode was later used for electrolysis experiments. The electrochemical activity of NiS was assessed using a cyclic voltammogram (CV) and linear sweep voltammogram (LSV) recorded at a scan rate of 10 mV s⁻¹ in a 1.0 M potassium hydroxide (KOH) solution with and without glucose, glycerol and ethylene glycol. Initially, the experiments were conducted in a single chamber using NiS/NF as the working electrode, with platinum wire and silver/silver chloride (Ag/AgCl) as the counter and reference electrodes, respectively. Further, the experiments were conducted in an anion exchange membrane (AEM) stack electrolyser using NiS/NF as both the anode and cathode. Electrochemical impedance spectroscopy (EIS) data were acquired with an applied potential over a frequency range from 1 kHz to 10 mHz. The long-term stability of the materials was assessed by recording chronopotentiometry. The recorded potentials from the three-electrode system were converted to the reversible hydrogen electrode (RHE) scale using Nernst's equation.

E_RHE = E_appl + (0.059 × 14) + E_Ag/AgCl (1)

2.4 Computational setup

Density functional theory (DFT) calculations were computed using the Cambridge Sequential Total Energy Package (CASTEP) code,²⁴ employing a pseudopotential approach. Geometry optimization was performed using the exchange–correlation functional of the Perdew–Burke–Ernzerhof (PBE) functional.²⁵ A full structural optimization and energy minimization calculation was performed with a cutoff energy and Monkhorst–Pack k-points of 500 eV and 4 × 4 × 1, respectively.²⁶ The Broyden–Fletcher–Goldfarb–Shanno (BFGS) minimization scheme algorithm was utilized for structural optimization. The DFT-D2 method developed by Grimme et al.²⁷ was employed to correct the effect of vdW interactions for evaluating the adsorption energies. 10⁻⁵ and 10⁻² were used as the electronic and ionic optimization requirements for structural geometry, respectively. To investigate the adsorption strength of the molecules (glucose, glycerol, and ethylene glycol) the NiS (110) surface was allowed to relax. The adsorption energy is expressed as:

E_ads = E_mol/surf − (E_mol + E_surf) (2)

where E_mol/surf, E_mol, and E_surf represent the total final energy of the molecule-NiS (110) surface, free molecule, and the clean surface of NiS (110), respectively. Adsorption energy values that are higher indicate less favoured (less stable) molecule adsorption, whereas those that are lower (more negative) suggest more favourable (more stable) adsorption.

3. Results and discussion

3.1 Structural and morphological characterization

A rapid and straightforward synthesis method was employed to produce NiS nanoparticles through a self-sustained exothermic solution combustion process. This approach involves the reaction between nickel nitrate and thioacetamide, resulting in a highly exothermic reaction and the generation of numerous gaseous molecules at elevated temperatures, leading to the formation of crystalline NiS nanoparticles within a remarkably short duration of only 10 min. The synthesis protocol intentionally incorporates a higher thioacetamide fuel content to maintain a sufficiently low combustion temperature necessary for the crystallization of NiS.²⁸ The formation mechanism of NiS was also elucidated based on the reactions described in the literature.

C₂H₂NS + 2O₂ → 2CO₂ + NH₃ + H₂S

Ni(NO₃)₂ → NiO + N₂ + O₂

NiO + H₂S → NiS + H₂O

However, it was observed that NiSO₄ was also formed as a byproduct along with NiS as shown in Fig. S1.† The formation of NiSO₄ may be explained by the following equations:²⁹

2NiS + 3O₂ → 2NiO + 2SO₂

NiO + SO₂ + ½O₂ → NiSO₄

As is known that NiSO₄ is water soluble, the NiSO₄ formed was removed by washing it with water. The crystallinity and phase formation of the prepared NiS nanoparticles were confirmed using the X-ray diffraction pattern displayed in Fig. 1a. The diffraction peaks observed in Fig. 1a match well with the hexagonal phase of crystalline NiS with lattice parameters a, b, and c, measured at 3.44 Å, 3.44 Å, and 5.11 Å, respectively [JCPDS 89-7141].

Fig. 1 (a) PXRD spectrum of the synthesized NiS nanoparticles; (b) XPS spectrum of Ni 2p and (c) XPS spectrum of S 2p presenting the surface chemical oxidation states of the Ni and S in the synthesized NiS nanoparticles.

