NiPS3 Nanosheet–Graphene Composites as Highly Efficient

Jun 14, 2018 - Developing new electrocatalysts is essentially important for efficient water splitting to produce hydrogen. Two-dimensional (2D) materi...
1 downloads 0 Views 8MB Size
Download PDF
www.acsnano.org

NiPS3 Nanosheet−Graphene Composites as Highly Efficient Electrocatalysts for Oxygen Evolution Reaction Sen Xue,†,‡,∥ Long Chen,†,§,∥ Zhibo Liu,† Hui-Ming Cheng,†,⊥ and Wencai Ren*,†

Downloaded via UNIV OF TOLEDO on June 14, 2018 at 23:01:27 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, 72 Wenhua Road, Shenyang 110016, P.R. China ‡ School of Materials Science and Engineering, Northeastern University, 3 Wenhua Road, Shenyang 110819, P.R. China § University of Chinese Academy of Sciences, 19 A Yuquan Road, Shijingshan District, Beijing 100049, P.R. China ⊥ Tsinghua−Berkeley Shenzhen Institute (TBSI), Tsinghua University, 1001 Xueyuan Road, Shenzhen 518055, P.R. China S Supporting Information *

ABSTRACT: Developing new electrocatalysts is essentially important for efficient water splitting to produce hydrogen. Two-dimensional (2D) materials provide great potential for high-performance electrocatalysts because of their high specific surface area, abundant active edges, and tunable electronic structure. Here, we report few-layer NiPS3 nanosheet− graphene composites for high-performance electrocatalysts for oxygen evolution reaction (OER). The pure NiPS3 nanosheets show an overpotential of 343 mV for a current density of 10 mA cm−2, which is comparable to that for IrO2 and RuO2 catalysts. More importantly, the NiPS3 nanosheet−graphene composites show significantly improved OER activity due to the synergistic effect. The optimized composite shows a very low overpotential of 294 mV for a current density of 10 mA cm−2, 351 mV for a current density of 100 mA cm−2, a small Tafel slope of 42.6 mV dec−1, and excellent stability. These overall performances are far better than those of the reported 2D materials and even better than those of many traditional materials even at a much lower mass loading of NiPS3. KEYWORDS: NiPS3 nanosheets, graphene, composites, electrocatalysts, oxygen evolution reaction high price, and poor durability.5,6 Therefore, it is urgent to develop new OER electrocatalysts with high catalytic activity, good durability, and low cost for practical applications. Recently, numerous efforts have been made on transitionmetal-based OER electrocatalysts such as transition metal oxides,7−11 sulfides,12,13 nitrides,14 carbides,15,16 phosphides,17−21 and transition metal@nitrogen/carbon materials.22,23 These materials usually show catalytic performance toward OER comparable to or even better than that of RuO2

A

s a promising renewable energy source, hydrogen (H2) has drawn great attention in replacing traditional fossil fuel.1 Electrocatalytic water splitting is one of the most efficient and cleanest methods to produce H2, which consists of two half reactions, hydrogen evolution reaction (HER) and oxygen evolution reaction (OER).2,3 Generally, the kinetics of a multistep proton-coupled electron transfer process in OER is more sluggish than the two-electron transfer process in HER. Therefore, OER has been regarded as the rate-limiting step in the water splitting and usually needs the assistance of electrocatalysts to facilitate the reaction rate.3 Noble metal oxides such as RuO2 and IrO2 are known as the most active OER electrocatalysts.4,5 However, they suffer from scarcity, © XXXX American Chemical Society

Received: December 26, 2017 Accepted: June 12, 2018

A

DOI: 10.1021/acsnano.7b09146 ACS Nano XXXX, XXX, XXX−XXX

Article

Cite This: ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 1. Preparation and characterization of NiPS3 nanosheets. (a) Photograph and (b) SEM image of bulk NiPS3 crystals. (c) Ultrathin NiPS3 nanosheets dispersed in water. (d) AFM image and (e) corresponding zoomed-in image of the area marked in (d) of NiPS3 nanosheets. (f) Thickness and (g) size distribution of NiPS3 nanosheets counted from (e). (h) XRD patterns and (i) Raman spectra of bulk NiPS3 crystals and NiPS3 nanosheets.

