2D nanosheet molybdenum disulphide (MoS₂) modified electrodes explored towards the hydrogen evolution reaction

Abstract

We explore the use of two-dimensional (2D) MoS₂ nanosheets as an electrocatalyst for the Hydrogen Evolution Reaction (HER). Using four commonly employed commercially available carbon based electrode support materials, namely edge plane pyrolytic graphite (EPPG), glassy carbon (GC), boron-doped diamond (BDD) and screen-printed graphite electrodes (SPE), we critically evaluate the reported electrocatalytic performance of unmodified and MoS₂ modified electrodes towards the HER. Surprisingly, current literature focuses almost exclusively on the use of GC as an underlying support electrode upon which HER materials are immobilised. 2D MoS₂ nanosheet modified electrodes are found to exhibit a coverage dependant electrocatalytic effect towards the HER. Modification of the supporting electrode surface with an optimal mass of 2D MoS₂ nanosheets results in a lowering of the HER onset potential by ca. 0.33, 0.57, 0.29 and 0.31 V at EPPG, GC, SPE and BDD electrodes compared to their unmodified counterparts respectively. The lowering of the HER onset potential is associated with each supporting electrode's individual electron transfer kinetics/properties and is thus distinct. The effect of MoS₂ coverage is also explored. We reveal that its ability to catalyse the HER is dependent on the mass deposited until a critical mass of 2D MoS₂ nanosheets is achieved, after which its electrocatalytic benefits and/or surface stability curtail. The active surface site density and turn over frequency for the 2D MoS₂ nanosheets is determined, characterised and found to be dependent on both the coverage of 2D MoS₂ nanosheets and the underlying/supporting substrate. This work is essential for those designing, fabricating and consequently electrochemically testing 2D nanosheet materials for the HER.

---

1. Introduction

The increasing scarcity of fossil fuels coupled with the consequences of anthropogenic climate change has produced a global need to find feasible alternatives to energy generation.^1–3 One scenario of change is creating a global hydrogen economy in which energy generation demands are met partially or entirely by hydrogen fuel cells.^4–6 Utilising hydrogen fuels in this manner would cause a dramatic decrease in the anthropogenic greenhouse emissions and ozone precursors released by fossil fuel combustion given that the major product of hydrogen oxidation is H₂O.^4,7 The main barrier to the implementation of such a hydrogen economy is, of course, the production of hydrogen. This requires an energy input that could feasibly be drawn from renewable energy sources such as wind, solar, and wave.^3,8–10

A common method of hydrogen production is the Hydrogen Evolution Reaction (HER) (2H⁺ + 2e⁻ → H₂), which involves the electrocatalytic splitting of water, known to occur via one of two routes; these being either the Volmer–Tafel or the Volmer–Heyrovsky reactions.^11–13 The efficiency of the HER is dependent on the choice of electrocatalyst, with an optimal catalyst having a binding energy for adsorbed H⁺ close to that of the reactant or product.⁷ Currently, the most proficient catalyst for the HER is platinum (Pt) which has a small binding energy for the reaction to occur, resulting in the reaction proceeding at a near zero over-potential.^7,11,14,15 Pt is a precious metal with a low natural abundance of 0.001–0.005 mg kg⁻¹ within the Earth's crust. This, coupled with its high demand, result in Pt having a very high cost.¹⁶ Consequently, current research is focused on finding a cheaper, more sustainable catalyst for the HER whilst maintaining performance and offering a binding energy towards H⁺ close to that of Pt,⁷ thus making a hydrogen based energy economy substantially more feasible.

Carbon based materials have long been utilised as electrodes in a plethora of analytical and industrial electrochemical applications.^17–23 They have the distinct advantage of being comparatively cheap and easily obtainable compared to the traditional noble metal based electrodes, particularly out-performing said metals in numerous significant areas due to carbon's structural polymorphism, chemical stability, wide operable potential windows and relative inert electrochemistry.^17,18 Carbon based electrodes are often used as the supporting material for a plethora of electrocatalytic materials which lower HER onset potentials and thus greatly improve the HER kinetics in comparison to the bare/unmodified carbon based electrode.^1,7,24–28

One example of where modified electrodes have been utilised to improve the HER has been reported by Liu et al.²⁹ who utilised carbon nanotubes decorated with CoP nanocrystals deposited onto GC electrodes with a coverage of 0.285 mg cm⁻² which resulted in the onset potential for the HER to reduce to −40 mV. Other prominent examples from the literature have used graphene as an electrocatalyst.^24,27,30,31 The current challenge within HER research is finding a low-cost, abundantly available, non-polluting catalyst which is capable of matching the HER onset potential observed when Pt is used as a catalyst. Towards this goal, Mo based catalysts have been explored towards the HER. Table 1 presents a literature overview of Mo based electrocatalysts explored towards the HER, which is a combination and adaptation of work presented in papers by Joesen, Li and Ji et al.,^32–34 in addition to recent literature reports. It is evident that MoS₂ and MoSe₂ have shown promising results as HER catalysts. MoS₂ is a semiconductor with a direct band gap of ca. 1.9 eV and excellent charge carrier mobility (reported to be no less than 200 cm² V⁻¹ S⁻¹).³⁵ It has been beneficially implemented in numerous electrochemical applications, such as in transistors, sensors, solar cells, and lithium ion batteries.^1,36,37

Table 1 Comparison of current literature reporting the use of MoS₂ as a catalyst explored towards the HER

Catalyst	Electrode/supporting material	Loading (μg cm^−2)	Electrolyte	HER onset (mV)	Tafel (mV dec^−1)	Ref.
Key: —: Value unknown; GC: glassy carbon; MCN: mesoporous carbon nanospheres; RGO: reduced graphene oxide; NAS: nano-assembled structures; PET: polyethylene terephthalate; EPPG: edge plane pyrolytic graphite; BDD: boron doped diamond; SPE: screen-printed graphite electrode.a Optimised loading.b Mechanical bent tensile-strain-induced two dimensional MoS2 nanosheets.
MoS2	Au(111)	—	0.5 M H2SO4	∼−150	55–60	83
Narrow sheet MoS2	GC	280	0.5 M H2SO4	−103	49	45
MoS2/CoSe2	GC	∼280	0.5 M H2SO4	11	36	84
MoS2/MCN	GC	190	0.5 M H2SO4	−100	41	85
MoS2/RGO	GC	280	0.5 M H2SO4	−100	41	27
MoS3	GC	32	1 M H2SO4	∼−100	54	86
Fe- MoS3	GC	—	1 M H2SO4	—	39	87
MoS2 nanosheets	GC	48	0.5 M H2SO4	∼−150 to −200	70	34
2D MoS2 nanoplates	GC	280^a	0.5 M H2SO4	−93	42	88
MoS2	GC	∼8.5	0.5 M H2SO4	—	86	72
MoS2 film	GC	—	0.5 M H2SO4	—	ca. 140	46
Defect rich MoS2 nanosheets	GC	285	0.5 M H2SO4	−120	50	77
MoS2 nanosheets	GC	—	0.5 M H2SO4	∼−100	ca. 40	43
Amorphous MoSX films	GC	—	1 M H2SO4	—	40	28
MoS2 NAS	GC	—	0.5 M H2SO4	−54	100	26
Annealed MoS2/Ag	Ag/PET	6	0.5 M H2SO4	—	154	33
MoS2/Ag Strain 0%^b	Ag/PET	6	0.5 M H2SO4	—	145	33
MoS2/Ag Strain 0.005%^b	Ag/PET	6	0.5 M H2SO4	—	141	33
MoS2/Ag strain 0.01%^b	Ag/PET	6	0.5 M H2SO4	—	138	33
MoS2/Ag strain 0.02%^b	Ag/PET	6	0.5 M H2SO4	—	135	33
MoS2/Ag	Ag/PET	2	0.5 M H2SO4	—	152	33
MoS2/Ag strain 0.01%^b	Ag/PET	2	0.5 M H2SO4	—	142	33
MoS2/Ag	Ag/PET	6	0.5 M H2SO4	—	142	33
MoS2/Ag strain 0.01%^b	Ag/PET	6	0.5 M H2SO4	—	138	33
MoS2/Ag	Ag/PET	12	0.5 M H2SO4	—	143	33
MoS2/Ag strain 0.01%^b	Ag/PET	12	0.5 M H2SO4	—	140	33
MoS2 nanoparticles	Graphite	—	—	−100 to −200	—	7
Ni–Mo nanopowder	Ti	1	2 M KOH	−70	—	89
Ni–Mo nanopowder	Ti	3	0.5 M H2SO4	−80	—	89
Ni–Mo nanopowder	Ti	13.4	2 M KOH	−100	—	89
Amorphous MoSx	GC	(10^17 sites cm^−2)	0.5 M H2SO4	−200	60	47
2D MoS2 nanosheets	GC	1.019	0.5 M H2SO4	∼−480	40	This work
2D MoS2 nanosheets	EPPG	1.267	0.5 M H2SO4	∼−450	74	This work
2D MoS2 nanosheets	BDD	1.267	0.5 M H2SO4	∼−450	90.9	This work
2D MoS2 nanosheets	SPE	1.267	0.5 M H2SO4	∼−440	92	This work