A detailed XPS analysis was carried out to determine the surface composition and electronic/valence states in the prepared NiS powder sample. Fig. S2a† displays the XPS survey spectrum of NiS, demonstrating the presence of Ni, S, and O, which are in good agreement with EDX results. The Ni 2p spectrum of NiS nanoparticles presented in Fig. 1b contains Ni 2p_3/2 and Ni 2p_1/2 spin–orbit doublets and the satellite peaks. The peaks observed at 855.5 and 873.4 eV were respectively assigned to 2p_3/2 and 2p_1/2 of Ni²⁺ (ref 30) while the peaks at 879.5 and 861.1 eV can be assigned to the shake-up satellite peaks of Ni 2p_1/2 and Ni 2p_3/2 respectively.³¹Fig. 1c presents the S 2p spectrum where the peak at a binding energy of 163.7 eV is assigned to S²⁻ bonded to Ni, and that at 165.1 eV is attributed to S_n²⁻ of polysulfide. The peaks at higher binding energies of 168.4 eV and 169.7 eV were assigned to the high valence state of SO₄²⁻ ions, formed due to oxidation of the surface of S²⁻ in air.^32,33 Further, the NiS nanoparticles' synthesis is carried out under ambient conditions, and thus, it is expected to have a small amount of nickel present in the form of NiO on the surface. The high-resolution XPS spectrum of O 1s presented in Fig. S2b† exhibits peaks at binding energies of 528.7 eV, 531.0 eV, and 532.9 eV. The peak at the binding energy of 528.7 eV is assigned to O²⁻ bonded to metal, 531.0 eV represents the hydroxide ion, and 532.9 eV is due to surface adsorbed water.^34,35

The microstructure of the NiS surface plays an important role in promoting electrochemical reactions. As shown in Fig. 2a, NiS exhibits an agglomerated micro–nanostructure. The micro–nanostructure is composed of Ni, S, and O (Fig. 2b). Further, the TEM image in Fig. 2c shows that the micro–nanostructure is composed of spherical-shaped particles of size 5–10 nm and the HRTEM presented in Fig. 2d and e exhibit an interlayer d spacing of 0.17 nm corresponding to the (110) plane of NiS. The surface area and mesoporous structure of the NiS are assessed using a BET study, where the solution combustion-derived NiS exhibits a surface area of 3 m² g⁻¹ with a type IV isotherm and mesoporous nature (Fig. S3†). It is anticipated that the unique micro–nanostructured rough surface with mesoporous nature of NiS offers enhanced exposed active sites for electrochemical reactions. The largely exposed active sites improve the mobility of the electrolyte and thereby speed up the faradaic redox reactions.

Fig. 2 (a) SEM image, (b) EDX spectrum, (c) TEM image, and (d and e) HRTEM images of the NiS prepared at 500 °C/10 min. Spherical-shaped nanostructure, small crystallite size, and interplanar spacing reveal the formation of NiS.

3.2 Electrocatalytic activity of hexagonal NiS towards the OER, and glucose, glycerol and ethylene glycol oxidation

Initially, the electrochemical performance of hexagonal NiS was evaluated using cyclic voltammograms (CV) and linear sweep voltammograms (LSV) recorded in 1.0 M KOH electrolyte with and without glucose, glycerol and ethylene glycol. The CV of NiS presented in Fig. S4† reveal that the anodic current increases with an increase in glucose, glycerol, and ethylene glycol concentrations, suggesting that hexagonal NiS was electrochemically active towards the oxidation of these organic substrates. To verify whether the electrochemical activity arises from the Ni foam, which is used as the substrate or NiS electrocatalyst, LSVs were recorded using a hexagonal NiS-coated glassy carbon (GC) electrode in KOH with and without glucose, glycerol, and ethylene glycol. The resulting voltammograms are presented in Fig. S5,† where it is observed that the onset potential significantly shifted to a lower potential in the presence of glucose, glycerol and ethylene glycol compared to KOH. Additionally, the NiS-coated GC exhibits higher current density in the presence of glucose, glycerol and ethylene glycol, which clearly demonstrates the superior electrocatalytic activity of NiS towards the oxidation of these organic substrates.

LSV profiles recorded in 1 M KOH using bare nickel foam and NiS-coated nickel foam are presented in Fig. 3a–c. The bare nickel foam exhibits an onset potential of 1.63 V vs. RHE in 1 M KOH, while NiS exhibits an onset potential of 1.56 V vs. RHE towards the oxygen evolution reaction. In contrast, the bare nickel foam electrode exhibits an onset potential of 1.34 V, 1.38 V and 1.39 V vs. RHE in 1 M KOH containing glucose, glycerol and ethylene glycol, respectively, with slightly increased current density. However, the hexagonal NiS electrode exhibits the lowest onset potential of 1.27 V vs. RHE in the presence of 50 mmol of glucose, followed by glycerol (1.36 V vs. RHE) and ethylene glycol (1.40 V vs. RHE), with very high current density compared to bare nickel foam. Further, the hexagonal NiS electrode exhibits a much lower overpotential of 1.28 V vs. RHE and 1.34 V vs. RHE at the current densities of 10 mA cm⁻² and 400 mA cm⁻², in 50 mmol of glucose. Similarly, the NiS electrode requires overpotentials of 1.37 V vs. RHE and 1.72 V vs. RHE towards glycerol and 1.41 V vs. RHE and 1.96 V vs. RHE towards ethylene glycol at the same current densities.

Fig. 3 Linear sweep voltammograms of NiS and bare Ni foam: (a) glucose, (b) glycerol, (c) ethylene glycol, (d) Tafel plot, (e) impedance spectrum and (f) chronopotentiometric (V × t) curves recorded in a 1.0 M KOH solution of 50 mmol glucose, glycerol and ethylene glycol. The black, red, blue and pink colored curves are indexed to KOH, glucose, glycerol and ethylene glycol, respectively.