synthesized simply by tip sonication in water, free from surfactant. The pure NiPS3 nanosheets show an overpotential of 343 mV for a current density of 10 mA cm−2 for OER electrocatalysis, which is comparable to that of the IrO2 and RuO2 catalysts. More importantly, the NiPS3-G composites show significantly improved OER activity due to the synergistic effect. The optimized composite shows a very low overpotential of 294 mV for a current density of 10 mA cm−2, 351 mV for a current density of 100 mA cm−2, a small Tafel slope of 42.6 mV dec−1, and excellent stability. These overall performances are far better than those of the reported 2D materials and even better than those of many traditional materials even at a much lower mass loading of NiPS3 nanosheets.

or IrO2. Compared to bulk materials, their two-dimensional (2D) counterparts have much higher specific surface area, more active edges, tunable electronic structure depending on the number of layers, and easiness of functionalization and, therefore, are more favorable for catalysis applications. For example, it has been demonstrated that defect-rich MoS2 ultrathin nanosheets with additional active edge sites have remarkbly improved HER catalytic activity compared to that of bulk MoS2,24 and adding every additional layer leads to a decrease in HER catalytic activity by a factor of ∼4.47.25 Layer-structured transition metal phosphorus trichalcogenides, MPX3 (M = Cr, Mn, Fe, Co, Ni, Zn or Cr; X = S or Se), have drawn great attention toward electrocatalysis. It has been reported that both NiPS3 and FePS3 nanosheets are active toward HER electrocatalysis in a wide pH range from 1 to 14.26,27 Moreover, the HER kinetics of NiPS3 nanosheets could be remarkably improved by doping with Co, showing catalytic activity very similar to that of the Pt/C catalyst.28 Very recent studies show that NiPS3 nanosheets have relatively low onset potential (1.48 V) and good stability in 0.1 M KOH solution toward OER.29 However, the nanosheets are absorbed with many CTAB surfactants during the exfoliation process. This surfactant contamination, together with the intrinsically low electrical conductivity of semiconducting NiPS3 crystals (∼1 × 10−7 S m−1),30 leads to a slow reaction rate with a relatively high Tafel slope (80 mV dec−1).29 Here, we report NiPS3 nanosheet−graphene (NiPS3-G) composites as high-performance OER electrocatalysts. Different from the previous report,29 NiPS3 nanosheets are

RESULTS AND DISCUSSION Bulk NiPS3 crystals were prepared by a chemical vapor transport method as reported previously.31 They show metallic luster (Figure 1a) and have size and thickness of ∼3 mm and 50 μm, respectively (Figure 1b and Figure S1a). Energydispersive spectrometry (EDS) (Figure S1b) indicates that the atomic ratio of Ni, P, and S is nearly 1:1:3, which agrees well with the stoichiometric ratio of NiPS3 crystals. Different from the previous report,29 we exfoliated these crystals to ultrathin NiPS3 nanosheets simply by tip sonication in water followed by centrifugation to remove the thick flakes (see Methods section). As a result, the NiPS3 nanosheets obtained are very clean without surfactant contaminations, which ensures full exposure of the active sites. Interestingly, the NiPS3 nanosheets B

DOI: 10.1021/acsnano.7b09146 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

