WO2018165449A1

WO2018165449A1 - Electrocatalyst for hydrogen evolution reaction

Info

Publication number: WO2018165449A1
Application number: PCT/US2018/021580
Authority: WO
Inventors: Yun-Hyuk Choi; Mohammed AL-HASHIMI; Lei Fang; Sarbajit Banerjee
Original assignee: Qatar Foundation For Education, Science And Community Development
Priority date: 2017-03-09
Filing date: 2018-03-08
Publication date: 2018-09-13
Also published as: CA3055828A1; DE112018001227T5; US20200048783A1

Abstract

The electrocatalyst for hydrogen evolution reaction includes nanosheets of molybdenum disulfide (MoS2) deposited on a carbon fiber substrate. The catalyst is formed in stepwise fashion by chemical vapor deposition of nanosheets of MoO3 onto the substrate, then reducing the MoO3 to nanosheets of MoO2 using sublimed sulfur, then by reaction of sulfur vapor with the MoO2 to form nanosheets of MoS2 on the carbon fiber substrate. The catalyst is multifaceted, having a large density of edges providing catalytically active sites for the hydrogen evolution reaction. The activity of the catalyst is enhanced by coating the catalyst with spherical fullerenes (nC60).

Description

ELECTROCATALYST FOR HYDROGEN EVOLUTION REACTION

TECHNICAL FIELD

The disclosure of the present patent application relates to an electrocatalyst for hydrogen evolution reaction, and particularly to an electrocatalyst for hydrogen evolution reaction that includes nanosheets of molybdenum disulfide (M0S₂) deposited on a carbon fiber substrate.

BACKGROUND ART

Sunlight shining on earth is intermittent. As such, a fundamental impediment to meaningful utilization of sunlight is the effective storage of solar energy. Water splitting, or the disproportionation of ¾0 into ¾ and O₂, represents a promising strategy but is non- trivial because it requires the concerted transfer of four electrons and four protons. In nature, the complex biological machinery for photosynthesis couples multiple platforms wherein the light harvesting, water oxidation, and proton reduction steps are each performed by discrete components. Consequently, there is much interest in hybrid structures wherein discrete components perform each of the individual reactions required for photocatalysis. A viable photocatalytic cycle can be constituted by coupling photocatalytic water oxidation with electrocatalytic hydrogen evolution. The latter hydrogen evolution reaction (HER), however, is beset by a distinctive set of challenges.

The Pt group metals are excellent catalysts for HER and evolve hydrogen at near- zero overpotentials in acidic media but are cost prohibitive and amongst the least abundant elements available to mankind. There has been a strong push to develop alternatives and some success was achieved with M0S₂ as well as transition metal phosphides. In those strategies, however, the electrocatalytic activity of M0S₂ was mainly derived from catalytically active edge sites. The basal planes were thought to be catalytically inert with some exceptions. Furthermore, the low charge carrier mobility of M0S₂ was an impediment to its use as an electrocatalyst. Also, those materials generally tend to evolve ¾ at high overpotentials in comparison to Pt.

Chemical vapor deposition is ubiquitously used to prepare well-crystallized M0S₂ architectures, typically using molybdenum oxide or chloride precursors. A major drawback of this method as applied to the growth of M0S₂ is that it necessitates the operation of several concurrent reactions. Consequently, previously obtained M0S₂ electrocatalyst samples are often plagued by poor size and shape homogeneity, with sparse substrate coverage.

Thus, an electrocatalyst for hydrogen evolution reaction solving the aforementioned problems are desired.

DISCLOSURE OF INVENTION

The electrocatalyst for hydrogen evolution reaction includes homogeneously sized nanosheets of molybdenum disulfide (M0S₂) integrated on a carbon fiber paper substrate, with exposed catalytically active edge sites. The nanosheets are well distributed and vertically oriented. The electrocatalyst is prepared by the stepwise reduction and sulfidation of M0O₃. The direct integration of edge-exposed M0S₂ nanosheets onto CFP yields a 3D architecture with a high surface-to- volume ratio desirable for electrocatalytic applications. The inherent HER activity of the edge-sites of M0S₂ can be enhanced significantly by interfacing with nC₆o nanoclusters, as a result of the enhancement of the conductivity of M0S₂ owing to charge transfer.

These and other features of the present disclosure will become readily apparent upon further review of the following specification and drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is schematic diagram illustrating the steps for preparing the electrocatalyst for hydrogen evolution reaction.

Fig. 2A is a FESEM image showing homogeneous distribution of M0S₂ nanosheets grown on textured CFP substrate.

Fig. 2B is a high-magnification SEM image of an individual M0S₂ nanosheet.

Fig. 2C is a low-magnification TEM image of a nanosheet depicting locations of "clean" well-faceted and "collapsed" edges.

Fig. 2D is an HRTEM image of a "clean" edge.

Fig. 2E is an HRTEM image of a discontinuous "collapsed" edge.

Fig. 3A is a FESEM image of M0O₃ harvested from a flat Si(100) substrate.

Fig. 3B is a FESEM image of M0O₂ harvested from a flat Si(100) substrate.

Fig. 3C is a FESEM image of M0S₂ harvested from a flat Si(100) substrate.

Fig. 3D is a FESEM image of M0S₂ nanosheets harvested from a flat Si(100) substrate.

Fig. 3E is a FESEM image of M0S₂ nanosheets harvested from a flat Si(100) substrate.

Fig. 3F is a FESEM image of M0S₂ nanosheets harvested from a flat Si(100) substrate, showing additional edge sites created on the basal plane of the nanosheets.