However, the increased glucose, glycerol, and ethylene glycol concentrations increase the overpotential (Fig. S6†). The higher concentration results in a larger number of oxidized products, which may be adsorbed on the active sites of hexagonal NiS, which hinders the electrochemical activity with increased overpotentials. The overpotential observed for glucose at 10 mA cm⁻² decreased by 0.290 V vs. RHE compared to the overpotential observed for the OER, indicating the competence of glucose oxidation in water electrolysis. The observed overpotentials were compared with the reported literature (Table S1†), where NiS exhibits competitive results. Overall, the glucose oxidation reaction over the hexagonal NiS electrocatalyst provides an enhanced current density by consuming less overpotential than ethylene glycol and glycerol. It is important to note here that the overpotential for glucose is much lower than that of the electrolyzers composed of ethylene glycol and glycerol, validating the proficient oxidation reaction.

The enhanced performance of NiS in KOH solution containing glucose, glycerol, and ethylene glycol was further substantiated by turnover frequency, mass activity, Tafel slope and charge transfer resistance values. The NiS exhibits high TOF and mass activity of 0.0764 s⁻¹ and 1.5 A g⁻¹, respectively, in the presence of glucose. In contrast, the NiS exhibits a comparatively low TOF and mass activity of 0.000445 s⁻¹ and 0.0116 A g⁻¹ in the presence of glycerol and 0.00001710 s⁻¹ and 0.0005 A g⁻¹ in the presence of ethylene glycol. The Tafel slope values were obtained by solving the Tafel equation, η = b × log(j + a), where j and b denote the current density and Tafel slope, while η and a represent the overpotential and Tafel constant, respectively. The observed Tafel slope values are presented in Fig. 3d, where hexagonal NiS exhibits Tafel slope values of 86 mV dec⁻¹, 94 mV dec⁻¹, 99 mV dec⁻¹ and 111 mV dec⁻¹ for glucose, glycerol, ethylene glycol and KOH, respectively, which confirms that NiS has fast kinetics in the presence of glucose relative to glycerol, ethylene glycol and KOH. Additionally, the lowest charge transfer resistance value found for glucose (1.07 Ω) as compared to glycerol (1.3 Ω), ethylene glycol (3.2 Ω) and KOH (4.7 Ω) further confirms that hexagonal NiS exhibits improved kinetics during glucose oxidation (Fig. 3e). The faster kinetics is due to the ultrafast charge transfer accompanied by excellent contact, and rapid charge transfer between NiS and the nickel substrate.³⁶ Moreover, chronopotentiometry assessment at a fixed current density of 10 mA cm⁻² over 24 hours revealed minimal potential degradation, indicating good stability of an organic substrate coupled electrolyzer (Fig. 3f). Thus, glucose oxidation presents a viable alternative to the OER in alkaline advanced water electrolysis.

It is important to note that extracting the oxidation current arising only from glucose oxidation is difficult, as it is interfered with by water oxidation. However, glucose oxidation and oxygen evolution reactions can be seen separately in the present case, indicating a higher selectivity of hexagonal NiS towards glucose. However, a similar trend is not observed with glycerol and ethylene glycol. The observed trend is due to the difference in the rate of oxidation of glucose, glycerol and ethylene glycol over the Ni(100) surface. DFT study reveals that the oxidation of glucose is more favorable on the NiS(110) surface, and thus the current increases rapidly. In contrast, the oxidation of glycerol and ethylene glycol is slow compared to that of glucose and thus the current saturates. Additionally, the viscosity of the electrolyte and the rate of desorption of oxidized products from the electrode surface also affect the rate of electrochemical reaction.³⁷

3.3 Electrocatalytic activity of hexagonal NiS towards the HER in the presence of glucose, glycerol and ethylene glycol

The impact of glucose, glycerol, and ethylene glycol on the HER catalyzed by hexagonal NiS was evaluated using LSV. As shown in Fig. 4a, hexagonal NiS exhibits low onset potential and an overpotential of −0.19 V vs. RHE, −0.24 V vs. RHE, −0.37 V vs. RHE, and −0.28 V vs. RHE in the presence of 1 M KOH, glucose, glycerol and ethylene glycol, respectively, at a current density of 400 mA cm⁻². One can observe that the slightly increased overpotential in the presence of glucose, glycerol and ethylene glycol may be due to the possibility that the oxidized product may adsorb on the electrocatalyst and block the active sites.³⁸ The observed results are consistent with the HER results observed for Co(OH)₂@HOS in the presence of methanol³⁷ and Ni₇S₆ in the presence of table sugar, orange juice and glucose.³² Tafel slopes provide important information on the kinetics of the HER where hexagonal NiS shows Tafel slope values of 85 mV dec⁻¹, 103 mV dec⁻¹, 97 mV dec⁻¹ and 78 mV dec⁻¹ in glucose, glycerol and ethylene glycol and 1 M KOH respectively (Fig. 4b), indicating faster catalytic kinetics and a lower energy barrier for electron transfer reactions.