image of a NiPS3 nanosheet, and Figure 2d−f shows the corresponding energy-dispersive X-ray (EDX) elemental mapping images of Ni−K, P−K, and S−K, respectively. It can be seen that Ni, P, and S elements are uniformly distributed throughout the whole nanosheet. High-resolution TEM (HRTEM) image shows clean and intact lattice fringes (Figure 2g), confirming that the NiPS3 nanosheets are very clean and maintain high quality after exfoliation. Figure S3a shows the atomic model of NiPS3, and Figure S3b presents the in-plane unit cell of NiPS3, of which the lattice constants are a = 0.581 nm and b = 1.007 nm (PDF #33-0952). Note that the measured lattice constants of a and b are 0.583 and 1.032 nm, respectively, which match well with that of NiPS3 crystal.34 As mentioned above, a semiconducting NiPS3 crystal intrinsically has a low electrical conductivity. Moreover, the NiPS3 nanosheets tend to easily aggregate after being dropped on the substrate (Figure 3d and Figure S1c), which severely reduces the amount of active sites. In order to overcome these problems, we prepared NiPS3-G composites by simply mixing the NiPS3 nanosheets with graphene sheets with different mass ratios in surfactant-free solution followed by drying. The graphene sheets have a thickness no more than 10 layers and lateral size mostly in the range of 1−5 μm (Figure 3a,b). More importantly, they have very high quality, showing a very sharp XRD (002) peak (Figure S4), a large Raman intensity ratio of G peak to D peak (IG/ID > 10, Figure 3c), and very high electrical conductivity greater than 1 × 105 S m−1.35 In the composites, the small NiPS3 nanosheets (Figure 3d,e) are attracted and uniformly attached on the basal planes of highquality large graphene sheets through van der Waals interaction because of the their 2D flexible characteristics, forming a face-to-face stacked structure (Figure 3g,h and Figure S5). On the one hand, this structure ensures a good contact between NiPS3 nanosheets and graphene sheets for electron transfer. As a result, the NiPS3-G-1:1 composite shows a dramatically increased electrical conductivity of ∼1.3 × 104 S m−1, which is about 11 orders of magnitude higher than that of the pure NiPS3 crystal (∼1 × 10−7 S m−1).36 On the other hand, this structure significantly prevents the aggregation of NiPS3 nanosheets and consequently ensures greatly increased exposed active sites, as shown below. Therefore, the NiPS3-G composites are expected to have excellent OER electrocatalysis performance. We then measured the OER performance of graphene, assynthesized NiPS3 nanosheets, and NiPS3-G composites in different mass ratios in 1 M KOH using a rotating disk electrode (RDE) with a total mass loading of 0.2 mg cm−2. The linear sweep voltammetry (LSV) in Figure 4a shows that pure graphene has very poor electrocatalytic activity toward OER. The as-synthesized NiPS3 nanosheets (NiPS3-G-1:0) exhibit good OER activity with an onset potential at 1.48 V and an overpotential at 343 mV to generate 10 mA cm−2, which is comparable to that of the reported IrO2 (320 mV)17 and RuO2 (366 mV).37 As expected, the NiPS3-G composites show greatly improved catalytic performance. Compared to pure NiPS3 nanosheets, for instance, after compositing with graphene with a mass ratio of 1:1, the onset potential reduces from 1.48 to 1.46 V, and the overpotential decreases from 343 to 300 mV for a current density of 10 mA cm−2. Moreover, the Tafel slope dramatically decreases from 58.9 to 42.4 mV dec−1 (Figure 4b and Table S1). We also compared the OER performance of NiPS3-G composites that were made by NiPS3 nanosheets with different thicknesses. As shown in Figure S6,

can be uniformly dispersed in water for months (Figure 1c). The measured ζ-potential of the NiPS3 nanosheets in water is −38.6 mV (Figure S2), which explains the good stability of the dispersion. Extensive atomic force microscopy (AFM) measurements show that the average thickness of the nanosheets is ∼3.5 nm, corresponding to ∼5 layers, and the average size is ∼140 nm (Figure 1d−g). Compared to the bulk crystals, the dramatically increased edge sites and decreased thickness of the nanosheets are expected to significantly improve the catalytic activity of NiPS3 materials. We characterized the structure of the bulk NiPS3 crystals and exfoliated NiPS3 nanosheets by X-ray diffraction (XRD) and Raman spectroscopy. As shown in Figure 1h, both of their XRD spectra show intense sign of NiPS3 (PDF #33−0952) with strong and sharp (001), (002), and (004) diffraction peaks, indicating their high quality. For NiPS3 nanosheets, the full width at half-maximum (fwhm) of these peaks increases 0.16, 0.14, and 0.19°, respectively, which could be explained by their smaller size compared to the bulk NiPS3. As shown in Figure 1i, the Raman peaks in the range of 200−700 cm−1 are attributed to the vibrational modes of PS3, the 434 cm−1 peak to the P−P bond vibrational mode, and the 135 cm−1 peak to the Ni2+ metal ions.29,32,33 Due to the relatively large thickness of the nanosheets (average of 5 layers), both the peak position and width of the Raman peaks of bulk crystals and nanosheets remain almost the same (Figure 1i). The unchanged Raman features also confirm that the nanosheets still maintain high crystallinity after exfoliation. We further used transmission electron microscopy (TEM) to study the atomic-level structure of the NiPS3 nanosheets. Figure 2a is the bright-field TEM image of one NiPS3 nanosheet, which clearly shows its layered structure. The clear and sharp selected area electron diffraction (SAED) pattern along the [001] axis in Figure 2b indicates its high crystallinity. Figure 2c shows the scanning TEM (STEM)