Fig. 4A shows XRD patterns of M0O₃ nanosheets, M0O₂ nanodisks, and M0S₂ nanosheets prepared on CFP.

Fig. 4B shows Raman spectra (514.5 nm laser excitation) of M0O₃ nanosheets, M0O₂ nanodisks, and M0S₂ nanosheets prepared on CFP.

Fig. 4C shows XPS spectra indicating Mo 3d binding energies.

Fig. 4D shows XPS spectra indicating O ls binding energies

Fig. 4E shows XPS spectra indicating S 2p binding energies

Fig. 5 shows Raman spectra of the nCeo cluster and hybrid nC₆o MoS₂ architectures.

Fig. 6 shows the mass spectrum of negatively charged ions emitted from the surface of the clusters deposited on CFP.

Fig. 7A shows polarization curves of various concentrations of nCeo clusters deposited directly onto CFP.

Fig. 7B shows polarization curves of various concentrations of nCeo clusters interfaced with 3D M0S₂ nanosheets on CFP.

Fig. 8A shows a cyclic voltammogram acquired in the range between 0.10-0.30 V vs. RHE for neat nCeo clusters deposited on CFP from solutions of C₆o concentration of 0.1 mg/mL.

Fig. 8B shows a cyclic voltammogram acquired in the range between 0.10-0.30 V vs. RHE for neat nC₆o clusters deposited on CFP from solutions of C₆o concentration of 0.5 mg/mL.

Fig. 8C shows a cyclic voltammogram acquired in the range between 0.10-0.30 V vs. RHE for neat nC₆o clusters deposited on CFP from solutions of C₆o concentration of 2.0 mg/mL.

Fig. 8D is a graph showing differences in current density Aj=j_a-j_c at 0.20 V versus RHE, plotted as a function of scan rate with each plot fitted to a straight line to determine the C_di values.

Fig. 9A shows a voltammogram acquired in the range of 0.15-0.30 V versus RHE for as-prepared 3D M0S₂.

Fig. 9B shows a voltammogram acquired in the range of 0.15-0.30 V versus RHE for hybrid nC₆₀ (0.1 mg/mL)/MoS₂.

Fig. 9C shows a voltammogram acquired in the range of 0.15-0.30 V versus RHE for for hybrid nC₆o (0.5 mg/mL)/MoS₂.

Fig. 9D shows a voltammogram acquired in the range of 0.15-0.30 V versus RHE for for hybrid nC₆₀ (2.0 mg/mL)/MoS₂.

Fig. 9E is a graph showing differences in current density (Aj = j_a-j_c) at 0.23 V versus RHE are plotted as a function of the scan rate (the C_di values are extrapolated from a linear fit to the plot).

Fig. 10A shows the Nyquist plots of as-prepared 3D M0S₂ nanosheets and hybrid nC₆o MoS₂ architectures prepared on CFP measured at -150 mV vs. RHE.

Fig. 10B shows R_ct values plotted as a function of potential for nCeo, as-prepared 3D M0S₂ nanosheets, and hybrid nC₆o MoS₂ architectures prepared on CFP.

Similar reference characters denote corresponding features consistently throughout the attached drawings.

BEST MODE(S) FOR CARRYING OUT THE INVENTION

The electrocatalyst for hydrogen evolution reaction includes high-edge-density M0S₂ nanosheets directly integrated with conductive carbon fiber paper (CFP). The nanosheets can be vertically aligned. The nanosheets can be 3-dimensional, polycrystalline nanosheets. The nanosheets are homogeneously dispersed across centimeter scales and have a high density of exposed edge sites. The nanosheets can include a mixture of faceted as well as discontinuous collapsed edges within the basal planes, induced by volume expansion accompanying topochemical sulfidation, as described herein. The M0S₂ nanosheets on the CFP can exhibit an overpotential ηιο value of about 245 mV at 10 mA/cm² a Tafel slope of about 81 mV/dec, and a turnover frequency (TOF) of about 1.28 ¾/s per active site at -0.2 V versus RHE in a 0.5 M acidic solution.

The catalyst can further include Buckminsterfullerenes or spherical fullerenes (nC₆o). For example, the 3D M0S₂ nanosheets can be interfaced with nCeo clusters by a facile solution-deposition method. The hybrid structures show greatly enhanced HER activity with an overpotential η₁₀ value of about 172 mV and a Tafel slope of about 60 mV/dec when the deposition concentration of C₆o is about 0.5 mg/mL. This condition corresponds to about a 2% coverage of the M0S₂ nanosheets by nCeo clusters. The improved activity of the hybrid catalysts is believed to derive from the interfacial charge transfer at nC₆o MoS₂ p— n heteroj unctions. An optimal coverage of nCeo with a homogeneous distribution can facilitate such interfacial doping. The catalyst can be formed in stepwise fashion by chemical vapor deposition of nanosheets of M0O₃ onto a carbon fiber substrate, reducing the M0O₃ to nanosheets of M0O₂ using sublimed sulfur, then reacting sulfur vapor with the M0O₂ to form nanosheets of M0S₂ on the carbon fiber substrate. The resulting catalyst is multifaceted, having a large density of edges providing catalytically active sites for hydrogen evolution reaction.

The stepwise vapor transport, reduction, and sublimation steps used to prepare edge- exposed M0S₂ nanosheets on CFP are schematically illustrated in Fig. 1. In the first step, M0O₃ nanosheets that are about 1-2 μιη in lateral dimensions are deposited onto CFP by the vapor transport of M0O₃ powder heated to 850°C.