---

Researchers have directed attention towards 2D MoS₂ where a single layer comprises two monoatomic planes of hexagonally arranged sulphur atoms linked to molybdenum atoms.^7,38 2D MoS₂ has previously been reported to be electrocatalytic towards the HER.^26,27,34,38 For example, MoS₂ has been reported to exhibit electrocatalytic behaviour, attributed to it having a Density Function Theory (DFT) calculated binding energy towards H⁺ of +0.08 eV at its edge.⁷ Such catalytic activity was however shown to be anisotropic, with the basal plane of the MoS₂ nanosheet being relatively inert and the exposed sulphur edges being the active sites of electron transfer.^{7,13,26,33,39,40} Resultantly, highly defected sheets of MoS₂ have a greater catalytic activity due to the larger number of exposed edges.¹⁴ The MoS₂ edge-terminated sites have high energy kinetics,²⁶ making their production difficult as thermodynamic instability results in the active edges forming fullerene-like structures, which have few exposed edge sites.²⁶ In its ‘bulk’ form, MoS₂ is an inefficient HER catalyst due to it possessing a low ratio of ‘exposed electroactive edges’ to ‘inert basal-like planes’, and thus a high resistance resulting in slow ion transfer.^1,34,41 Interestingly, 2D MoS₂ nanosheets have been reported to possess 13× more active sites compared to the alternative bulk MoS₂.^42,50

Throughout the current literature (see Table 1), glassy carbon (GC) has been used exclusively as a supporting electrode material, with few attempts made to use alternative carbon based supports; this is clearly evident from inspection of Table 1.^{21,34,43–45} For instance, Voiry et al.⁴³ reported a low HER onset potential of ca. 100 mV for the HER using typical MoS₂ nanosheets deposited on a GC electrode. Yu et al.⁴⁶ demonstrated a layer dependent electrocatalysis using MoS₂ (grown via chemical vapour disposition (CVD)) deposited on GC, which is correlated with electron hopping in the vertical direction of the MoS₂ layers. Other work utilising modified GC electrodes has demonstrated that edge exposed MoS₂ nanosheets are efficient HER catalysts and that bulk MoS₂ has low activity.^42,47 This work indicates that an MoS₂ structure with a greater proportion of MoS₂ edge sites to basal planes will likely give rise to improved HER kinetics, with an optimal material having a small geometric basal plane contribution (which is reportedly less active towards the HER).⁴² It has been shown that electrocatalytic activity towards the HER correlates linearly with the number of MoS₂ edge sites.¹⁴

The above studies are elegant in their approaches towards HER; however, they are lacking significantly from not altering the underlying electrode material and in doing so neglecting the ability to de-convolute the true electrochemical performance of 2D MoS₂ nanosheets. It is also important to realise that a cheap electrode support will be required in the application of electrocatalysts utilised in the HER. Graphite screen-printed electrodes (SPEs) meet this criteria due their advantages over other carbon based electrodes, which include scales of economy resulting in ultra-low cost of production, competitive electron transfer performance/properties, versatility, and the ability to tailor and mass-produce such electrodes.⁴⁸ Note that the performance of MoS₂ can only be truly understood via immobilisation using a range of support materials with varying electrode kinetics (electrochemical activities). Such control experiments are explored herein, which are usually overlooked in the current literature. Secondly, it is usual practise within the literature to modify electrodes with only one mass (coverage) of MoS₂, which again makes it difficult to extrapolate a true understanding of 2D MoS₂ nanosheets electrochemical behaviour; this critical parameter is explored in this paper.

Inspired by the above insights and attempts in the literature to utilise MoS₂ as an alternative catalyst to Pt for the HER, this paper explores the use of 2D MoS₂ nanosheet modified carbon based electrodes towards the possible electrocatalysis of the HER. The 2D MoS₂ nanosheets are thoroughly characterised with scanning electron microscopy (SEM), transmission electron Microscopy (TEM), Raman spectroscopy, energy dispersive spectroscopy (EDS), X-ray diffraction (XRD), UV-visual spectroscopy and X-ray photoelectron spectroscopy (XPS). Linear sweep voltammetry is utilised to measure the onset of the HER with four carbon based electrodes as the underlying support materials, namely edge plane pyrolytic graphite (EPPG), glassy carbon (GC), boron-doped diamond (BDD) and screen-printed graphite electrodes (SPEs). This work is distinct from the literature given that we explore a range of electrode substrates, correlate our electrochemical responses with supplementary Raman mapping of the electrode surface (and other complementary physicochemical characterisation) and explore the effect of MoS₂ coverage; each noted component is routinely overlooked in the literature. We also benchmark the electrochemical performance of the 2D MoS₂ nanosheets towards the HER (utilising turn over frequency (ToF) calculations) and compare our results to Pt and literature reports.

2. Experimental section

All chemicals used were of analytical grade and were used as received from Sigma-Aldrich without any further purification. All solutions were prepared with deionised water of resistivity not less than 18.2 MΩ cm and were vigorously degassed prior to electrochemical measurements with high purity, oxygen free nitrogen. The above ensures the removal of any trace of oxygen from test solutions, which if present would convolute the observed results for HER with the competing oxygen evolution reaction (OER); this is common practice in the literature.^49,50 All measurements were performed in 0.5 M H₂SO₄.

Electrochemical measurements were performed using an Ivium Compactstat™ (Netherlands) potentiostat. Measurements were carried out using a typical three electrode system with a Pt wire counter and a saturated calomel electrode (SCE) reference electrode. The working electrodes used were as follows: an edge plane pyrolytic graphite (EPPG) (Le Carbone, Ltd Sussex, UK) electrode, which was machined into a 4.9 mm diameter, with the disc face parallel with the edge plane as required from a slab of highly ordered pyrolytic graphite (HOPG, highest grade available with a lateral grain size, L_a of 1–10 μm and 0.4 ± 0.1° mosaic spread); a glassy carbon (GC) electrode (3 mm diameter, BAS, USA); a boron-doped diamond (BDD) electrode (3 mm diameter, BAS, USA); a Pt electrode (3 mm diameter, BAS, USA); and screen-printed graphite electrodes (SPE), which have a 3 mm diameter working electrode. The SPEs were fabricated in-house with appropriate stencil designs using a DEK 248 screen-printing machine (DEK, Weymouth, U.K.). For their fabrication first, a carbon–graphite ink formulation (product code C2000802P2; Gwent Electronic Materials Ltd, U.K.) was screen-printed onto a polyester (Autostat, 250 μm thickness) flexible film (denoted throughout as standard-SPE); these electrodes have been used extensively in other work.^48,51–53 This layer was cured in a fan oven at 60 °C for 30 minutes. Next, a silver/silver chloride reference electrode was included by screen-printing Ag/AgCl paste (product code C2040308D2; Gwent Electronic Materials Ltd, U.K.) onto the polyester substrates. Finally, a dielectric paste (product code D2070423D5; Gwent Electronic Materials Ltd, U.K.) was then printed onto the polyester substrate to cover the connections. After curing at 60 °C for 30 minutes the SPEs are ready to be used. SPEs have been reported previously and shown to exhibit a heterogeneous electron transfer rate constant, k⁰, of ca. 3.66 ×10⁻³ cm s⁻¹, as measured using the [Ru(NH₃)₆]^3+/2+ redox probe.^48,54–57 Note that for the purpose of this work, electrochemical experiments were performed using the working electrode of the SPEs only and that external reference and counter electrodes were utilised as detailed earlier to allow a direct comparison between all the utilised electrodes.

The 2D MoS₂ nanosheets were commercially procured from ‘Graphene Supermarket’ (Reading, MA, USA).⁵⁸ The 2D MoS₂ nanosheets have a reported purity of >99% and are dispersed in ethanol at a concentration of 18 mg L⁻¹.⁵⁸ The suspended flakes are reported to have an average lateral flake size of 100–400 nm and a thickness of between 1 and 8 monolayers.⁵⁸ The modification of each electrode was carried out using a drop casting approach, where an aliquot of the 2D MoS₂ nanosheet suspension was deposited onto the desired supporting electrode surface using a micropipette.⁵⁴ This deposition was then allowed 5 minutes to dry (at 35 °C) to ensure complete ethanol evaporation. Finally, the electrode was allowed to cool to ambient temperature, after which the process was repeated until the desired mass was deposited onto the surface. The electrode was then ready to use.