Figure 2. TEM characterization of NiPS3 nanosheets. (a) Brightfield TEM image of one NiPS3 nanosheet. (b) Corresponding SAED patterns of the NiPS3 nanosheet in (a). (c) STEM image of a NiPS3 nanosheet. EDX elemental mappings of (d) Ni−K, (e) P−K, and (f) S−K across the NiPS3 nanosheet in (c). (g) HRTEM image of the NiPS3 nanosheet. C

DOI: 10.1021/acsnano.7b09146 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 3. Characterization of graphene sheets, NiPS3 nanosheets, and NiPS3-G-1:1 composite. (a) SEM image, (b) TEM image, and (c) Raman spectrum of the graphene sheets, indicating that they have a few layers with high quality. (d) SEM image, (e) TEM image, and (f) Raman spectrum of the NiPS3 nanosheets, showing that they are a few hundred nanometers in size with 2D nature. (g) SEM image, (h) TEM image, and (i) Raman spectrum of the NiPS3-G-1:1 composite, showing that the small NiPS3 nanosheets are well attached on the surface of larger graphene sheets.

cm−2). As shown in Figure 4e,f, the NiPS3-G-1:1 composite with a NiPS3 nanosheet loading of 0.2 mg cm−2 also has excellent stability and durability, showing very small change in potential under a constant current density of 10 mA cm−2 for 30 h and in the LSV curves after 1000 continuous cyclic voltammogram (CV) tests. It is important to note that an oxidation peak at ∼1.36 V appears during the OER process (Figure 4a,c), which is the sign of the oxidation of Ni2+ to Ni3+, indicating that nickel oxides or hydroxides were in situ generated.17,29,44,45 To confirm this, the structure of NiPS3 nanosheets before and after the OER test was characterized by X-ray photoelectron spectroscopy (XPS), Raman spectroscopy, and TEM. As shown in Figure S8, the overall survey XPS spectrum before the OER test confirms that the sample contains Ni, P, and S elements. The Ni 2p XPS spectrum (Figure 5a) shows a main peak at 854.6 eV for the state of Ni 2p3/2, with two satellite peaks at 860.0 and 864.8 eV, and another main peak at 872.3 eV for the state of Ni 2p1/2, with satellite peaks at 876.9 and 882.2 eV. The P 2p1/2 and 2p3/2 are located at 131.4 and 132.4 eV, respectively (Figure 5b). The S 2p XPS spectrum (Figure 5c) shows two peaks at 161.9 eV (2p3/2) and 163.2 eV (2p1/2). Note that the separation between Ni 2p3/2 and S 2p3/2 is 692.7 eV, suggesting that S is octahedrally coordinated around the Ni atom in an ionic structure.29 After the OER test, the Ni 2p XPS peak shifts to higher binding energy, where two main peaks at 855.9 eV (2p3/2) and 873.8 eV (2p1/2) with two satellite peaks at 861.7 and 879.9 eV are related to the Ni3+ oxidation state for