In the next step, reaction with sublimed sulfur at 400°C as per:

2M0O3 (s) + S (g) 2M0O2 (s) + SO2 (g)...(1)

yields M0O₂ nanosheets with retention of the vertical growth orientation, although the edges are slightly rounded. Finally, the topochemical sulfidation of M0O₂ at 850°C as per:

M0O2 (s) + 3S (g) MoS₂ (s) + SO2 (g)...(2)

yields faceted M0S₂ nanosheets that are uniformly dispersed and vertically oriented across a large area (about 2 cm²) of the CFP (Fig. 2A). Figs. 2B-2E illustrate "clean," well- faceted, and "collapsed" edges of an individual M0S₂ nanosheet.

Figs. 3A-3F illustrate the morphologies of vapor transported M0O₃ collected on a Si (100) substrate before and after stepwise reduction and sulfidation. These images further enable visualization of the vertical growth direction and high density of edge sites. Notably, the vertical growth orientation is achieved without mediation of a catalyst. Interestingly, the faceted M0O₃ nanosheets are transformed to thicker rounded M0O₂ discs upon reduction and finally converted to faceted M0S₂ nanosheets during sulfidation. The edge geometries in large measure reflect the intrinsic crystal structures of the phases. Orthorhombic α-Μοθ₃ crystallizes in a layered structure and thus faceted nanosheets are obtained comprising stacked layers. Reduction to monoclinic M0O₂ yields rounded edges, whereas topochemical transformation to 2H-M0S₂ again yields faceted structures reflecting the layered stacking of M0S₂ sheets. The considerable lattice mismatch between M0O₂ and M0S₂ results in a substantial volume change, which creates a distinctive discontinuous motif characterized by faceted "clean" and discontinuous "collapsed" domains along the M0S₂ basal planes. The latter is important as it allows for exposure of an increased density of catalytically active edge-sites.

Figs. 4A-4E corroborate the phase identification of the prepared materials based on X-ray diffraction (XRD) and Raman microprobe analysis. The XRD patterns acquired on CFP are dominated by the (002) reflections of the graphitic substrate. However, reflections corresponding to the deposited materials are discernible and are indexed to orthorhombic a- M0O₃ (Joint Committee on Powder Diffraction Standards (JCPDS) 76-1003), monoclinic M0O2 (JCPDS 86-0135), and hexagonal 2H-MoS₂ (JCPDS 87-2416) as shown in Fig. 4A. Clearer phase assignment is enabled from the Raman spectra shown in Fig. 4B since the graphitic D and G bands from the substrate are only observed above 1300 cm^"1. The Raman bands of the nanosheets formed in the first step are well matched with the Raman active modes of orthorhombic α-Μοθ₃ reported in the literature. The detailed Raman band assignments of the prepared α-Μοθ₃ nanosheets are listed in Table 1.

Table 1 : Photon Mode Assignments for Raman bands measured for α-Μοθ₃ nanosheets

α-Μοθ₃ nanosheet

(produced by the 1st process)

Bands (cm-1) Raman modes Assignments

995 Ag v_as 0=Mo stretch

819 Ag v_s 0=Mo=0 stretch

665 B_2g/B_3g v_as Mo-O-Mo stretch

472 A_g v_as Mo-O-Mo stretch and bend

378 B ig δ 0=Mo=0 scissoring

364 Ag δ 0=Mo=0 scissoring

337 Ag/B_lg δ Mo-O-Mo bend

290 B_3g δ 0=Mo=0 wagging

282 B_2g δ 0=Mo=0 wagging

245 B_3g τ 0=Mo=0 twist

216 Ag rotational rigid Mo0₄-chain mode, R_c

196 B_2g τ 0=Mo=0 twist

157 Ag/B ig translational rigid Mo0₄-chain mode, Tb

127 B_3g translational rigid Mo0₄-chain mode, Tc

114 B_2g translational rigid Mo0₄-chain mode, T_c

The Raman spectra of the nanodiscs formed by the reduction of the α-Μοθ₃ nanosheet using sulfur are an excellent match for phonon modes of monoclinic Mo0₂ as reported previously in the literature. The sulfide structures on CFP show clear Raman signatures of 2-H MoS₂ including Raman bands at 282, 377, and 404 cm^"1, which can be ascribed to modes of Ei_g, E^¹, and Ai_g symmetry, respectively. X-ray photoelectron spectroscopy (XPS) analysis was performed by acquiring Mo 3d, O Is, and S 2p core level spectra of each product, to investigate the evolution of the chemical composition (Figs. 4C - 4E). The Mo 3d core level spectra are characterized by a distinctive doublet at 233.20 and 236.35 eV ascribed to the binding energies of Mo 3ds_/2 and 3d_3/2 states, respectively; these binding energies are characteristic of hexavalent molybdenum, verifying stabilization of the M0O₃ phase. The O Is singlet at 530.95 eV is further in good agreement with the value expected for an inorganic oxide.