Fig. 4 Effect of glucose, glycerol and ethylene glycol on the HER performance of NiS; (a) linear sweep voltammograms and (b) corresponding Tafel slopes. The black, red, blue and pink colored curves are indexed to KOH, glucose, glycerol and ethylene glycol, respectively.

3.4 Designing an electrolyzer (hexagonal NiS‖hexagonal NiS) by coupling the HER with glucose, glycerol and ethylene glycol oxidation

Replacing the sluggish OER in water electrolysis with thermodynamically more favourable ones by adding reactive substrates significantly reduces the consumption of electrical energy.^10,39 In addition to reducing energy consumption, it offers other benefits, such as the production of value-added products, and also avoids the formation of reactive/explosive gaseous mixtures. A two-electrode electrolyzer was assembled using NiS/NF as an anode and cathode, demonstrating the catalyst's suitability for water electrolysis with integrated glucose, glycerol and ethylene glycol oxidation. The observed results are compared with a standard electrolysis cell, employing NiS as an anode and cathode and 1.0 M KOH as the electrolyte. Polarization curves obtained from both setups are depicted in Fig. 5a.

Fig. 5 (a) LSV curves of NiS nano–microstructure used as both the anode and cathode (hexagonal NiS‖hexagonal NiS) in the presence of KOH-black, glucose-, glycerol-, ethylene glycol-. Post stability (b) XRD spectrum, and high resolution XPS spectra of (c) Ni 2p and (d) S 2p.

The standard alkaline electrolysis cell required a high cell voltage of 1.75 V to achieve a current density of 10 mA cm⁻², indicating a higher electrical energy consumption. The high energy consumption is due to the complex and sluggish multi-electron transfer, multi-step oxygen evolution reaction.⁴⁰ The multielectron transfer OER over hexagonal NiS can be represented as shown in Scheme 1.

Scheme 1 Mechanism of the multielectron transfer reaction in the OER.

The addition of 50 mmol glucose, glycerol and ethylene glycol to the electrolyte notably reduced the applied voltage to 1.57 V, 1.63 V and 1.61 V for the same current density. The % of energy saving competence of the proposed glucose-assisted electrolyzer is examined using the following equation

The % energy saving competence is observed to be 10.3% at the current density of 10 mA cm⁻² when glucose oxidation is substituted in place of the OER, manifesting the energy-saving hydrogen generation.

Post-stability characterizations, including XRD, XPS and SEM analysis, were used to evaluate the stability of the NiS electrode. Post-stability characterization of the XRD result revealed that there were no shifts in peak positions or the appearance of new diffraction peaks apart from the nickel foam diffraction peaks represented by , indicating that the crystalline structure of hexagonal NiS remained unchanged before and after the stability test (Fig. 5b). The post-stability high-resolution XPS spectra of Ni 2p, S 2p and O 1s remain almost consistent with the pre-test results (Fig. 5c and d). Nevertheless, the new peak at 853.2 eV can be attributed to Ni³⁺, indicating slight oxidation of Ni²⁺. Meanwhile, the emergence of a new disulfide peak at 161.6 eV in the post-stability XPS spectrum of S 2p indicates that some amount of sulfide has undergone oxidation due to the prolonged stability conditions.⁴¹ Similarly, the morphological analysis using SEM revealed that NiS retains an agglomerated morphology without undergoing any significant modification (Fig. S7†). The elemental composition and mapping images revealed the presence of Ni, S, and O even after the stability of a long duration of 24 h. The consistency in the XRD diffraction patterns, XPS spectra and SEM morphology before and after the stability test confirm that NiS retains its structural integrity even under rigorous conditions.

The selectivity of the NiS towards the oxidation of ethylene glycol, glycerol and glucose was assessed using ¹H NMR spectra. Glucose, glycerol and ethylene glycol give only formate (formic acid) according to Scheme 2 (ref. 42–44) confirmed by ¹H NMR spectra presented in Fig. S9–S11.†

Scheme 2 Representation of the oxidized products of glucose, ethylene glycol and glycerol.

On the other hand, it is reported that the electrochemical oxidation of glucose over the MnO₂/Ti electrocatalyst in alkaline electrolytes results in the formation of glucaric and gluconic acids.⁴⁵ A similar example from the literature indicates that lactate, glycerate, glycolate and formate were the oxidized products of glycerol in an alkaline electrolyte over the ZnFe_xCo_2−xO₄ electrocatalyst⁴⁶ and glycolate, oxalate and formate for ethylene glycol over Co and Ni.⁴⁷

3.5 Designing a glucose-assisted single-stack cell

Due to the excellent results of hexagonal NiS when glucose oxidation is coupled with the HER a single stacked cell with a working electrode area of 2.2 cm × 2.2 cm² is constructed. A representation of the single-stack cell assembly is shown in Fig. 6a and b.