the above investigated samples made by NiPS3 nanosheets with moderate thickness (average of 5 layers) show the best OER performance among the three composites investigated. In order to further improve the catalytic performance, NiPS3-G-1:1 composites with different NiPS3 nanosheet loadings were also studied. As shown in Figure 4c,d and Table S1, when the NiPS3 nanosheet loading is increased, the OER catalytic activity of NiPS3-G-1:1 sample first increases and then decreases. The NiPS3-G-1:1 sample with a NiPS3 nanosheet loading of 0.2 mg cm−2 needs only 294 mV to achieve an OER current density of 10 mA cm−2 (Tafel slope is 42.6 mV dec−1), which is better than that of the reported 2D materials-based OER electrocatalysts, such as BP-CNTs (∼320 mV, 59.84 mV dec−1),38 holey Ni(OH)2 nanosheets (∼335 mV, 65 mV dec−1),39 Co−Mn LDH (324 mV, 43 mV dec−1),40 and Ni3S2 film (400 mV, 51 mV dec−1),41 and comparable to that of Co−Bi NS/G (290 mV, 53 mV dec−1)42 and Ni2P nanoparticles (290 mV, 59 mV dec−1).17 Moreover, only 351 mV is needed to achieve a high current density of 100 mA cm−2. In contrast, the pure NiPS3 nanosheets cannot reach this value even at a high overpotential of 487 mV. As shown in Table S1, this value is better than that of most reported highperformance catalysts such as NF-Ni3Se2/Ni (353 mV),12 CoP-MNA/Ni (∼363 mV),18 and IrO2/Ni (∼460 mV),18 except for that of Fe(PO3)2/Ni (∼221 mV).43 Moreover, it is worth noting that the mass loading of NiPS3 nanosheets (less than 1 mg cm−2) in our composites is much lower than those of the catalysts mentioned above (usually larger than 5 mg D

DOI: 10.1021/acsnano.7b09146 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 4. Electrocatalytic performance of the NiPS3-G composites. (a) LSV curves and (b) Tafel plots of the NiPS3-G composites at different mass ratios. (c) LSV curves and (d) Tafel plots of the NiPS3-G-1:1 composites with different NiPS3 mass loadings. (e) Chronopotentiometric curve of the NiPS3-G-1:1 composite (NiPS3-0.2 mg cm−2) at a constant current density of 10 mA cm−2 for 30 h. (f) LSV curves of the NiPS3-G-1:1 composite (NiPS3-0.2 mg cm−2) before and after 1000 CV tests.

NiOOH (Figure 5a).46 For the P 2p and S 2p counterparts, the peak intensity becomes very weak. The higher binding energy peak at 133.3 eV for P 2p is related to the oxidized phosphate,44 and the peak at 168.5 eV for S 2p is related to the −SO3H in the Nafion residue.47 In addition to these peaks, there are still weak P 2p and S 2p peaks that are related to pure NiPS3. Similarly, two characteristic Raman peaks at 474 and 554 cm−1 that are attributed to NiOOH are observed in the samples after OER, in addition to the characteristic Raman peaks of NiPS3 (129, 176, 254, 383, and 588 cm−1) (Figure 5d).48,49 These results indicate the appearance of NiOOH in the samples after OER in addition to the NiPS3. HRTEM images give clear information on the detailed structure of the catalyst after OER. As shown in Figure 5e, except for the lattice fringe of 0.209 nm that corresponds to the (221) planes of NiPS3, lattice fringes of 0.241 and 0.235 nm corresponding to the (101) and (102) planes of NiOOH are also observed.50 This confirms the formation of NiOOH-NiPS3 in-plane heterostructures. EDX elemental mappings (Figure 5f−j) show that Ni and O distributions are closely correlated, whereas P and S distributions are closely correlated, giving more evidence of NiOOH-NiPS3 heterostructures. These NiOOH-NiPS3 heterostructures are believed to serve as the main active sites for OER electrocatalysis in our composites.29