Upon reduction with sulfur, the XPS spectra for the nanodiscs shows a substantial alteration of the Mo 3d binding energies to 229.80 and 233.05 eV for the Mo 3ds_/2 and 3d_3/2 states, suggesting the stabilization of a tetravalent oxide of molybdenum. A remnant shoulder at 236.35 eV attributable to the binding energy of Mo 3d_3/2 for hexavalent molybdenum indicates incomplete reduction. Corresponding features in the O Is core level spectra at 530.75 and 531.85 eV, can be attributed to M0O₂ and M0O₃ respectively. Furthermore, a distinctive doublet is discernible in S 2p core level spectra at 162.75 and 163.80 eV and can be ascribed to S 2p_3/2 and S 2p binding energies, respectively, revealing surface sulfidation forms some M0S₂ even at a temperature of 400°C. The nanosheets after sulfidation at 850°C show Mo 3d core level spectra at 229.70 and 232.85 eV attributable to binding energies for Mo 3ds_/2 and Mo 3d_3/2, respectively; these values are characteristic of M0S₂. The small shoulder at 226.95 eV is attributed to S 2s. A much more pronounced doublet is observed in S core level spectra at 162.65 and 163.75 eV assigned to S 2ρ₃β and S 2pi_/2 binding energies, respectively. These values verify the sulfidation of M0O₂. A broad O Is spectrum with a peak at 532.60 eV is attributed to surface-adsorbed oxygen species.

The enthalpy of hydrogen adsorption on M0S₂ edges has been estimated to be endo thermic by c a. 0.08 eV and the extent of H-coverage is limited to one in four atoms at the edges of M0S₂. Reducing the overpotential and increasing catalytic efficiency requires a further decrease of the hydrogen adsorption enthalpy and an increase of the extent of H- coverage. One approach involves polarizing Mo— S bonds at the edges via electronic coupling with electron-donating or withdrawing moieties, ideally other semiconductors. Here, the faceted M0S₂ nanosheets were interfaced with nCeo clusters deposited from solution to prepare hybrid architectures. Upon solution deposition from chlorobenzene solution (nC₆o of 0.5 mg/mL), nCeo clusters that are ca. 7 μιη in diameter are deposited onto the fibers of CFP . Similar morphologies of nCeo were grown on the M0S₂ nanosheets. Energy dispersive X-ray spectroscopy (EDS) maps acquired at C, Mo, and S elemental edges verified the co- localization of the C₆o clusters atop the M0S₂ nanosheets. EDS line profiles further verified the co-localization of nCeo clusters on the M0S₂ basal planes. The Raman spectra of the nCeo cluster and hybrid nC₆o MoS₂ architectures are shown in Fig. 5. Distinctive Raman modes of C₆o are evidence in both spectra with bands assigned to phonons of A_g(l,2) and H_g(l-8) symmetry. Both M0S₂ and C₆o modes are discernible in the hybrid architecture. The coverage of nCeo clusters strongly depends on the concentration of the precursor solution. The size of the nCeo clusters increases with increasing concentration from 0.1 to 2.0 mg/mL. Notably, at a concentration of 0.1 mg/mL, the relatively small nC₆o clusters are homogenously distributed throughout the sample; however, upon increasing the concentration to 0.5 mg/mL, the homogeneity is somewhat reduced. Upon increasing the precursor concentration to 2.0 mg/mL, the nCeo clusters are mostly present as large agglomerations that are rather sparsely distributed across the surface. In the concentration range examined, the clusters do not appear to form a continuous percolative network. In order to evaluate quantitatively the coverage of nCeo clusters for a precursor concentration of 0.5 mg/mL, the sample deposited onto CFP has been examined by secondary ion mass spectrometry (SIMS) using 50 keV C₆o²⁺ ions as the source. The mass spectrum of negatively charged ions emitted from the surface of the clusters deposited on CFP is shown in Fig. 6. The ions emitted were carbon, hydrocarbon clusters, and intact molecular cluster ions of C₆o with a yield ( number of secondary ions detected per single projectile impact) of 0.08%. The C₆o clusters are ca. 7 μιη in diameter and cover ca. 2% of the total surface area of the carbon fibers of CFP. Indeed, the SEM and SIMS measurements indicate that optimal concentrations of the C₆o solution are necessary to maximize interfacial interactions and prevent crystallization into larger nCeo clusters.

The electrocatalytic HER performance of CFP based samples with nCeo clusters alone, as-prepared 3D M0S₂ nanosheets, and hybrid nC₆o MoS₂ architectures were investigated in a 0.5 M aqueous solution of H₂SO₄, using a conventional three-electrode setup. Bare CFP was contrasted as a control and was essentially catalytically inert towards HER. In contrast, nCeo (0.5 mg/mL) clusters on CFP exhibited a finite cathodic current density with an overpotential of 353 mV, reaching a current density of 10 mA/cm² (ηιο) and a Tafel slope of 169 mV/dec. The 3D faceted M0S₂ nanosheets on CFP showed HER activity with a ηιο value of 245 mV and a Tafel slope of 81 mV/dec. Remarkably, interfacing the M0S₂ nanosheets with nCeo resulted in a much lower overpotential. Hybrid nCeo (0.5 mg/mL)/MoS₂ structures had a ηιο value of 172 mV and a Tafel slope of 60 mV/dec. These results clearly indicate the synergistic enhancement of HER activity as a result of coupling between nCeo and M0S2. A high Tafel slope value of pristine nCeo on CFP (>120 mV/dec) indicated that HER proceeds through the Volmer mechanism, wherein proton reduction yielding hydrogen ad- atoms bound to the active sites represents the rate determining step. In contrast, low Tafel slope values measured for as-prepared 3D M0S2 and hybrid nC₆o MoS2 (60 and 80 mV/dec) suggest the operation of the Volmer-Heyrovsky mechanism wherein the rate-determining steps involve both proton reduction and hydrogen desorption. It is noteworthy that the HER performance of the 3D array of M0S2 nanosheets with a high density of exposed edge-sites and their hybrid structures interfaced with nC₆o are either higher or comparable to previously reported values for bulk or nanostructured M0S2. In addition, the hybrid materials reported here possess the advantages of well-defined architectures, conductive substrates, and scalability to centimeter-sized dimensions.