Fig. 6 (a and b) Schematic representation of the glucose-assisted stack cell electrolyser, (c) LSV curves and (d) stability profile recorded at low and high current density.

The LSV obtained using a stack cell in the presence and absence of glucose is presented in Fig. 6c, where hexagonal NiS exhibits an overpotential of 1.33 V at 10 mA cm⁻² in the presence of glucose, while the same cell exhibits 1.47 V at 10 mA cm⁻² in 1.0 M KOH. The percentage energy efficiency for the glucose-assisted single-stack cell is found to be 10% compared to the conventional single-stack cell operated in 1.0 M KOH. Chronoamperometry was carried out for 24 h to check the stability of NiS in the presence of glucose at low and high current densities (Fig. 6d). The NiS exhibits good stability over 24 h at low current density, while it exhibits stable current over 9 h at high current density. The electrolyte was collected after 9 h and subjected to ¹H NMR analysis, where the signal pertaining to glucose is not observed, indicating complete oxidation of glucose. Thus, the decrease in current density after 9 h is due to the non-availability of glucose.

Faradaic efficiency for hydrogen was calculated by the water displacement method employing hexagonal NiS(+)‖NiS(−) as both the anode and cathode along with an anion exchange membrane as a separator at a constant potential of 1.6 V. By dividing the amount of H₂ gas generated empirically by the amount of H₂ gas produced theoretically, the faradaic efficiency (FE) of the hydrogen evolution process aided by glucose, glycerol and ethylene glycol is calculated. The charge transmitted across the electrode determines the theoretical quantity of H₂ gas.⁴⁸

FE for H₂ = V_exp/V_theor = V_exp/[(Q/(n × F)) × V_gas] × 100

where Q, N and F signify total charge, number of moles of product formed and Faraday constant (96 485 C mol⁻¹) respectively. The number of electrons required to produce 1 molecule of H₂ is denoted by n. V_gas represents the molar volume of gas, which is 24.1 L mol⁻¹ at 293 K. A faradaic efficiency of 90%, 88% and 84% was observed for hydrogen evolution in the presence of glucose, glycerol and ethylene glycol (Fig. S8†).

Similarly, the electrolyte from the anode compartment was collected after 20 minutes and subjected to ¹H NMR analysis to determine the faradaic efficiency of the oxidized products. The faradaic efficiency for formate formation is calculated based on the following (ref. 49).

FE for formate = N/[Q/(n × F)] × 100

where Q, N and F signify total charge, number of moles of product formed and Faraday constant (96 485 C mol⁻¹). The number of electrons (n) required by 1 mole of glucose to produce 1 mole of formate is 2.⁵⁰ Likewise, 1 mole of glycerol produces 3 moles of formate, requiring 8 electrons, for which n is 2.67 (ref. 51) and ethylene glycol requires 3 electrons.⁵² The number of moles of product formed was determined by the integrated area of the peak signal for formate from the ¹H NMR spectra using the equation below.
where I_product and I_standard are the integrated area of the peak signal for the product and internal standard, respectively. n_product and n_standard are the number of protons contributing to the product and internal standard peak. The faradaic efficiency of formate formation is found to be 98%, 25.6% and 30.3%, respectively, for glucose, glycerol and ethylene glycol.

3.6 Computational studies

3.6.1 Adsorption by physisorption. All the molecules of glucose, glycerol, and ethylene glycol were exposed to the NiS (110) surface by being positioned above the surface, and a complete structural optimization was carried out. Fig. S12 (a–c)† presents the adsorption of optimized atomic structures for glucose, glycerol, and ethylene glycol molecules. The adsorption mechanism was investigated by calculating the adsorption energy using expression (2). All the adsorption energy values were found to be negative, demonstrating thermodynamic stability and an exothermic process. Moreover, the stronger adsorption was characterized by the highest negative adsorption energy. The adsorption energy values of glucose, glycerol, and ethylene glycol molecules are observed to be as follows: E_ads = −9.286 eV, E_ads = −8.794 eV and E_ads = 0.557 eV, respectively. The adsorption energy value corresponding to glucose adsorption was 0.5 eV times larger in comparison to the adsorption of glycerol. Further, it was discovered that the ethylene glycol molecule is the least stable and unfavourable with positive adsorption energy. The adsorption energy stability follows the trend:

E^glucoes_ads > E^glycerol_ads > E^{ethylene glycol}_ads

This implies that the adsorption of glucose molecules was more stable and favorably than the adsorption of glycerol and ethylene glycol molecules on the NiS (110) surface.

3.6.2 Adsorption by chemisorption. Adsorption by chemisorption was considered for investigating the adsorption of glucose, glycerol, and ethylene glycol molecules at different adsorption sites, such as top and bridge sites. Fig. 7 presents the adsorption of the optimized structures of (a–c) glucose, (d–f) glycerol, and (g–i) ethylene glycol molecules. After the optimization, from the adsorption configuration it was observed that the adsorbed molecule binds to the NiS (110) surface via oxygen (molecules), forming Ni–O and S–O interactions. The equilibrium bond lengths between the interacting atoms differed with the adsorbent. The bond lengths for the adsorption of the glucose molecule on the NiS (110) surface atoms were found to be: d_O–Ni = 2.343 Å, d_O–S = 2.637 Å, d_O–Ni/Ni = 2.586 Å, and d_O–Ni/S = 2.543 Å, wherein the bond length of the O–Ni interaction was observed to be shorter, while that of O–S was larger.