An Agilent 8453 UV-visible Spectroscopy System (equipped with a tungsten lamp assembly, G1315A, 8453 for absorption between 250 nm and 1500 nm and a deuterium lamp, 2140-0605 for absorption between 200 nm and 400 nm) was used to obtain the absorption spectroscopy. The absorption spectra was analysed using the UV-Visible ChemStation software. Scanning electron microscope (SEM) images and surface element analysis were obtained using a JEOL JSM-5600LV model SEM equipped with an energy-dispersive X-ray microanalysis (EDS) package. Raman Spectroscopy was next performed using a ‘Renishaw InVia’ spectrometer equipped with a confocal microscope (×50 objective) and an argon laser (514.3 nm excitation). Measurements were performed at a very low laser power level (0.8 mW) to avoid any heating effects. XRD was performed using an “X'pert powder PANalytical model” with a copper source of K_α radiation (of 1.54 Å) and K_β radiation (of 1.39 Å), using a thin sheet of nickel with an absorption edge of 1.49 Å to absorb K_β radiation. The Omega was set to 3.00 and the 2θ range was set between 10 and 100 2θ in correspondence with literature.³² Additionally, to ensure well defined peaks an exposure of 100 seconds per 2θ step was implemented. 2D MoS₂ nanosheets were utilised after deposition onto a sterilised glass slide (coated with excess 2D MoS₂ nanosheets in ethanol then allowed to dry). The XPS data was acquired using a bespoke ultra-high vacuum system fitted with a Specs GmbH Focus 500 monochromated Al Kα X-ray source, Specs GmbH Phoibos 150 mm mean radius hemispherical analyser with 9-channeltron detection, and a Specs GmbH FG20 charge neutralising electron gun.⁵⁹ Survey spectra were acquired over the binding energy range 1100–0 eV using a pass energy of 50 eV and high resolution scans were made over the C 1s and O 1s lines using a pass energy of 20 eV. Under these conditions the full width at half maximum of the Ag 3d_5/2 reference line is ∼0.7 eV. In each case, the analysis was an area-average over a region approximately 1.4 mm in diameter on the sample surface, using the 7 mm diameter aperture and lens magnification of ×5. The energy scale of the instrument is calibrated according to ISO 15472, and the intensity scale is calibrated using an in-house method traceable to the UK National Physical Laboratory.⁶⁰ Data were quantified using Scofield cross sections corrected for the energy dependencies of the electron attenuation lengths and the instrument transmission.⁶¹ Data interpretation was carried out using CasaXPS software v2.3.16.⁶²

3. Results and discussion

3.1 Characterisation of 2D MoS₂ nanosheets

3.1.1 UV-VIS and lateral grain size calculations. The lateral length (L_a) and number of the commercially procured 2D MoS₂ nanosheets can be readily deduced from absorption spectroscopy. It has been observed that actually the terminology is more correctly optical extinction spectroscopy since the optical beam interacts with the dispersed nanosheets by both absorption and scattering.^63–65 The extinction coefficient of dispersed 2D MoS₂ nanosheets is 6820 L g⁻¹ m⁻¹ at the local minimum of 345 nm, using this information along with a spectra it is possible to determine the concentration of dispersed 2D MoS₂ nanosheets.⁶⁴ Varrla et al.⁶³ uses this information to calculate the concentration as a function of mixing parameters whilst also showing the extinction spectra can be used to determine information regarding the 2D MoS₂ nanosheet length and thickness. ESI Fig. 1† shows the optical extinction spectra of the commercially procured MoS₂ nanosheets which are dispersed in ethanol. It is readily evident that the spectra displays A- and B-excitonic transitions as well as other pertinent features consistent with the 2H polytype of MoS₂ and is consistent with MoS₂ nanosheets produced via shear exfoliation.^63,64,66 The extinction spectrum of the 2D MoS₂ nanosheets allows one to readily determine the mean nanosheet lateral length due to the effect 2D MoS₂ nanosheet edges have upon the spectral profile. The extinction spectrum also allows the number of layers (thickness) to be determined as a result of quantum confinement effects causing a well-defined shift A-exciton position corresponding to nanosheet thickness.⁶⁴ The lateral length, L (μm) of the MoS₂ can be deduced from the following equation:where (Ext_B/Ext₃₄₅) is the ratio of extinction at the B-excition to that at 345 nm since the spectral profile is dependent upon the lateral length of the MoS₂. Further information can be obtained in terms of the number of nanosheets, N_MoS2 expressed as the number of monolayers per nanosheet can be determined from the wavelength associated with the A-excition, since the quantum conferment effects result in well-defined shifts in the A-excition position with the thickness of the nanosheet; this is summarised by the following equation:

From the spectra presented in ESI Fig. 1† the lateral length and number of MoS₂ nanosheets are determined to correspond to 61.5 nm and 3 (2.89) respectively. We note that the lateral size is smaller than the value given by the commercial supplier. The average of 3 monosheets per nanosheets agrees strongly with the commercial supplier who notes the number of monolayers per nanosheet to be between 1–8 in solution.⁵⁸ It is important to point out that the determined lateral size and number of MoS₂ sheets are for when these are in solution; when immobilised upon a surface these will deviate from the measured values, as we will see later, but this is a common issue throughout the literature.

3.1.2 SEM, TEM, Raman, EDS and XRD. Independent physicochemical characterisation of the 2D MoS₂ nanosheets was performed. SEM and TEM images of the commercially sourced 2D MoS₂ nanosheets are shown in ESI Fig. 2 and 3.† Despite some aggregation, which is the case for all nanosheet materials, upon close inspection a lateral size of ca. 100–400 nm is evident. The 2D MoS₂ nanosheets immobilised upon the silicon wafer generally exhibit a uniform coverage. Complimentarily EDS mapping analysis was performed to offer insight into the elemental composition of the area shown in ESI Fig. 2.† Analysis of the EDS map shows uniform distribution of Mo and S atoms with a ratio of 0.55% At. and 1.35% At. and this composition correlates with expected values for the structure of 2D MoS₂ nanosheets (ca. 1 : 2 ratio of Mo and S respectively) agreeing with independent literature.³⁵ Additionally, XRD analysis showed characteristic diffraction peaks for the MoS₂ nanosheets, with 2θ corresponding to 14.2° and indicating the presence of MoS₂via the reflection of separated MoS₂ layers (see ESI Fig. 4†) which is in agreement with literature reports.^32,41 The peak at 28° 2θ is attributed to the supporting glass slide. Last, the Raman spectra for the 2D MoS₂ nanosheets after deposition on a serialised silicon disk are presented in Fig. 1. The characteristic peaks associated with MoS₂ at ca. 380 and 405 cm⁻¹ are clearly evident, corresponding to that reported in the literature.⁶⁷ Further analysis is described later where Raman mapping is utilised to explore MoS₂ mass modifications deposited onto an electrode surface.

Fig. 1 Raman spectra of the commercially sourced 2D MoS₂ nanosheets immobilised upon a silicon wafer.

3.1.3 XPS. MoS₂ nanosheets suspended in ethanol were prepared for X-ray photoelectron spectroscopy (XPS) analysis by pipetting a few drops of the suspension onto a fragment of a clean Si (111) wafer and allowing the ethanol to evaporate. The XPS spectrum is shown in ESI Fig. 5† and the results of the surface composition analysis (excluding hydrogen) are shown in ESI Table 1.† The C and O present is likely a result of residuals from the ethanol used to disperse the MoS₂. This is supported by the presence of a component peak at ca. 286.6 eV showing that alcohol groups were present on the sample surface. The Si present can be accounted for by the use of a Si (111) wafer as an underlying support material for the drop coating. Na is likely present from an organic Na contaminant or via contributions from the supporting wafer. The Mo to S % atomic concentrations are observed at a 1.0 : 2.2 ratio respectively, agreeing well with the Raman and EDS analysis performed above (further indicating the presence of the target material). Shin et al.⁶⁸ theorises the deviation from an expected stoichiometry ratio of 1 : 2 for Mo and S respectively in 2D MoS₂ nanosheets is due to the presence of MoS₃.⁶³ It is of note that Mo is present in three valence states, with each of the valence states consisting of a Mo 3d_5/2 and Mo 3d_3/2 doublet. The curve-fitted high resolution Mo 3d spectral region is shown in ESI Fig. 6.† The area ratios of the doublets are constrained to the ratios of the Scofield cross sections (i.e. Mo 3d_3/2 = 0.6904 × Mo 3d_5/2), the two components of each doublet are constrained to the same line shape and ‘full width at half maximum’, and their separations are fixed at the known reference value of 3.13 eV. The 3d_5/2 components were found at 229.4 eV (Mo⁴⁺), 231.6 eV (Mo⁵⁺) and 233.1 eV (Mo⁶⁺). The spectral region also includes the S 2s peak at 226.6 eV.

3.2 Electrochemical activity at assigned coverage

The electrochemical response of the 2D MoS₂ nanosheets (1266.7 ng cm⁻²) immobilised upon SPEs were studied using cyclic voltammetry (CV). Fig. 2 shows a typically observed CV where two oxidation peaks at +0.65 and +1.0 V are clearly visible in the first cycle followed by several minor reduction peaks in the −0.5 to −1.5 V range. The initiation of the HER is evident in the cathodic response, with an onset potential of ca. −1.1 V. In terms of the anodic response, the two prominent oxidation peaks are due to irreversible reactions, likely the oxidation of Mo⁴⁺ to Mo⁶⁺ at edge- and basal-plane sites. Such oxidations are not observable in subsequent scans, which is likely due to the MoO₃ dissolving into solution or being reduced to Mo³⁺, at which point it dissolves, explaining the origin of the observed reduction peak. This is represented in the equation below.⁶⁹

MoS₂ + 7H₂O → MoO₃ + SO₄²⁻ + ½S₂²⁻ + 14H⁺ + 11e⁻ (3)

Fig. 2 Typical cyclic voltammetric response of 2D MoS₂ nanosheets (1266.7 ng cm⁻²) immobilised upon a SPE in pH 7 phosphate buffer solution (PBS). First scan: solid black line. Second scan (representative of subsequent scans): red dotted line. Scan rate: 5 mV s⁻¹ (vs. SCE).

It may also be possible that the Mo within our 2D MoS₂ nanosheets has two dominant valence states, both of which are oxidised, resulting in the presence the double peak. This inference is supported by the independent XPS analysis (see earlier), which indicated the presence of Mo in three different valance states for the 2D MoS₂ nanosheet sample analysed. The double oxidation peak observed in Fig. 2 is of interest and further study is required to determine its exact cause. The observed voltammetry (see Fig. 2) is in good agreement with Bonde et al.⁶⁹ who explored nano-MoS₂ deposited onto toray carbon paper in 0.5 M H₂SO₄ towards the HER. If the potential window is kept below the range where the reported oxidation peaks occur, the HER activity of nano-MoS₂ remains stable.⁷⁰ The reduction peaks can also be accounted for by the reduction of S₂²⁻ to H₂S.⁶⁹ The double oxidation peak has previously been mischaracterised as a single oxidation peak due to the peaks merging: occurring as a result of performing cyclic voltammetry at too high a scan rate.⁷¹ This is important to note as the double peak convolutes the understanding of the electrochemical process occurring.