Figs. 7A-7B contrast the polarization curves of various concentrations of nCeo clusters either deposited directly onto CFP or interfaced with 3D M0S2 nanosheets on CFP. The cathodic current density of the neat nCeo cluster formed on CFP measured at -0.4 V versus RHE is gradually decreased from 22.6 to 18.5 to 14.6 mA/cm² as the concentration of C₆o deposition solution is increased from 0.1 to 0.5 to 2.0 mg/mL. With increasing concentration of C₆o solution, the overpotential ηιο is also increased from 331 to 353 to 363 mV. As noted above, the hybrid nC₆o MoS2 electrocatalyst prepared using 0.5 mg/mL C₆o deposition solution shows the best HER performance with the highest current density (7₀.₂v = 18.0 mA/cm² at 0.2 V vs. RHE), lowest ηι₀ value (172 mV), and the lowest Tafel slope (60 mV/dec). The nCeo (0.1 mg/mL)/MoS2 sample (7₀.₂v = 5.2 mA/cm², ηιο = 245 mV, and Tafel slope = 74 mV/dec) exhibits substantially worse performance that is analogous to the 3D M0S2 nanosheets without C₆o hybridization (7₀.₂v = 5.0 mA/cm², ηιο = 245 mV, and Tafel slope = 81 mV/dec). At such low concentrations, the limited nCeo coverage likely limits the extent to which the edge reactivity is modulated. At substantially higher solution concentrations of C₆o, the HER performance is diminished as well. The nCeo (2.0 mg/mL)/MoS2 sample is characterized by values of /o.2v = 3.9 mA/cm², ηιο = 273 mV, and a Tafel slope = 80 mV/dec. The lack of synergistic enhancement in the latter case can be attributed to the sparse and heterogeneous distribution of nCeo (2.0 mg/mL). The large agglomerations observed suggest that the buckministerfullerene clusters are not effectively interfaced with M0S2, which likely perturbs the electronic coupling necessary for improved HER performance as described below.

In order to examine the mechanistic basis for the observed modulation of electrocatalytic properties upon interfacing with nCeo, electrochemically active surface areas (ECSA) of all the samples were estimated by measuring the double-layer capacitance (C_di) from cyclic voltammetry (CV) data across a potential range with no Faradaic current. The voltammograms were collected at various scan rates (20— 100 mV/s) in the potential range of 0.10— 0.30 V versus RHE, where the current is preponderantly due to the charging of the double layer (and not due to proton reduction). Figs. 8A-8D and 9A-9E depict CV curves acquired for nCeo, 3D nanosheets of M0S2, and hybrid nC₆o MoS2 architectures with various C₆o concentrations. The differences (Δ/^') of anodic and cathodic current densities at 0.20 and 0.23 V versus RHE for each CV plot is shown as a function of the scan rate in Fig. 8D and Fig. 9E. The slope of each Aj versus scan rate plot is equal to a value of 2C_di. The ECSA were obtained from the ratio of the measured C_di with respect to the specific capacitance of flat crystalline M0S2 (ca. 66.7 μΡ/cm²). The resulting C_di and ECSA values are displayed as a function of C₆o concentration in Fig. 8C. Significantly, the C_di and ECSA of nCeo on CFP (9A) and hybrid nC₆o MoS2 on CFP (9B and 9C) are respectively lower and higher than those of 3D M0S2 nanosheets on CFP (9D), and are decreased with increasing C₆o concentration. From these results, it can be inferred that the nCeo clusters formed on CFP or M0S2/CFP are increasingly agglomerated and crystallized with increasing C₆o concentration in solution, which is consistent with the morphologies observed by SEM. Furthermore, the C_di and ECSA of hybrid nC₆o MoS2 appear to be the sum of those of nCeo and 3D M0S2 nanosheets on CFP. However, an increased concentration of electrochemically active sites does not necessarily translate to increased HER activity since the nCeo clusters alone are much less active as compared to the 3D M0S2 architectures. The decrease of C_di and ECSA with increasing C₆o concentration of the precursor solution leads only to a slight deterioration of the cathodic current density for nC₆o/CFP and the Tafel slope is mostly preserved, indicating that the changes in C_di, ECSA, and the resulting number of active sites do not fundamentally alter the HER mechanism (i.e., Volmer reaction in the neat C₆o) and rate (Fig. 9E). These two sets of observations suggest that the improved HER performance observed for the hybrid nCeo (0.5 mg/mL)/MoS2 electrocatalyst is derived from an intrinsic enhancement of the inherent catalytic activity of M0S2 for HER rather than an increase in the number of active sites upon C₆o deposition.