Nonmetal elements such as P and S usually do not contribute to the OER performance directly. However, they can affect the local electronic structure of Ni due to their strong electronegativity, which facilitates the oxidation of Ni2+ to Ni3+. For example, it has been reported that P can induce steric strain into the lattice of NiOOH, which significantly promotes its OER performance.51 The CV curves show that the oxidation process is quasi-reversible for NiPS3 nanosheets containing catalysts (Figure S7a). In other nickel-based electrocatalysts,52,53 it has been reported that NiOOH can be easily reduced to Ni(OH)2 in alkaline solution. Therefore, we suggest that the product after reduction process should be Ni(OH)2. To further understand the role of graphene in improving the OER performance of NiPS3 nanosheets, the electrochemically active surface areas (ECSA) of pure NiPS3 nanosheets and NiPS3-G-1:1 (NiPS3-0.2 mg cm−2) were evaluated by the electrochemical double-layer capacitance (Cdl), which was obtained through CV tests at different scan rates (Figure S7b− d). As shown in Figure 6a, the Cdl value of NiPS3-G-1:1 (0.454 mF cm−2) is ∼3.2 times larger than that of NiPS3 nanosheets (0.141 mF cm−2). In addition, the electrochemical active site numbers and turn over frequency (TOF) values were calculated by integrating the Ni3+/2+ reduction wave (Figure S7a), assuming 1 e− per Ni atom.53 As shown in Table S2, the E

DOI: 10.1021/acsnano.7b09146 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 5. Mechanism analysis of the electrocatalytic process for NiPS3 nanosheets. XPS spectra of (a) Ni 2p, (b) P 2p, and (c) S 2p for the NiPS3 nanosheets before and after the OER test at 1.5 V for 1 h. The signals of NiPS3 nanosheets after the OER test in (b,c) are multiplied 10 times for clarity. (d) Raman spectra of the NiPS3 nanosheets before and after the OER test at 1.5 V for 1 h. (e) HRTEM image and (f) STEM image of NiPS3 nanosheets after the OER process. EDX elemental mappings of (g) O−K, (h) Ni−K, (i) P−K, and (j) S−K across the NiPS3 nanosheet in (f).

electrochemical active site number and TOF value of NiPS3-G1:0 (NiPS3-0.2 mg cm−2) are 4.37 × 10−8 mol and 0.0249 s−1, respectively, whereas those of NiPS3-G-1:1 (NiPS3-0.1 mg cm−2) are 9.10 × 10−8 mol and 0.0564 s−1, respectively, even with lower mass loading of NiPS3 nanosheets.The electrochemical impedance spectroscopy (EIS) analyses were also carried out on pure NiPS3 nanosheets and NiPS3-G-1:1 in 1 M KOH at an overpotential of 314 mV. As shown in Figure 6b, the Nyquist plots of the electrocatalysts exhibit a twofrequency feature. In general, the resistance at high frequency is associated with the solution resistance (Rs), and the

resistance at low frequency is associated with the charge transfer process of OER.54 It is worth noting that the charge transfer resistance (Rct) of the NiPS3-G-1:1 (NiPS3-0.2 mg cm−2) sample (35 Ω) is much smaller than that of the pure NiPS3 nanosheet sample (191 Ω), indicating that the incorporation of highly conductive graphene sheets leads to faster kinetics of the OER process. Based on the above studies, the excellent OER catalytic performances of NiPS3-G composites can be explained from the following four aspects. First, the nanometer-scaled lateral size and atomically thin thickness of NiPS3 nanosheets enables F