Next, attention was turned to benchmarking our electrochemical system for the HER using commonly available electrodes in 0.5 M H₂SO₄, as is widespread practise within the literature.^14,69,72 As is evident in Fig. 3(A), the four unmodified carbon based electrodes are all significantly inferior to a Pt electrode with respect to the potential required for HER onset and the current density reached. Note that the onset of HER is analysed as the potential at which the observed current initially begins to deviate from background current. Pt′s superior performance is to be expected given it is a pure metal which has a very small binding energy for H⁺.⁷ It is evident that the onset potential for the HER occurs at ca. −1.05, −0.78, −0.76 −0.73, and −0.25 V at the GC, EPPG, BDD, SPE, and Pt electrodes respectively. The SPEs exhibit the lowest onset potential for the HER when compared to all of the carbon-based electrodes utilised herein. The bare GC electrode exhibited the largest HER onset potential, indicating that it is not a beneficial electrode for the HER.

Fig. 3 (A) Linear sweep voltammetry (LSV) of unmodified EPPG, GC, SPE, BDD and Pt electrodes showing the onset of the HER. In all cases, scan rate: 25 mV s⁻¹ (vs. SCE). Solution composition: 0.5 M H₂SO₄. (B) Tafel analysis; potential vs. natural log (ln) of current density for faradaic section of the LSV presented in (A).

It is interesting to consider the current density obtained at each of the unmodified electrodes towards the HER and how this progressively alters during the course of the measurement. Although it is apparent that a higher potential is required to initiate the HER at the GC electrode (by ca. −0.3 V in contrast to the alternative carbon based electrodes, making it unfavourable in this sense), the current density recorded at this electrode appears to surpass that of the alternative materials.

3.2.1 Tafel analysis. A common approach in the literature is to employ Tafel analysis, allowing the most likely electrochemical process to be theorised. Literature has suggested three possible steps in the reaction, each of which is capable of being the rate-determining step of the HER; this analysis is dependent on the corresponding Tafel slope. The initial H⁺ discharge step being the Volmer reaction, leading to the following equation:^11–13

H₃O⁺ (aq) + e⁻ + catalyst → H (ads) + H₂O (l) (4)

The Volmer step can then be followed by one of two possible steps; either the Heyrovsky step:

H (ads) + H₃O⁺ (aq) + e⁻ → H₂ (g) + H₂O (l) (5)

or the Tafel step:

H (ads) + H (ads) → H₂ (g) (6)

where the transfer coefficient (α) is 0.5, F is the Faraday constant, R is the universal gas constant and T is the temperature at which the electrochemical experiment was performed. Values from the Tafel analysis (presented below each equation) are an indication of the reaction mechanism. Tafel analysis was performed on the Faradaic sections of the LSV plots which can be observed in Fig. 3(A). Tafel analysis was performed on the data presented in Fig. 3(A) where the corresponding analysis is presented in Fig. 3(B), which yielded Tafel values of ca. 89.2, 64.4, 94.4 and 81.2 mV dec⁻¹ for the unmodified EPPG, GC, SPE and BDD electrodes respectively. Using the above values for the unmodified electrodes, interpretation of the Tafel slopes reveals that the adsorption Volmer step is likely rate limiting for the SPE and EPPG electrodes. The discharge Heyrovsky step is most likely the rate limiting step at the GC electrode.^11–13 In the case of BDD, Tafel analysis does not allow a definitive mechanism to be estimated.

We next explore the use of 2D MoS₂ nanosheet modified carbon based electrodes towards the HER. As detailed in the introduction, the aim of this paper is to tackle the current issue of finding a cheap, more abundant, electrocatalyst alternative to Pt for the HER.^7,11 We investigate the potential current state of the art (2D MoS₂ nanosheets) and compare this to our benchmarking experiments with the aim of revealing valuable insights. First, the range of carbon based electrodes utilised above (namely GC, BDD, EPPG and SPEs) are modified with different masses of the 2D MoS₂ nanosheets. Note that using various underlying support materials is seldom seen in the current literature and thus it is important that such control experiments are explored and reported herein for the first time. Inspection of Fig. 4(A) reveals that using a 1267 ng cm⁻² modification of 2D MoS₂ nanosheets results in a lowering of the potential required for onset of the HER and is accompanied by an increase in the observed current density, signifying an improved electrochemical response at each of the underlying electrode substrates utilised. Specifically, the HER onset potential was lowered to ca. −0.45, −0.48, −0.44 and −0.45 V for EPPG, GC, SPE, and BDD electrodes respectively. Clearly, these newly obtained values are significantly closer to the reported value using a Pt electrode (ca. −0.25 V) than the initial values reported above at the unmodified carbon based electrodes. This implies that the 2D MoS₂ nanosheet is an effective electrocatalyst for the HER.

Fig. 4 (A) LSV of 1267 ng cm⁻² 2D MoS₂ nanosheet modified EPPG, GC, SPE, BDD electrodes and an unmodified Pt electrode showing the onset of the HER. In all cases, scan rate: 25 mV s⁻¹ (vs. SCE). Solution composition: 0.5 M H₂SO₄. (B) Tafel analysis; potential vs. natural log (ln) of current density for Faradaic section of the LSV presented in (A).

Tafel analysis was performed on the modified LSV profiles. Tafel slope values of ca. 74.7, 41.4, 90.0, and 90.9 mV dec⁻¹ were estimated at the EPPG, GC, SPE and BDD electrodes respectively (shown in Fig. 4(B)). Comparison of the Tafel values suggests that modification of the support electrodes with 2D MoS₂ nanosheets does not cause a significant alteration in the mechanism or indeed the Faradic current density of the HER.⁴² This however, is not the case when utilising the GC electrode, which exhibited a reasonable increase in current density resulting in the ‘discharge Heyrovsky’ step more likely becoming the rate limiting step. This implies that modification with the 2D MoS₂ nanosheets allows for increased and sufficient H⁺ adsorption, thus in turn catalysing the HER process.

In terms of analysing the current densities obtained, deposition of the 2D MoS₂ nanosheets at a coverage of 1266.7 ng cm⁻² induces higher exchange current densities when compared to the unmodified alternative. It is likely that this results from the early onset of the HER (i.e. the decreased over-potential) and is not directly due to an overall increased reaction rate at the modified SPE and BDD electrodes given that comparison is based on analysis at specific potentials. For the GC and EPPG electrodes there was an increase in the current density, which is due not only the earlier HER onset potential but also an increased current density slope. This is likely a result of the structure of MoS₂ on these electrodes having a high number of exposed active edge sites allowing for a greater amount of H⁺ adsorption.

3.3 Electrochemical activity at differing 2D MoS₂ nanosheet coverages

3.3.1 Raman analysis. The experimental data discussed above considers the use of 1267 ng cm⁻² (2D MoS₂ nanosheet) modified electrodes. Currently it is common practice in the literature to choose only one mass of MoS₂ coverage when exploring its electrochemical performance (in addition to exploring this on only one underlying electrode surface, typically GC). As highlighted in the introduction, this practice results in convoluted interpretations of the performance of 2D MoS₂ nanosheets and thus diligent controls are required in order to reveal the true electrochemical response. We next consider this parameter and explore the effects of different mass coverages of 2D MoS₂ nanosheets on a given electrode.

In order to ascertain the level of MoS₂ coverage and relate this to the observed voltammetry, we first investigate the respective Raman properties of our modified electrodes. Fig. 5 depicts the effect that larger deposition quantities of MoS₂ onto a SPE has upon the recorded Raman Spectroscopy, where the evolution of the two characteristic peaks is evident and is discussed in more detail below. We first analyse the effect of coverage on the electrode surface using Raman mapping and a SPE as the underlying support material (as a representative model). Through comparison of the underlying graphite peak area at ca. 1550 cm⁻¹ against the area of the MoS₂ Raman peaks at ca. 380 and 405 cm⁻¹, the surface area coverage of the electrode was investigated. It is evident that increasing the mass deposition of MoS₂ on the SPE surface results in an increased intensity in the respective assigned Raman peaks (see Fig. 5(A)). Raman maps are presented in Fig. 6 which concur with the previous inference and show that with increased mass deposition a thicker (largely uniform) layered coverage is achieved across the entire electrode surface. Through analysis of the respective Raman maps depicted in Fig. 6 and 7, it is likely that the deposition of 504 ng cm⁻² 2D MoS₂ nanosheets results in the complete coverage of the underlying SPE support material (which has a surface area of 0.0707 cm²). With each additional modification this layer of deposited MoS₂ will thicken (see Fig. 7). It is possible to determine the stacking number by comparison of E¹_2g and A_1g vibrational bands (VB) as the observed Raman spectrum evolves with the number of layers present. The E¹_2g VB results (at ca. 382 cm⁻²) due to the opposite vibration of two S atoms in respect to a Mo atom, whereas the A_1g peak (at ca. 407 cm⁻²) represents the S atoms vibrating in opposite directions and out of plane.^67,73 Literature suggests that as MoS₂ moves from single layer to bulk the E¹_2g VB downshifts from 384 to 382 cm⁻², whilst A_1g VB shifts upwards from 403 to 408 cm⁻¹, where a separation of ca. 19 cm⁻¹ between the VBs is indicative of single layer MoS₂ and a value of ca. 25 cm⁻¹ represents the bulk material.^26,40,73 In this work we observe an increase in the Raman shift of the E¹_2g VB with greater mass additions from ca. 377 to 380 cm⁻¹,(see Fig. 5(A)), however the response observed herein may result from the specific morphology and stacking structures of our MoS₂ on the underlying support material given that previous studies have reported a similar shift relating to the E¹_2g band and attributed this to uniaxial strain or heterostructure stacking.⁷⁴ Interestingly, analysis of the A_1g peak corresponds to the predicted theorem for a transition from single-layer to multi-layer 2D MoS₂ nanosheets with an increasing Raman shift from ca. 402 to 405 cm⁻¹.^40,67

Fig. 5 (A) Raman spectra peak intensity and position for 504 (black), 1009 (red), 2019 (blue) and 2533 (green) ng cm⁻² 2D MoS₂ nanosheet modifications on SPEs. (B) Depicts 2D MoS₂ nanosheet coverage plotted against Raman peak intensity for E¹_2g (black) and A_1g (red) vibrational bands, showing a constant peak distance of 24.7 cm⁻¹ at all coverages.