The turnover frequency (TOF), defined as the number of ¾ molecules evolved per active site per unit time, is an essential parameter to contrast the inherent catalytic activity of different systems. The TOF can be calculated using the expression TOF = 7NA/2Fn(ECSA), where / is the current density, NA is Avogadro's number, 2 represents the stoichiometric number of electrons consumed at the electrode during HER, F is Faraday's constant, n is the number of active sites (1.164 x 10¹⁵ cm^"2) on a flat surface of crystalline M0S₂, and ECSA is the electrochemically active surface area of the electrode. Fig. 7D plots the TOF (per active site) of the 3D M0S₂/CFP and hybrid nCeo (0.5 mg/mL)/MoS₂ structure prepared on CFP in the applied potential range of -0.1 to -0.3 V versus RHE; in this regime, the HER is controlled by electrode kinetics with minimal influence from other effects. The measured TOF of the hybrid nCeo (0.5 mg/mL)/MoS₂ structure at -0.2 V (2.33 ¾/s per active site) is nearly twice as high as that of 3D M0S₂ nanosheets (1.28 H₂/s per active site) on CFP. These results highlight the synergistic enhancement of the inherent catalytic activity of the edge sites of the M0S₂ nanosheets upon nCeo hybridization. The 3D hybrid architectures constructed on mesoporous CFP clearly represent viable electrocatalysts.

In order to further investigate a possible origin of the enhanced HER performance observed for the hybrid nCeo (0.5 mg/mL)/MoS₂ structure, electrochemical impedance measurements were performed at various potentials between 10 and -250 mV by sweeping the frequency from 200 kHz to 100 mHz with an AC amplitude of 10 mV. Fig. 10A shows the Nyquist plots of as-prepared 3D M0S₂ nanosheets and hybrid nC₆o MoS₂ architectures prepared on CFP measured at -150 mV vs. RHE. The Nyquist plots are fitted to an equivalent circuit model (inset of Fig. 10A) comprising the following elements: an ohmic resistance (R_s), a charge-transfer resistance (R_ct), constant phase element (Q), and a Warburg constant (W). The obtained R_ct values are plotted as a function of potential in Fig. 10B for nCeo, as-prepared 3D M0S₂ nanosheets, and hybrid nC₆o oS₂ architectures prepared on CFP. The kinetics of electrocatalytic HER on the different electrode samples can be evaluated based on their respective R_ct values with a lower R_ct value corresponding to a faster reaction rate. The resulting R_ct values reveal a similar trend as the ηιο and Tafel slope values deduced from the cathodic current density in polarization curves (Figs. 7A-7B and 8A-8B); specifically, the R_ct values of 3D M0S₂/CFP and hybrid nC₆o MoS₂ prepared on CFP are nearly two orders of magnitude lower than those of nC₆o/CFP. Furthermore, the lowest R_ct values are obtained for the hybrid nCeo (0.5 mg/mL)/MoS₂ structure. Taken together, these results suggest that the enhanced HER performance observed upon interfacing with nCeo derive in large measure from the increased conductance of the hybrid constructs when C₆o is appropriately interfaced with M0S₂.

Indeed, recent ab initio density functional theory calculations of C₆₀ M0S₂ constructs are particularly instructive in understanding the nature of the interface formed between these two semiconductors. Prior studies have determined that the lowest energy configuration for these heterostructures corresponds to the hexagonal rings of C₆o situating directly above S sites on the basal planes of M0S2 resulting in buckministerfullerene molecules being able to rotate freely on the surface. This configuration yields a Type-II interface with charge depletion from C₆o and charge accumulation on M0S2 estimated to be ca. 0.055 e^" per C₆o unit. This directional charge transfer is thought to be key to the reduced resistance of the hybrid constructs. Indeed, the Type-II alignment has been further verified by recent theoretical and experimental studies of C₆₀ M0S2 hybrids. Studies have predicted that the valence band edge of M0S2 (-4.5 eV) resides lower than that of C₆o (-3.8 eV), resulting in charge transfer and electron accumulation on M0S2 when the two semiconductors are interfaced. Upon application of an electric field, the steadily increasing electron density in M0S2 reduces the junction-barrier height, further allowing facile electron tunneling and transport and giving rise to conductive pathways along the interfaces of the resulting C₆₀ M0S2 p— n heteroj unctions. Therefore, based on the measured TOF, deduced resistance values, and charge transfer resistance values extrapolated from EIS data, the enhanced HER performance of the hybrid nCeo (0.5 mg/mL)/MoS2 structure likely derives from a charge transfer mechanism. The nCeo clusters donate electron density to M0S2 and give rise to a conductive interfacial layer that is much more effective at charge transport as compared to the relatively insulating basal planes of M0S2. Such charge transfer may also polarize the Mo— S bonds reducing the enthalpy of hydrogen adsorption. Notably, this mechanism, essentially invoking interfacial doping of M0S2, is quite distinct from hybrid MoS2/carbon nanotube and MoS2/graphene heterostructures wherein the latter components actually form conductive pathways for electron transport between the CFP electrodes and the catalytically active edge sites, thereby mitigating the poor transport characteristics of the basal planes of 2H-M0S2. The role of interfacial doping is further underscored by the dependence of HER performance on the concentration of the C₆o precursor solution and the morphology of the nCeo clusters. Agglomerated C₆o clusters that are homogeneously dispersed across the M0S2 basal planes will be ineffective at modulating the electronic structure of M0S2 through electron transfer.

To assess the long-term stability of nCeo (0.5 mg/mL)/CFP, 3D M0S2/CFP, and hybrid nCeo (0.5 mg/mL)/MoS2 on CFP as electrocatalysts for HER, CV sweeps have been performed for 1000 cycles in a 0.5 M aqueous solution of H2SO4 in the range between -0.2 and 0.2 V versus RHE at a scan rate of 100 mV/s. The polarization curve for the 3D M0S2 on CFP is almost exactly superimposable upon the initial data suggesting no degradation in performance (Fig. 7B). In contrast, after 1000 cycles, the hybrid nC₆o MoS2 catalysts show a slight increase of the overpotential ηιο to 181 mV and the Tafel slope is changed to 65 mV/dec. The observed changes are suggestive of the partial loss of C₆o clusters upon prolonged electrocatalytic cycling, which likely disrupts some of the interfacial charge transfer and thereby disrupts charge transport between the CFP substrate and active catalytic edges.