Fig. 7 Optimized molecule (glucose, glycerol and ethylene glycol) adsorption on the NiS (110) surface at different positions: (a) glucose-Ni site, (b) glucose-S site, (c) glucose-bridge site, (d) glycerol-Ni site, (e) glycerol-S site, (f) glycerol-bridge site, (g) ethylene glycol-Ni site, (h) ethylene glycol-S site and (i) ethylene glycol-bridge site.

This suggests that the stronger interaction of glucose on the NiS surface was through Ni bonding. In addition, the equilibrium bond lengths for glycerol and ethylene glycol molecules range from 2.328–2.844 Å and 2.023–2.997 Å, respectively. One can observe a slight decrease in the adsorption of the corresponding inter-atomic bond lengths between glucose < glycerol< ethylene glycol.

The effect of adsorbate on the adsorbent was investigated by adsorbing at different surface sites, including top and bridge sites. The interaction strength was determined by calculating the adsorption energy using expression (2). A comparison of the interaction stability of ethylene glycol, glucose, and glycerol at various adsorption sites is presented in Table 1. Negative adsorption energies were observed for all adsorbates, indicating exothermic and spontaneous reactions. Variations in adsorption energy were noted based on the adsorption site. The S site (top) was found to be the least favourable for adsorption interaction and bonding. In contrast, the Ni site demonstrated the most stable and advantageous adsorption of molecules like ethylene glycol, glucose, and glycerol, suggesting a strong attachment to the NiS surface through interaction with Ni atoms.

Table 1 Calculated adsorption energies (E_ads) of glucose, glycerol, and ethylene glycol molecules on the NiS (110) surface at different adsorption sites

Adsorption sites	Glucose	Glycerol	Ethylene glycol
S	−4.436	−6.149	−3.912
Ni	−9.087	−8.572	−7.415
Bridge Ni–Ni	−8.833	−8.291	−6.703
Bridge Ni–S	−8.895	−8.359	−7.236

---

Moreover, a more negative adsorption energy value indicates a stronger contact between the adsorbent and substrate. Glucose exhibited a higher adsorption energy strength on the top site (Ni) by 0.515 eV compared to glycerol and ethylene glycol by 1.672 eV, indicating a greater ease of adsorption on the NiS surface for glucose than glycerol and ethylene glycol.

3.6.3 Density of states. Fig. S13† examines and depicts the partial density of states (PDOS) to provide a deeper insight into the electronic interaction between the adsorbates and NiS (110) surfaces. The PDOS plots of the pure surface and adsorbed species are primarily characterized by -s, -p, and -d orbitals, featuring a pseudo-gap at the Fermi energy (E_F) with a lesser contribution from -s- and -p orbitals at the Fermi level. Upon adsorption, a slight decrease in the magnitude of the -d orbital of the adsorbed surface compared to the clean NiS surface was observed, indicating a reduction in total energy following molecule adsorption. However, with ethylene glycol adsorption, the PDOS plots reveal minimal modifications compared to glucose and glycerol. Initially, a prominent-s orbital peak is observed at approximately −15 eV, but post-adsorption it broadens and merges with the electronic peak. Also, heightened p–d hybridization is noted at around −9 eV towards the EF level, with adsorption leading to a slight increase in peak hybridization compared to the pure surface.

3.6.4 Charge density difference. The charge density difference (CDD) was computed to elucidate the electronic charge distribution within the structure and discern the type of interatomic bonding between specific pairs of atoms. Due to atomic interactions, charge density undergoes redistribution due to electronic hybridization among multiple atomic orbitals. The CDD was utilized to investigate charge and electron distribution across the surface. In Fig. 8, CDD profiles for ethylene glycol, glucose, and glycerol adsorption are depicted, with various colored regions indicating the amount of charge, where yellow denotes depletion and blue signifies accumulation. The findings highlight that charge distribution predominantly occurs at the interaction between O atoms (molecules) and surface atoms. Furthermore, it was observed that the Ni atom donated more charge to the O atom (glucose) compared to glycerol and ethylene glycol adsorption, suggesting a donor–acceptor interaction.

Fig. 8 Charge density redistribution for the adsorbed NiS (110) surface: (a) glucose, (b) glycerol and (c) ethylene glycol; blue: charge depletion and yellow: charge accumulation.