Fig. 6 Raman maps representing various MoS₂ coverages onto an underlying SPE. Coverages of 2D MoS₂ nanosheets: 0 (A), 504 (B), 1009 (C), 2019 (D), and 2533 (E) ng cm⁻². Raman intensities recorded at 405 cm⁻¹.

Fig. 7 Raman maps of SPE surface, each point showing the intensity ratio between the sum of the characteristic MoS₂ peak areas (380 and 405 cm⁻¹) against the area of the underlying graphite peak (1550 cm⁻¹). Using varying surface coverages of our 2D MoS₂ nanosheets; 504 (A), 1009 (B), 2019 (C) and 2533 (D) ng cm⁻². The grey maps are the modified electrodes and the black map in each represents an unmodified electrode surface.

Furthermore, when considering the separation of the two VBs (given that this value is also indicative of the number of MoS₂ layers present) the shift in the difference of the VB positions between A_1g and E¹_2g gives a consistent value of 24.7 cm⁻¹ with increasing coverages, indicating that +4 layers (i.e. bulk) MoS₂ is present at all four mass coverages utilised herein (see Fig. 5(B)). Consideration of the above factors in conjunction with our independent lateral grain size calculations (UV-VIS) indicates that the structural model of the 2D MoS₂ nanosheets utilised herein is likely that of re-assembly, with the few-layer nanosheets forming bulk MoS₂ upon deposition onto the electrode surface. Further work on whether the morphology of a SPEs surface causes uniaxial strain and/or heterostructure stacking of the 2D MoS₂ nanosheets would be of great interest. Note that confirmatory tests were performed in which the aliquots deposited above were immobilised onto an alternative silicon wafer support in order to overcome potential issues with the underlying surface. Such control experiments (data not shown) exhibited the same separation values as identified above (ca. 25.1 cm⁻¹) between the two VBs and thus confirms that our 2D MoS₂ nanosheets likely form bulk MoS₂ once deposited/immobilised onto a support surface. It is likely that further additions of our material result in the formation of thicker or rougher ‘bulk’ layers. We next explore the electrochemical implications of this, often overlooked in the literature.

3.3.2 Electrochemical HER: critical mass of 2D MoS₂ nanosheets. Returning to the analysis of the HER, the effects of different mass deposition on the different electrode surfaces was next explored. Using a fixed potential (–0.75 V) at which each electrode exhibits an observable current, the current densities relating to the HER were recorded. It is evident through inspection of Fig. 8 that a trend of increased current density (corresponding to increased 2D MoS₂ nanosheet coverage (ng cm⁻²)) is subsequently followed by a decrease in current density and/or plateauing effect. This is apparent upon modification of each of the four underlying electrode materials studied with our material of interest. Over the modification range tested, the maximum/optimal current density was found to correlate to a 2D MoS₂ nanosheet coverage of ca. 1014.4, 1266.7, 1266.7, 1266.7 ng cm⁻² for the GC, EPPG, SPE and BDD electrodes respectively. Tafel analysis on these optimal mass modifications reveals values of ca. 40, 74, 92 and 90.9 mV dec⁻¹. GC and BDD when modified with the 2D MoS₂ nanosheets demonstrate a fast discharge mechanism, with H⁺ adsorption no longer the rate-limiting step.¹ Through analysis of the values reported above and comparison to the values reported in section 3.2.1 it is clear that of the four electrodes modified, electrodes with slower kinetics (such as the GC and BDD) exhibit a favourable electrocatalytic effect when modified with 2D MoS₂ nanosheets towards the HER. Contrary to this, the EPPG, which possesses faster underlying rate kinetics, exhibited only a slight change towards an altered reaction mechanism and further in the case of the SPE there was no observable change.

Fig. 8 Current density values taken at −0.75 V from linear sweep voltammetry from 0, 128.6, 252, 504, 762. 1009, 1267, 1524 and 1771 ng cm⁻² 2D MoS₂ nanosheets immobilised upon: (A) EPPG, (B) GC, (C) SPE (D) BDD. Each graph showing the plateauing effect of current density when a critical mass of the 2D MoS₂ nanosheets is deposited onto the electrodes surface. Error bars are the average and standard deviation of 3 replicates.

3.4 Assessment of 2D MoS₂ nanosheets catalytic activity

An array of approaches are employed to study the intrinsic catalytic activity of 2D MoS₂ nanosheets within the literature. In order to deduce how the observable catalytic activity alters with changes in the mass of 2D MoS₂ nanosheet modification we employ two commonly used techniques below: the assessment of catalytic turn over frequency (ToF) and the number of active sites present on the surface of the electrode.^47,68

3.4.1 Turn over frequency (ToF). In order to evaluate how the intrinsic catalytic activity of the 2D MoS₂ nanosheets alters with varying modification on a ‘per active site’ basis, the ToF was deduced using a method reported previously by Benck et al.⁴⁷ their deriviation is repeated here for clarity using values associated with the 252 ng cm⁻² modified SPE.^47,75 In this calculation, it is assumed that the surface of the 2D MoS₂ nanosheet is atomically flat (although the true modification will have a finite roughness).⁴⁷ Taking the sulphur to sulphur bond distance to be 3.15 Å (corresponding to an area of 4.296 Ų per S atom)^47,76 which can be used to calculate the surface area occupied by each MoS₂:

Using the derived area for a MoS₂ molecule (corresponding to the number of surface sites for a flat standard) it is possible to determine the number of MoS₂ molecules per cm² geometric area:

The number of electrochemically accessible surface sites can be determined from the following:

It is also essential to accurately determine the roughness factor (R_F) for each modified electrode surface which was performed using white light profilometry (WLP) (See ESI Fig. 7†) as is common within the literature. The described WLP technique yielded R_F values which represent the entire surface area of the electrode rather than electroactive area (see later). In the case of the 252 ng cm⁻² modified SPE, the number of surface sites cm⁻² corresponds to 3.16 × 10¹⁶ surface sites per cm². The following allows the ToF on a per-site basis to be determined:

Taking the value of current density (mA cm⁻²) at the potential of −0.75 V (at a 100 mV s⁻¹ scan rate) and using the R_F calculated via WLP, per-site the ToF can be deduced from the following:

Using eqn (12) and a value derived from formula (11), it is possible to determine a value for the ToF:

At the chosen potential (−0.75 V) the current densities were found to correspond to 0.83, 1.16 and 1.17 mA cm⁻² for SPEs (SPEs used as representative example of the carbon based electrodes utilised) modified with 252, 1009 and 2019 ng cm⁻² of 2D MoS₂ nanosheets. The R_F values calculated via the WLP method (which provides the topography of the uppermost surface, see roughness factor calculation method 1 in the ESI†) were found to correspond to 1.918, 1.934 and 1.924 respectively. Using these values the ToF values deduced from the above equations were found to correspond to 0.81, 1.14 and 1.15 . The stability of the WLP R_F values across the range of modifications results in relatively little variation of the ToF values obtained. Benck et al.⁴⁷ suggest that the upper and lower possible ToF could be one order of magnitude greater or less than the given value. If the chosen potential is altered to −1.5 V the current densities at this potential correspond to 4.25, 4.43 and 4.33 mA cm⁻² for the 252, 1009 and 2019 ng cm⁻² 2D MoS₂ nanosheet modified SPEs respectively. Using the same technique and WLP R_F values above, however replacing the current densities at −0.75 V for those at −1.5 V the ToF values are 0.59, 0.61 and 0.6 for the 252, 1009 and 2019 ng cm⁻² 2D MoS₂ nanosheets modified SPEs respectively. Clearly, altering the potential at which the current densities are recorded has a significant affect upon the ToF recorded.