The electrocatalyst for hydrogen evolution reaction will now be illustrated by the following examples, which do not limit the scope defined by the appended claims.

Example 1

Preparation of M0S2 nanosheets, C₆o clusters, and their hybrid structures

The CVD processes were performed using a 1 -inch-diameter horizontal cold- wall quartz tube furnace equipped with gas flow controls. In the first step to prepare M0O₃ nanosheets, 15.0 mg of M0O₃ powder (Sigma-Aldrich, purity >99.5%) was placed within an alumina boat, which was placed at the center of tube. A bare CFP substrate (Toray Paper 120) with dimensions of 7 cm x 1 cm size was placed downstream from the M0O₃ source at a distance of 15 cm from the alumina boat. After an initial Ar purge for 30 min, the M0O₃ powder was heated to 850°C at a ramp rate of 20°C/min and transported under a 68.3 seem Ar flow at 1 atm. After holding at 850°C for 10 min, the furnace was allowed to cool naturally to room temperature. Subsequently, M0O₃ nanosheets integrated onto ca. 2 cm² areas of the CFP were recovered. Such nanosheets were reproducibly formed at a distance of ca. 18— 20 cm from the alumina boat. The MoCVdeposited CFP was cut to dimensions of 4 cm x 1 cm thereby preserving margins on all sides. This substrate was then placed at the center of the tube furnace but downstream at a distance of 20 cm from an alumina boat containing 100 mg of elemental sulfur powder (Alfa Aesar, 99.5% purity). Next, after purging with Ar, the reactor was heated to a temperature of 400°C at a ramp rate of 20°C/min under an Ar flow of 100 seem at 1 atm to facilitate the reaction of sublimed sulfur with the M0O₃ nanosheets. After holding at 400°C for 20 min, the furnace was then naturally cooled to room temperature. Subsequently, a final CVD step was performed by replacing the spent sulfur in the alumina boat with an additional 100 mg of fresh elemental sulfur. The reactor was heated to 850°C at a ramp rate of 20°C/min under a 100 seem flow of Ar at 1 atm for 20 min after which the furnace was allowed to naturally cool to room temperature. The CFP paper was then removed from the center of the furnace for characterization and electrocatalytic evaluation.

In order to prepare hybrid nC₆o MoS2 structures on CFP, C₆o powder (Strem Chemicals Inc., 99.9% purity) was dissolved in chlorobenzene at concentrations of 0.1, 0.5, and 2.0 mg/mL, respectively. The M0S2/CFP (as well as bare CFP as a control) were immersed within the chlorobenzene solutions for 1 min and then removed. Subsequently, the samples were annealed at 160°C for 10 min under a flowing Ar atmosphere.

Example 2

Structural characterization

The morphology of the prepared materials was examined by field-emission scanning electron microscopy using a JEOL JSM-7500F instrument. The edge-sites of M0S2 flakes harvested from the M0S2/CFP sample by ultrasonication for 1 h in toluene were examined by high-resolution transmission electron microscopy using a JEOL JEM-2010 instrument operated at an accelerating voltage of 200 keV. Phase assignment was performed with the help of X-ray diffraction using a Bruker D8-Advance instrument equipped with a Cu Ka source (λ = 1.5418 A) as well as by Raman microprobe analysis using a Jobin-Yvon HORIBA LabRAM HR800 instrument coupled to an Olympus BX41 microscope. Raman spectra were collected with excitation from the 514.5 nm line of an Ar-ion laser; the laser power was kept below 10 mW to minimize photooxidation. The chemical composition and oxidation states of M0O₃, M0O2, and M0S2 prepared on CFP were investigated by X-ray photoelectron spectroscopy (XPS, Omicron XPS) with Mg Ka radiation (1253.6 eV). Energy calibration was achieved by setting the Cls line from adventitious hydrocarbons to 284.8 eV. The elemental composition of the C₆o clusters deposited on CFP and MoS2-deposited CFP was examined by energy-dispersive X-ray spectroscopy (EDS) coupled to the FE-SEM system. The coverage of C₆o (0.5 mg/mL) clusters deposited on CFP was measured on a custom-made secondary ion mass spectrometer (SIMS) using C₆o²⁺ projectiles with an energy of 50 keV as the source.

Example 3

Electrochemical characterization

The HER performance of the prepared materials was evaluated using a three-electrode cell with the help of a Bio-Logic potentiostat (SP-200). All of the measurements were performed in a 0.5 M aqueous solution of H2SO4 purged with N2 gas. M0S2/CFP, C₆o/CFP, and the hybrid structures prepared on CFP were individually used as the working electrodes. A saturated calomel electrode (SCE) and a Pt plate were used as reference and counter electrodes, respectively. The potential versus SCE (ESCE) was converted to the potential versus the reversible hydrogen electrode (RHE) (ERHE) using the relation ERHE = ESCE + 0.279 V. Polarization curves for HER were measured using linear sweep voltammetry (LSV) in the range between 0.1 and -0.4 V versus RHE at a scan rate of 8 mV/s. The polarization curves were corrected for the ohmic potential drop (iR) losses, where R is the series resistance of the electrochemical cell as determined by electrochemical impedance spectroscopy (EIS) measurements. EIS measurements were performed in the range between 200 kHz and 50 mHz using an AC amplitude of 25 mV. The EIS measurements for obtaining the charge- transfer resistance (R_ct) values were performed at various potentials between 10 and -250 mV by sweeping the frequency from 200 kHz to 100 mHz using an AC amplitude of 10 mV. In order to estimate the electrochemically active surface area (ECSA) of the samples, the double-layer capacitance (C_di) of the samples was determined by cyclic voltammetry (CV) in the potential range of 0.10— 0.30 V versus RHE at scan rates between 20— 100 mV/s.