4. Conclusion

A rapid and robust synthesis protocol was proposed for producing highly active hexagonal NiS micro and nanostructures, facilitating the creation of a high-performance organic substrate (glucose, glycerol, and ethylene glycol) oxidation-assisted water electrolyzer to produce hydrogen at reduced electrical energy consumption along with value-added formate. Hexagonal NiS demonstrates superior efficiency in oxidizing glucose at ultra-low overpotentials compared to glycerol and ethylene glycol. Consequently, a glucose-assisted single stack cell composed of hexagonal NiS(+)‖hexagonal NiS(−) requires decreased overpotential and reduced electrical energy consumption compared to a conventional water electrolyzer. DFT studies also revealed that glucose is the most preferred adsorption molecule on the NiS (110) surface as determined by chemisorption and physisorption reaction mechanisms. The results suggest glucose molecules are more readily adsorbed on the NiS surface than glycerol and ethylene glycol. These findings highlight the potential of glucose-assisted electrolyzers to significantly reduce energy consumption and production costs, paving the way for sustainable and cost-effective hydrogen generation technologies.

Data availability

The data that support the findings of this study are available upon reasonable request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors acknowledge the management of REVA University for financial support (RU/R&D/SEED/CHE/2024/09 and RU/R&D/SEED/CHE/2023/22) to carry out this research work. MSS also thanks CSIR for the financial support under the R&D Seed fund (Project No. CSPS24/RDSF/CIMFR/IHP24/03).