In order to deduce R_F values which are representative of the true electroactive area of an electrode, a double layer capacitance technique can be employed (see roughness factor calculation method 2 in the ESI†). The double layer capacitance technique is preferential to the WLP technique as it describes the true electroactive surface area of the electrode, including accessible pores and the thickness of multiple layers deposited, whereas WLP is limited in that it bases its calculated value off only a scan of the topography of the uppermost surface in question; simply, the WLP only probes the outermost layer and cannot determine a relatively thick layer from a thin layer. The double layer capacitance technique for determining R_F is reported in the ESI.†

Taking the current density values corresponding to a −0.75 V potential and replacing the WLP R_F values for 1.0, 3.3 and 4.8 which are the R_F values calculated via double layer capacitance, for SPEs modified with 252, 1009 and 2019 ng cm⁻² of 2D MoS₂ nanosheets, the ToF was found to correspond to 1.58, 0.6 and 0.46 . The values estimated for the ToF herein are in rough agreement with the range of ToF values Xie et al.⁷⁷ states should be expected for various types of MoS₂ structures. It is evident that, as a result of increasing the mass of 2D MoS₂ nanosheets deposited onto an SPE, there is a resulting decrease in the ToF . This is possibly the result of larger masses of 2D MoS₂ nanosheets, once deposited, forming a MoS₂ structure whereby there is increased shielding of active edge sites by inactive basal planes.^42,47 Again, if the current densities are altered to those corresponding to a −1.5 V potential whilst using the same technique and double layer capacitance R_F values, above, the ToF values deduced are 1.14, 0.37 and 0.24 for the 252, 1009 and 2019 ng cm⁻² 2D MoS₂ nanosheets modified SPEs respectively. Of note is the significant affect that altering the potential at which the current densities are measured has upon the determined ToF values. This ability to manipulate the ToF value deduced makes it essential for literature to create an industry standard for what potentials should be used when dealing with ToF calculations. This would allow for more ready comparison between different studies involving ToF.

As previously stated the double layer capacitance technique provides a true electrocatalyic surface area of the given electrode whereas the WLP technique provides a scan of topography of the uppermost surface. It can therefore be assumed that the R_F values deduced via double layer capacitance offer more accurate ToF values.

3.4.2 Number of active sites. Determining the number of active sites present on the surface of an electrode offers valuable insight into its catalytic properties. Shin and co-workers⁶⁸ have shown that it is possible to derive the number of 2D MoS₂ nanosheet active sites (N) present on the surface of the catalyst using the following equation:

N = R_F(N_Ad/M_f)^2/3 (13)

where N_A is Avogadros number, d is the film density (ca. 2.35 × 10⁻⁴ g cm⁻³, which was derived via the use of WLP to observe the step height, in this case 21.9 μm, between the bare electrode surface and a mass of 514 ng cm⁻² 2D MoS2 nanosheets deposited), M_f is the formula weight of MoS₂, and R_F is the roughness factor, which in this case is defined as the ratio of the real surface area to the geometric area. The geometric area (and the electrochemically active area) of the electrode surface can vary significantly due to the surface roughness and porosity of the sample. In this case the R_F was derived using double layer capacitance as it is linearly proportional to catalytic surface area (see roughness factor calculation method 2 in the ESI†).⁷⁸ The double layer capacitance values determined (via cyclic voltammetry, see ESI Fig. 8†) are 8.7, 68, 218, 322 μF cm⁻² for SPEs modified with 0, 252, 1009 and 2019 ng cm⁻² 2D MoS₂ nanosheets respectively. As a benchmark the double layer capacitance value for amorphous MoS₂ is reported to be 66.7 μF cm⁻².⁴⁶R_F values for SPEs modified with 252, 1009 and 2019 ng cm⁻² of 2D MoS₂ nanosheets were estimated to be ca 1.02, 3.26 and 4.83 respectively. The R_F values derived using double layer capacitance show an increment with greater masses of 2D MoS₂ nanosheet modification. R_F values for BDD, EPPG, GC and SPE post modification are presented in Table 2. The number of active sites per cm⁻² are summarised in Table 2 where there is a (positive) linear correlation between number of active sites and mass of MoS₂ deposited. It can therefore be asserted that there is a physicochemical change within the structure of the 2D MoS₂ present upon the surface of the electrode, which leads to the exposure of less active sites per additional ng cm⁻² modification after a certain ‘critical mass’ (this is also evident through inspection of Fig. 8), or no further increase in the number of the active sites accessible to the solution. It is clear that the number of active sites (and related HER performance) increase (improve) with the addition of larger quantities of MoS₂ onto the underlying electrode surface up until the specified critical mass is achieved.

Table 2 The determined roughness factor (R_F) values and the number of active sites (per cm²) for BDD, EPPG, GC and SPE all of which had been modified with 252, 1009, 2019 ng cm⁻² 2D MoS₂ nanosheets. Values determined using double layer capacitance obtained via cyclic voltammetry between the potential range of 0.01 V and 0.1 V

Electrode	MoS2 modification (ng cm^−2)	Roughness factor	Number of active sites
BDD	2019	2.2	2.01 × 10^12
	1009	1.5	1.38 × 10^11
	252	0.3	3.03 × 10^11
EPPG	2019	7.7	7.05 × 10^12
	1009	5.5	5.05 × 10^12
	252	4.2	3.84 × 10^12
GC	2019	2.6	2.35 × 10^12
	1009	2.1	1.95 × 10^12
	252	1.0	8.95 × 10^11
SPE	2019	4.8	4.43 × 10^12
	1009	3.3	3.00 × 10^12
	252	1.0	9.36 × 10^11

---

It is evident that the increased mass deposition of 2D MoS₂ nanosheets on a given electrode surface results in an improvement in the current passed in addition to a lowered HER onset potential (improved electrochemical response). As is apparent from the above discussion (and inspection of Fig. 8) this increase in the catalytic performance at a given modified electrode material (which corresponds to the addition of 2D MoS₂ nanosheets) reaches a plateau after which further additions of our target material do not result in an improved electrochemical performance. This ‘critical mass’ of modification is likely due to the structure of the 2D MoS₂ nanosheet altering to that of a bulk formation (see earlier), thereby exposing less edge sites and thus inhibiting the beneficial electrochemical properties of single-, few-, quasi- MoS₂ nanosheets. Alternatively, this plateau could signify the mass (a critical mass of ca. 1009 ng cm⁻² of the 2D MoS₂ nanosheets) at which the structure of MoS₂ can no longer structurally support itself upon the electrode surface (becoming unstable), diminishing and detaching from the surface of the electrode, eliminating the catalytic advantages of further additions (such that in some cases the catalytic response begins to deteriorate). Similar observations have been reported for the case of graphene.^79–82 This could arise due to the disconnection of 2D MoS₂ nanosheet layers during the course of the experiment (i.e. instability of the modified layer on the electrode surface due to the large quantity/mass present) which is bought about once such a ‘critical mass’ is achieved.

The intra-repeatability of the modified and unmodified SPEs were tested (N = 3). The % Relative Standard Deviation (% RSD) in the onset potential of the HER was found to be ca. 0.8, 1.4, 1.5, and 4.6% at the 0 (unmodified), 252, 1009 and 2019 ng cm⁻² 2D MoS₂ nanosheet modified SPEs respectively. It is clear that the % RSD increases with greater modifications of the MoS₂, likely a result of the factors stated above and potentially resulting in the reduced catalytic effects. Furthermore, the % RSD in the current densities observed were found to be ca. 12.4, 13.5, 10.5, and 15.5% at the 0 (unmodified), 252, 1009 and 1771 ng cm⁻² 2D MoS₂ nanosheet modified SPEs respectively. The high RSD values in this case are indicative of the structural instability of the deposited 2D MoS₂ nanosheet film on the underlying electrode surface, which likely leads to delamination and thus high levels of variation within the effective surface area and currents passed.

From the above inferences, we suggest that the structural model of 2D MoS₂ nanosheets is likely that of re-assembly, such that upon modification with increasing amounts of 2D MoS₂ nanosheets, bulk layers of MoS₂ materialise upon the surface of our support electrodes.^79,81

Electrochemical impedance spectroscopy (EIS) was used to determine the impedance of the electrode system as the coverage of 2D MoS₂ nanosheets was increased, see ESI Fig. 10.† ⁷⁰ It was observed that the charge transfer resistance (Ω) for all electrodes decreased after modification with 252 ng cm⁻² 2D MoS₂ nanosheets and further decreased after modification with 1009.5 ng cm⁻². Again a plateauing was observed in the response with increased coverage (for example after modification with 2019 ng cm⁻²). The Ω values of the unmodified SPE (3.51 × 10⁵ Ω) reduced to 1.69 × 10⁵ Ω after 252 ng cm⁻² and then to 3.27 × 10³ Ω upon a 1009 ng cm⁻² 2D MoS₂ nanosheet modification, after which the response plateaued, having an impedance value of 3.21 × 10³ Ω by modification with 2019 ng cm⁻². Error values for the aforementioned results were recorded as 1.78 × 10⁻⁴, 1.73 × 10⁻³, 1.2 × 10⁻², and 1.21 × 10⁻² Ω respectively. This EIS supports the above inferences and indicates that the 2D MoS₂ is an effective electrocatalyst with respect to the HER when deposited on a carbon electrode surface.