It is to be understood that the electrocatalyst for hydrogen evolution reaction is not limited to the specific embodiments described above, but encompasses any and all embodiments within the scope of the generic language of the following claims enabled by the embodiments described herein, or otherwise shown in the drawings or described above in terms sufficient to enable one of ordinary skill in the art to make and use the claimed electrocatalyst for hydrogen evolution reaction.

Claims

CLAIMS We claim:

1. An electrocatalyst for hydrogen evolution reaction, comprising:

a carbon fiber paper substrate; and

a plurality of nanosheets of M0S₂ on the carbon fiber substrate, the

nanosheets having a plurality of catalytically active edge sites along basal planes thereof.

2. The electrocatalyst for hydrogen evolution reaction according to claim 1, wherein the basal planes comprise faceted edges.

3. The electrocatalyst for hydrogen evolution reaction according to claim 2, wherein the basal planes comprise collapsed edges.

4. The electrocatalyst for hydrogen evolution reaction according to claim 1, wherein the nanosheets further comprise spherical fullerene nanoclusters.

5. The electrocatalyst for hydrogen evolution reaction according to claim 4, wherein the spherical fullerene nanoclusters are about 7 μηι in diameter.

6. The electrocatalyst for hydrogen evolution reaction according to claim 5, wherein the spherical fullerene nanoclusters cover about 2% of the total surface area of carbon fibers of the carbon fiber paper substrate.

7. The electrocatalyst for hydrogen evolution reaction according to claim 1, wherein the nanosheets are dispersed across an area of about 2 cm² of the carbon fiber paper substrate.

8. A method of making an electrocatalyst for hydrogen evolution reaction, comprising the steps of:

depositing nanosheets of M0O₃ onto a carbon fiber paper by chemical vapor deposition;

reducing the nanosheets of M0O₃ to nanosheets of M0O₂ by reaction with sublimed sulfur; and

sulfiding the nanosheets of M0O₂ to form nanosheets of M0S₂ integrated with the carbon fiber paper substrate, the nanosheets of M0S₂ integrated with the carbon fiber paper substrate providing the electrocatalyst for hydrogen evolution reaction.

9. The method of making an electrocatalyst for hydrogen evolution reaction according to claim 8, wherein the nanosheets of M0O₃ deposited onto the carbon fiber paper are about 1-2 μιη in lateral dimensions.

10. The method of making an electrocatalyst for hydrogen evolution reaction according to claim 8, wherein the reaction with sublimed sulfur occurs at a temperature of about 400°C.

11. The method of making an electrocatalyst for hydrogen evolution reaction according to claim 8, wherein the sulfiding of the nanosheets of M0O₂ occurs at a temperature of about 850°C.

12. The method of making an electrocatalyst for hydrogen evolution reaction according to claim 8, wherein the nanosheets of M0S₂ comprise a plurality of catalytically active edge sites along basal planes thereof.

13. The method of making an electrocatalyst for hydrogen evolution reaction according to claim 8, further comprising: immersing the nanosheets of M0S₂ integrated with the carbon fiber paper substrate in a solution of spherical fullerenes (nC₆o); and annealing the nanosheets of M0S₂ integrated with the carbon fiber paper substrate after immersion in the fullerenes.

14. An electrocatalyst for hydrogen evolution reaction prepared according to the method of claim 8.

15. A method of making an electrocatalyst for hydrogen evolution reaction, comprising the steps of:

depositing nanosheets of M0O₃ onto carbon fiber paper by chemical vapor deposition; reducing the nanosheets of M0O₃ to nanosheets of M0O₂ by reaction with sublimed sulfur;

sulfiding the nanosheets of M0O₂ to form nanosheets of M0S₂ integrated with the carbon fiber paper substrate, the nanosheets of M0S₂ integrated with the carbon fiber paper substrate providing the electrocatalyst for hydrogen evolution reaction; immersing the nanosheets of M0S₂ integrated with the carbon fiber paper substrate in a solution of spherical fullerenes (nC₆o); and annealing the nanosheets of M0S₂ integrated with the carbon fiber paper substrate after immersion in the fullerenes.

16. The method of making an electrocatalyst for hydrogen evolution reaction, according to claim 15, wherein the nanosheets of M0O₃ deposited onto the carbon fiber paper are about 1-2 μιη in lateral dimensions.

17. The method of making an electrocatalyst for hydrogen evolution reaction, according to claim 15, wherein the reaction with sublimed sulfur occurs at a temperature of about 400°C.

18. The method of making an electrocatalyst for hydrogen evolution reaction according to claim 15, wherein the sulfiding of the nanosheets of M0O₂ occurs at a temperature of about 850°C.

19. The method of making an electrocatalyst for hydrogen evolution reaction according to claim 15, wherein the nanosheets of M0S₂ comprise a plurality of catalytically active edge sites along basal planes thereof.

20. An electrocatalyst for hydrogen evolution reaction prepared according to the method of claim 15.