References

1. X. Kong, R. Zhang, H. Shi, L. Pan, C. Wang, W. Wang and W. Lv, J. Alloys Compd., 2024, 1002, 175467.
2. P. C. Meenu and S. Roy, Catal. Today, 2024, 439, 114802.
3. F. Liu, J. Dang, C. Zhao, B. Yuan, H. Qiu, Q. Wang, C. Zhang, L. Xiao, H. Miao and J. Yuan, J. Alloys Compd., 2024, 981, 173745.
4. J. Dang, H. Xu, X. Zhang, J. Jian, J. Huang, C. Zhang, Q. Wang, H. Miao and J. Yuan, J. Alloys Compd., 2024, 1004, 175935.
5. S. Anwar, F. Khan, Y. Zhang and A. Djire, Int. J. Hydrogen Energy, 2021, 46, 32284–32317.
6. Y. Li, X. Wei, S. Han, L. Chen and J. Shi, Angew. Chem., 2021, 133, 21634–21642.
7. D. Li, Z. Li, R. Zou, G. Shi, Y. Huang, W. Yang, W. Yang, C. Liu and X. Peng, Appl. Catal., B, 2022, 307, 121170.
8. Z. Lv, Z. Li, X. Tan, Z. Li, R. Wang, M. Wen, X. Liu, G. Wang, G. Xie and L. Jiang, Appl. Surf. Sci., 2021, 552, 149514.
9. R. Guo, Y. Zhang, X. Zhang, H. Yang and T. Hu, Chem. Eng. J., 2024, 491, 152028.
10. L. Du, Y. Sun and B. You, Mater. Rep.: Energy, 2021, 1, 100004.
11. C. Lin, H. Li, P. Zhang, C. Deng, L. Meng, Q. Zhou, S. Wang, J. Wu, C. Liu, J. Tian and Y. Qian, J. Electroanal. Chem., 2020, 861, 113946.
12. S. G. Anastassiadis, World J. Biol. Biotechnol., 2016, 1, 1.
13. C. Zhang, C. Xu, L. Yang, X. Tao, J. Liu, R. Su, X. Meng, R. Yang, D. Xue and K. Zheng, Chem. Eng. J., 2025, 507, 160396.
14. Y.-B. Liu and C.-Y. Chiang, Chem. Eng. J., 2025, 507, 160619.
15. Q. Lin, Y. Wei, W. Liu, Y. Yu and J. Hu, Int. J. Hydrogen Energy, 2017, 42, 1403–1411.
16. S. Chandrasekaran, L. Yao, L. Deng, C. Bowen, Y. Zhang, S. Chen, Z. Lin, F. Peng and P. Zhang, Chem. Soc. Rev., 2019, 48, 4178–4280.
17. B. Fei, Z. Chen, J. Liu, H. Xu, X. Yan, H. Qing, M. Chen and R. Wu, Adv. Energy Mater., 2020, 10, 202001963.
18. Y.-T. Wang, X.-F. He, G.-Y. He, C. Meng, X.-M. Chen, F.-T. Li and Y. Zhou, Crit. Rev. Solid State Mater. Sci., 2023, 48, 502–518.
19. H. Yaldiz, A. C. Kocal, O. Gultepe, E. Gür and F. Atay, Electrochim. Acta, 2025, 526, 146197.
20. Z. Wang, S. Liu, Y. Ma, J. Guo, B. Xin, C. Wang, C. Wang and J. Tong, J. Energy Chem., 2025, 105, 872–884.
21. K. N. Nguyen, T. K. Bui, K. Q. Nguyen and V. Van Pham, Phys. Scr., 2024, 99, 055402.
22. M. Vijayarangan, A. Gayathri, V. Ashok, M. Sangamithirai, J. Jayabharathi and V. Thanikachalam, Energy Fuels, 2024, 38, 1364–1372.
23. S. S. Manoharan, N. R. S. Kumar and K. C. Patil, Mater. Res. Bull., 1990, 25, 731–738.
24. S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. I. J. Probert, K. Refson and M. C. Payne, Z. Kristallogr.–Cryst. Mater., 2005, 220, 567–570.
25. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
26. H. J. Monkhorst and J. D. Pack, Phys. Rev. B, 1976, 13, 5188–5192.
27. S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456–1465.
28. N. Ardebilchi Marand, S. M. Masoudpanah and M. S. Bafghi, Ceram. Int., 2018, 44, 17277–17282.
29. G. Zhao, Y. Zhang, L. Yang, Y. Jiang, Y. Zhang, W. Hong, Y. Tian, H. Zhao, J. Hu, L. Zhou, H. Hou, X. Ji and L. Mai, Adv. Funct. Mater., 2018, 28, 201803690.
30. P. Luo, H. Zhang, L. Liu, Y. Zhang, J. Deng, C. Xu, N. Hu and Y. Wang, ACS Appl. Mater. Interfaces, 2017, 9, 2500–2508.
31. D. D. Yang, M. Zhao, R. D. Zhang, Y. Zhang, C. C. Yang and Q. Jiang, Nanoscale Adv., 2020, 2, 512–519.
32. S. A. Patil, A. C. Khot, K. D. Kadam, H. T. Bui, H. Im and N. K. Shrestha, Inorg. Chem. Front., 2023, 10, 7204–7211.
33. Y. Li, H. Guo, Y. Zhang, H. Zhang, J. Zhao and R. Song, J. Mater. Chem. A, 2022, 10, 18989–18999.
34. H. Zhang, H. Jiang, Y. Hu, H. Jiang and C. Li, Green Energy Environ., 2017, 2, 112–118.
35. S. Bhowmick, M. K. Mohanta and M. Qureshi, Sustainable Energy Fuels, 2021, 5, 6392–6405.
36. X. Peng, L. Wang, L. Hu, Y. Li, B. Gao, H. Song, C. Huang, X. Zhang, J. Fu, K. Huo and P. K. Chu, Nano Energy, 2017, 34, 1–7.
37. K. Xiang, D. Wu, X. Deng, M. Li, S. Chen, P. Hao, X. Guo, J. Luo and X. Fu, Adv. Funct. Mater., 2020, 30, 201909610.
38. K. Dharmaraj, R. Hanna, I. Lauermann, R. Bagacki, F. Xi, E. Kemppainen, R. Schlatmann and S. Calnan, ACS Sustain. Chem. Eng., 2024, 12, 9908–9921.
39. Z. Chen, W. Wei, L. Song and B.-J. Ni, Sustainable Horiz., 2022, 1, 100002.
40. X. Xie, L. Du, L. Yan, S. Park, Y. Qiu, J. Sokolowski, W. Wang and Y. Shao, Adv. Funct. Mater., 2022, 32, 202110036.
41. C. E. Mitchell, D. Santos-Carballal, A. M. Beale, W. Jones, D. J. Morgan, M. Sankar and N. H. de Leeuw, Faraday Discuss., 2021, 230, 30–51.
42. H. Yang, G. Vijaykumar, Z. Chen, J. N. Hausmann, I. Mondal, S. Ghosh, V. C. J. Nicolaus, K. Laun, I. Zebger, M. Driess and P. W. Menezes, Adv. Funct. Mater., 2023, 33, 2303702.
43. A. Herbst and C. Janiak, New J. Chem., 2016, 40, 7958–7967.
44. N. Thakur, D. Mehta, A. Chaturvedi, D. Mandal and T. C. Nagaiah, J. Mater. Chem. A, 2023, 11, 15868–15877.
45. D. Bin, H. Wang, J. Li, H. Wang, Z. Yin, J. Kang, B. He and Z. Li, Electrochim. Acta, 2014, 130, 170–178.
46. J. Sankar, E. N. Onyeozili and E. E. Kalu, ChemEngineering, 2017, 1, 1–14.
47. X.-Y. Ma, H.-Z. Ma, S.-H. He, Y. Zhang, Y.-N. Yi and Y.-Y. Yang, Mater. Today Phys., 2023, 37, 101191.
48. Y. Li, J. Chen, P. Cai and Z. Wen, J. Mater. Chem. A, 2018, 6, 4948–4954.
49. Z. Duan, T. Ren, Q. Mao, H. Yu, K. Deng, Y. Xu, Z. Wang, L. Wang and H. Wang, J. Mater. Chem. A, 2022, 10, 18126–18131.
50. R. Zhao, Y. Wang, G. Ji, F. Zhang, Y. Wang, Y. Zhao, B. Han and Z. Liu, Chem. Eng. J., 2024, 486, 150280.
51. Y. Pei, Z. Pi, H. Zhong, J. Cheng and F. Jin, J. Mater. Chem. A, 2022, 10, 1309–1319.
52. K. Liu, Y. Wang, F. Liu, C. Liu, R. Shi and Y. Chen, Chem. Eng. J., 2023, 473, 145292.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5se00790a

---