Finally, it is essential to assess the electrochemical stability of 2D MoS₂ nanosheets as a catalyst for the HER when drop coated onto an electrode surface (following on from the reported % RSD values noted above). This is a practical consideration for real applications where durability and longevity are necessary.^2,27,68 SPEs were used as a representative example for the four carbon based electrodes used within this study. A 1000 cycle voltammetry scan from 0 to −0.8 V at 25 mV s⁻¹ was performed on SPEs modified with 252, 1009, 2019 ng cm⁻² of 2D MoS₂ nanosheets. A decrease in the catalytic activity of each electrode was observed (see Fig. 9). At 0.6 V the decrease from the initial scan current was 6.8%, 32.3%, 26.9% for the SPEs modified with 252, 1009 and 2019 ng cm⁻² respectively. The observed activity loss is significant, especially for the 1009 and 2019 ng cm⁻² modified electrodes. According to Shin et al.⁶⁸ high stability of amorphous MoS₂ is rarely reported with the decrease in activity being associated with either: surface absorbatives which may poison the active sites of the 2D MoS2 nanosheets;⁴⁷ or the delamination of the 2D MoS2 nanosheets from the substrate.⁴⁷ However, it is likely a combination of these two effects with the delamination of 2D MoS₂ nanosheets becoming more prevalent at higher modifications. This inference is supported by the EIS observations detailed earlier.

Fig. 9 Stability studies using SPEs modified with (A) 252, (B) 1009 and (C) 2019 ng cm⁻² 2D MoS₂ nanosheets. Cyclic Voltammetry was performed between the potential range of 0 to −0.8 V, repeated for 1000 cycles, these figures show the initial (black line) and 1000^th (red line) scans. Scan rate: 25 mVs₋₁ (vs. SCE).

Another consideration in terms of industrial application is cost of the four carbon based electrodes studied, the GC offered the greatest current density and most noted HER onset potential decrease upon modification with 2D MoS₂ nanosheets. GC's production cost makes its application as a catalyst on an industrial scale economically unfeasible. The SPE offers an attractive alternative to using GC in real-world industrial applications. After modification with 2D MoS₂ nanosheets, the HER onset potential is lowered to a value equivalent of GC's. One issue could potentially be that the modified SPE exhibits a lower current density compared to that of GC, however, due to the nature of SPE's production they have can be produced to a wide manner of tailorability, varying in shape, surface area and carbon composition. As such SPEs can be produced with a larger surface area than that of GC, thereby increasing the currents possible and ultimately the amount of hydrogen produced. This combined with the ultra-low cost of fabricating SPEs makes them an exceptionally cost effective and easily producible supporting electrode material for HER. There is potential to incorporate 2D MoS₂ nanosheets into the carbon based inks used to produce SPEs, thereby eliminating the time consuming modification step. However, further research aimed at identifying and resolving the reason for the poor catalytic stability is essential if 2D MoS₂ nanosheet modified SPEs have a future industrial application.

4. Conclusions

We have explored the catalytic performance of 2D MoS₂ nanosheets towards the HER. In order to overcome issues prevalent within the literature and to enable us to ascertain the true electrochemical performance of our commercially sourced MoS₂ we modified a range of carbon based underlying support electrodes, namely GC, BDD, EPPG and SPEs; this approach is usually neglected within the literature. Application of the MoS₂ modified electrodes revealed a catalytic performance towards the HER, with lower onset potentials and higher current densities observed when utilising our target material. The supporting electrode was found to be of key importance, influencing the improvements observed in the electrochemical performance. This work indicates that 2D MoS₂ nanosheet modified electrodes can potentially serve as a viable, low cost and more abundant replacement to current Pt based electro-catalysts.

This work is distinct from the literature given that we have also correlated our electrochemical responses with supplementary Raman mapping of the electrode surface (and other complementary physicochemical characterisation) whilst exploring the effect of MoS₂ coverage. Coverage studies revealed that the catalytic effect of the 2D MoS₂ nanosheets increased (as indicated by ToF and ‘number of active sites’ calculations) until a ‘critical mass’ (coverage) was achieved, after which the response was observed to plateau. The likely cause of this effect is inferred herein and has clear implications (in this case and) when employing other 2D nanosheet modified electrodes within the literature.

We have provided insights into the observable electrochemistry and HER mechanism prevalent at 2D MoS₂ nanosheet modified electrodes, which has clear potential to be beneficially applied/utilised as an electrocatalyst if the diligent control measures reported herein are sufficiently applied.

Acknowledgements

D. A. C. Brownson acknowledges funding from the Ramsay Memorial Fellowships Trust and a British Council Institutional Link grant (No. 172726574) for the support of this research.

References

1. Y. Z. Yufei Zhao, Z. Yang, Y. Yan and K. Sun, Sci. Technol. Adv. Mater., 2013, 14.
2. T. Wang, L. Liu, Z. Zhu, P. Papakonstantinou, J. Hu, H. Liu and M. Li, Energy Environ. Sci., 2013, 6, 625–633.
3. A. J. López-Menéndez, R. Pérez and B. Moreno, J. Environ. Manage., 2014, 145, 368–373.
4. T. D. M. G. Schultz, G. P. Brasseur and W. Zittel, Science, 2003, 302, 624–627.
5. T. Hua, R. Ahluwalia, L. Eudy, G. Singer, B. Jermer, N. Asselin-Miller, S. Wessel, T. Patterson and J. Marcinkoski, J. Power Sources, 2014, 269, 975–993.
6. J. Andrews and B. Shabani, WIREs Energy Environ., 2014, 3, 474–489.
7. B. Hinnemann, P. G. Moses, J. Bonde, K. P. Jørgensen, J. H. Nielsen, S. Horch, I. Chorkendorff and J. K. Nørskov, J. Am. Chem. Soc., 2005, 127, 5308–5309.
8. M. Gratzel, Inorg. Chem., 2005, 44, 6841–6851.
9. J. P. Barton and D. G. Infield, IEEE Trans. Energy Convers, 2004, 19, 441–448.
10. F. Blaabjerg, Z. Chen and S. B. Kjaer, IEEE Trans. Power Electron., 2004, 19, 1184–1194.
11. N. M. Marković and P. N. Ross Jr., Surf. Sci. Rep., 2002, 45, 117–229.
12. N. M. Marković, B. N. Grgur and P. N. Ross, J. Phys. Chem. B, 1997, 101, 5405–5413.
13. A. B. Laursen, S. Kegnaes, S. Dahl and I. Chorkendorff, Energy Environ. Sci., 2012, 5, 5577–5591.
14. T. F. Jaramillo, K. P. Jørgensen, J. Bonde, J. H. Nielsen, S. Horch and I. Chorkendorff, Science, 2007, 317, 100–102.
15. B. X. Ya Yan, Z. Xu and X. Wang, ACS Catal., 2014, 693–1705.
16. K. Ravindra, L. Bencs and R. Van Grieken, Sci. Total Environ., 2004, 318, 1–43.
17. R. L. McCreery, Chem. Rev., 2008, 108, 2646–2687.
18. D. A. C. Brownson and C. E. Banks, Electrochem. Commun., 2011, 13, 111–113.
19. M. Lu, K. E. Toghill and R. G. Compton, Electroanalysis, 2011, 23, 1089–1094.
20. M. Musameh, J. Wang, A. Merkoci and Y. Lin, Electrochem. Commun., 2002, 4, 743–746.
21. A. Koca, Int. J. Hydrogen Energy Int., 2009, 34, 2107–2112.
22. Y. Shen, J. Liu, J. Jiang, B. Liu and S. Dong, J. Phys. Chem. B, 2003, 107, 9744–9748.
23. J. Yin, J. B. Jia and L. D. Zhu, In, 2008, 33, 7444–7447.
24. Y. Zheng, Y. Jiao, Y. Zhu, L. H. Li, Y. Han, Y. Chen, A. Du, M. Jaroniec and S. Z. Qiao, Nat. Commun., 2014, 5.
25. Z. Chen, D. Cummins, B. N. Reinecke, E. Clark, M. K. Sunkara and T. F. Jaramillo, Nano Lett., 2011, 11, 4168–4175.
26. D. Y. Chung, S.-K. Park, Y.-H. Chung, S.-H. Yu, D.-H. Lim, N. Jung, H. C. Ham, H.-Y. Park, Y. Piao, S. J. Yoo and Y.-E. Sung, Nanoscale, 2014, 6, 2131–2136.
27. Y. Li, H. Wang, L. Xie, Y. Liang, G. Hong and H. Dai, J. Am. Chem. Soc., 2011, 133, 7296–7299.
28. D. Merki, S. Fierro, H. Vrubel and X. Hu, Chem. Sci., 2011, 2, 1262–1267.
29. Q. Liu, J. Tian, W. Cui, P. Jiang, N. Cheng, A. M. Asiri and X. Sun, Angew. Chem., Int. Ed., 2014, 126, 6828–6832.
30. L. F. Pan, Y. H. Li, S. Yang, P. F. Liu, M. Q. Yu and H. G. Yang, Electrochem. Commun., 2014, 50, 13135–13137.
31. Y. H. Chang, C. T. Lin, T. Y. Chen, C. L. Hsu, Y. H. Lee, W. J. Zhang, K. H. Wei and L. J. Li, Adv. Mater., 2013, 25, 756–760.
32. P. Joensen, E. D. Crozier, N. Alberding and R. F. Frindt, J. Phys. C: Solid State Phys., 1987, 20, 4043–4053.
33. J. H. Lee, W. S. Jang, S. W. Han and H. K. Baik, Langmuir, 2014, 30, 9866–9873.
34. S. Ji, Z. Yang, C. Zhang, Z. Liu, W. W. Tjiu, I. Y. Phang, Z. Zhang, J. Pan and T. Liu, Electrochim. Acta, 2013, 109, 269–275.
35. B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti and A. Kis, Nat. Nanotechnol., 2011, 6, 147–150.
36. S. Tongay, J. Zhou, C. Ataca, K. Lo, T. S. Matthews, J. Li, J. C. Grossman and J. Wu, Nano Lett., 2012, 12, 5576–5580.
37. M.-L. Tsai, S.-H. Su, J.-K. Chang, D.-S. Tsai, C.-H. Chen, C.-I. Wu, L.-J. Li, L.-J. Chen and J.-H. He, ACS Nano, 2014, 8, 8317–8322.
38. C. Ataca, M. Topsakal, E. Aktürk and S. Ciraci, J. Phys. Chem. C, 2011, 115, 16354–16361.
39. N. A. Dhas and K. S. Suslick, J. Am. Chem. Soc., 2005, 127, 2368–2369.
40. F. Cesano, S. Bertarione, A. Piovano, G. Agostini, M. M. Rahman, E. Groppo, F. Bonino, D. Scarano, C. Lamberti, S. Bordiga, L. Montanari, L. Bonoldi, R. Millini and A. Zecchina, Catal. Sci. Technol., 2011, 1, 123–136.
41. S. S. Chou, M. De, J. Kim, S. Byun, C. Dykstra, J. Yu, J. Huang and V. P. Dravid, J. Am. Chem. Soc., 2013, 135, 4584–4587.
42. D. Wang, Z. Pan, Z. Wu, Z. Wang and Z. Liu, J. Power Sources, 2014, 264, 229–234.
43. D. Voiry, M. Salehi, R. Silva, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nano Lett., 2013, 13, 6222–6227.
44. N. Tateishi, K. Yahikozawa, K. Nishimura and Y. Takasu, Electrochim. Acta, 1992, 37, 2427–2432.
45. M.-R. Gao, M. K. Y. Chan and Y. Sun, Nat. Commun., 2015, 6.
46. Y. Yu, S.-Y. Huang, Y. Li, S. N. Steinmann, W. Yang and L. Cao, Nano Lett., 2014, 14, 553–558.
47. J. D. Benck, Z. Chen, L. Y. Kuritzky, A. J. Forman and T. F. Jaramillo, ACS Catal., 2012, 2, 1916–1923.
48. C. W. Foster, J. P. Metters and C. E. Banks, Electroanalysis, 2013, 25, 2275–2282.
49. A. B. Laursen, A. S. Varela, F. Dionigi, H. Fanchiu, C. Miller, O. L. Trinhammer, J. Rossmeisl and S. Dahl, J. Chem. Educ., 2012, 89, 1595–1599.
50. D. Marin, F. Medicuti and C. Teijeiro, J. Chem. Educ., 1994, 71, A277.
51. L. R. Cumba, J. P. Smith, D. A. C. Brownson, J. Iniesta, J. P. Metters, D. R. d. Carmo and C. E. Banks, Analyst, 2015, 140, 1543–1550.
52. C. W. Foster, J. Pillay, J. P. Metters and C. E. Banks, Sensors, 2014, 14, 21905–21922.
53. J. P. Metters, M. Gomez-Mingot, J. Iniesta, R. O. Kadara and C. E. Banks, Sens. Actuators, B, 2013, 177, 1043–1052.
54. E. P. Randviir, D. A. C. Brownson, J. P. Metters, R. O. Kadara and C. E. Banks, Phys. Chem. Chem. Phys., 2014, 16, 4598–4611.
55. J. P. Smith, J. P. Metters, O. I. G. Khreit, O. B. Sutcliffe and C. E. Banks, Anal. Chem., 2014, 86, 9985–9992.
56. O. Ramdani, J. P. Metters, L. C. S. Figueiredo-Filho, O. Fatibello-Filho and C. E. Banks, Analyst, 2013, 138, 1053–1059.
57. J. P. Smith, J. P. Metters, D. K. Kampouris, C. Lledo-Fernandez, O. B. Sutcliffe and C. E. Banks, Analyst, 2013, 138, 6185–6191.
58. Graphene Supermarket, https://graphene-supermarket.com/MoS2-Pristine-Flakes-in-Solution.html, Accessed [24/11/2014].
59. Innovation in Surface Spectroscopy and Microscopy Systems, http://www.specs.de/cms/front_content.php?idcat=209, Accessed [22/05/2015].
60. M. P. Seah and S. J. Spencer, J. Electron. Spectrosc. Relat. Phenom., 2006, 151, 178–181.
61. J. H. Scofield, J. Electron. Spectrosc. Relat. Phenom., 1976, 8, 129–137.
62. CasaXPS, http://www.casaxps.com/, Accessed 22/05/2015.
63. E. Varrla, C. Backes, K. R. Paton, A. Harvey, Z. Gholamvand, J. McCauley and J. N. Coleman, Chem. Mater., 2015, 27, 1129–1139.
64. C. Backes, R. J. Smith, N. McEvoy, N. C. Berner, D. McCloskey, H. C. Nerl, A. O'Neill, P. J. King, T. Higgins, D. Hanlon, N. Scheuschner, J. Maultzsch, L. Houben, G. S. Duesberg, J. F. Donegan, V. Nicolosi and J. N. Coleman, Nat. Commun., 2014, 5.
65. L. Yadgarov, C. L. Choi, A. Sedova, A. Cohen, R. Rosentsveig, O. Bar-Elli, D. Oron, H. Dai and R. Tenne, ACS Nano, 2014, 8, 3575–3583.
66. J. A. Wilson and A. D. Yoffe, Adv. Phys., 1969, 19, 193–335.
67. H. Li, Q. Zhang, C. C. R. Yap, B. K. Tay, T. H. T. Edwin, A. Olivier and D. Baillargeat, Adv. Funct. Mater., 2012, 22, 1385–1390.
68. S. Shin, Z. Jin, D. H. Kwon, R. Bose and Y.-S. Min, Langmuir, 2015, 31, 1196–1202.
69. J. Bonde, P. G. Moses, T. F. Jaramillo, J. K. Norskov and I. Chorkendorff, Faraday Discuss., 2009, 140, 219–231.
70. S. Ding, D. Zhang, J. S. Chen and X. W. Lou, Nanoscale, 2012, 4, 95–98.
71. A. H. Loo, A. Bonanni, A. Ambrosi and M. Pumera, Nanoscale, 2014, 6, 11971–11975.
72. D. Kong, H. Wang, J. J. Cha, M. Pasta, K. J. Koski, J. Yao and Y. Cui, Nano Lett., 2013, 13, 1341–1347.
73. C. Lee, H. Yan, L. E. Brus, T. F. Heinz, J. Hone and S. Ryu, ACS Nano, 2010, 4, 2695–2700.
74. K. Zhang, S. Hu, Y. Zhang, T. Zhang, X. Zhou, Y. Sun, T.-X. Li, H. J. Fan, G. Shen, X. Chen and N. Dai, ACS Nano, 2015, 9, 2704–2710.
75. J. D. Benck, T. R. Hellstern, J. Kibsgaard, P. Chakthranont and T. F. Jaramillo, ACS Catal., 2014, 4, 3957–3971.
76. S. Helveg, J. V. Lauritsen, E. Lægsgaard, I. Stensgaard, J. K. Nørskov, B. S. Clausen, H. Topsøe and F. Besenbacher, Phys. Rev. Lett., 2000, 84, 951–954.
77. J. Z. Xie, H. Li, S. Wang, R. Sun, X. Zhou, M. Zhou, J. Lou, X. Wen and X. Yi, Adv. Mater., 2013, 25, 5807–5813.
78. M. A. Lukowski, A. S. Daniel, F. Meng, A. Forticaux, L. Li and S. Jin, J. Am. Chem. Soc., 2013, 135, 10274–10277.
79. D. A. C. Brownson, A. C. Lacombe, D. K. Kampouris and C. E. Banks, Analyst, 2012, 137, 420–423.
80. D. A. C. Brownson, L. C. S. Figueiredo-Filho, X. Ji, M. Gomez-Mingot, J. Iniesta, O. Fatibello-Filho, D. K. Kampouris and C. E. Banks, J. Mater. Chem. A, 2013, 1, 5962–5972.
81. D. A. C. Brownson, L. J. Munro, D. K. Kampouris and C. E. Banks, RSC Adv., 2011, 1, 978–988.
82. D. A. C. Brownson, D. K. Kampouris and C. E. Banks, Chem. Soc. Rev., 2012, 41, 6944–6976.
83. T. F. Jaramillo, K. P. Jørgensen, J. Bonde, J. H. Nielsen, S. Horch and I. Chorkendorff, Science, 2007, 317, 100–102.
84. M.-R. Gao, J.-Z. Liang, Y.-R. Zheng, Y.-F. Xu, J. Jiang, Q. Gao, J. Li and S.-H. Yu, Nat. Commun., 2015, 6.
85. X. Bian, J. Zhu, L. Liao, M. D. Scanlon, P. Ge, C. Ji, H. H. Girault and B. Liu, Electrochem. Commun., 2012, 22, 128–132.
86. H. Vrubel, D. Merki and X. Hu, Energy Environ. Sci., 2012, 5, 6136–6144.
87. D. Merki, H. Vrubel, L. Rovelli, S. Fierro and X. Hu, Chem. Sci., 2012, 3, 2515–2525.
88. H. Zhu, F. Lyu, M. Du, M. Zhang, Q. Wang, J. Yao and B. Guo, ACS Appl. Mater. Interfaces, 2014, 6, 22126–22137.
89. J. R. McKone, B. F. Sadtler, C. A. Werlang, N. S. Lewis and H. B. Gray, ACS Catal., 2012, 3, 166–169.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5nr05164a|ART

---