DOI: 10.1021/acsnano.7b09146 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 6. Analysis on the role of graphene in improving the OER performance NiPS3-G composites. (a) Capacitive current (Δj = ja − jc) at 0.934 V versus the scan rates for graphene, NiPS3-G-1:0, and NiPS3-G-1:1 (NiPS3-0.2 mg cm−2). (b) Nyquist plots of NiPS3-G-1:0 and NiPS3G-1:1 (NiPS3-0.2 mg cm−2) at an overpotential of 314 mV. ampule was horizontally placed in the middle of a tube furnace (Lindberg Blue M (TF55035KC-1)) and heated to 700 °C within 1 h. After the mixture was reacted for 5 days, the furnace was turned off to cool the ampule to room temperature. Finally, black colored bulk NiPS3 crystals formed at the cold end were picked out and washed by acetone to remove the residual iodine. Preparation of NiPS3-G Composites. The obtained bulk NiPS3 crystals (200 mg) were exfoliated by tip sonication (Scientz-IID ultrasonic homogenizer, power = 285 W) in deionized water (30 mL) without any surfactant for 6 h, followed by centrifugation at different speeds to decrease the size of the NiPS3 nanosheets. In a typical procedure, the dispersion was first centrifuged twice at 2500 rpm for 10 min and then centrifuged twice at 5000 rpm for 10 min. The supernatant was dehydrated by vacuum filtration and dried at 40 °C for 24 h. Graphene sheets (Deyang Carbonene Technology Co., Ltd.) contain ∼98 atom % C element, and the C/O ratio is larger than 48. NiPS3-G composites (total mass is 2.0 mg) were prepared by mixing the obtained NiPS3 nanosheets with graphene powders in different mass ratios (1:10, 1:5, 1:0, 1:1, 1:0.2, 1:0.1) in 0.96 mL of ethanol mixed with 40 μL of Nafion solution (5 wt %) by sonication for 1 h to obtain uniform dispersions (catalyst inks). Electrochemical Measurements. All of the electrochemical measurements were carried out with a three-electrode system using CHI 760E and Autolab M204 (Metrohm) electrochemical workstations at room temperature. A glassy carbon disk electrode with a diameter of 5 mm was used as the working electrode, Pt foil as the counter electrode, and saturated calomel electrode (SCE) as the reference electrode. Twenty microliters of catalyst ink as mentioned above was drop-coated on the glassy carbon electrode polished with 3 μm alumina paste and dried in air at room temperature. All measured potentials were converted to reversible hydrogen electrode (RHE): ERHE = ESCE + 0.242 V + 0.059 × pH. LSV was employed using RDE (Pine Instrument Company) with a scan rate of 1 mV s−1 at a rotating speed of 1600 rpm in 1 M KOH solution (pH 13.6) saturated with O2 to obtain the polarization curves. Before the OER test, the samples were continuously swept from 1.0 to 1.6 V until a steady voltammogram curve was obtained. All data were presented with IR compensation. To evaluate the ECSA of these catalysts, CVs were obtained from 0.84 to 1.02 V versus RHE with a series of scan rates of 20, 40, 60, 80, and 100 mV s−1. EIS measurements were performed in a frequency range from 100 kHz to 0.01 Hz at an overpotential of 314 mV with 20 mV amplitude. Characterizations. The morphology and structure of the catalysts were characterized by scanning electron microscopy (Nova NanoSEM 430, 15 kV), transmission electron microscopy (FEI Tecnai F30, 300 kV; FEI Tecnai F20, 200 kV; FEI Tecnai T12, 120 kV; FEI Titan Cube Themis G2 300, 300 kV; JEOL JEM 2010, 200 kV), Raman spectroscopy (Jobin Yvon HR800), XRD (Rigaku diffractometer with Cu Kα radiation), and AFM (MultiMode 8). The compositions of the samples were detected by XPS (ESCALAB 250, Al Kα) and Elementar (vario MICRO cube). The ζ-potential of the catalyst was obtained by a Zetasizer Nano instrument (Malvern).

a great number of active edge sites. Second, the in situ formed NiOOH-NiPS3 heterostructures are fully effective to catalyze the OER process due to the synergistic effect between NiOOH and NiPS3 nanosheets.29 Third, the NiPS3 nanosheets are uniformly dispersed on the basal plane of graphene to avoid the heavy aggregation, which greatly improves the electrochemically active surface areas and TOFs. Fourth, due to the good interaction between graphene and NiPS3 nanosheets, the NiPS3-G composites have high electrical conductivity and low electron transfer resistance, which can facilitate charge transfer between the electrode and the catalyst. It has been reported that the OER performance of Co3O4 and CoFe LDH nanosheets can be greatly improved by inducing defects through Ar plasma treatment.14,55 We found that the NiPS3 nanosheets after Ar plasma treatment for 2 min could also be stably dispersed in water for a long time. Moreover, as shown in Figure S9, they show improved OER performance compared to the pristine NiPS3 nanosheets. Therefore, defect engineering would be a promising way to further enhance the OER performance of NiPS3 nanosheets and NiPS3-G composites in the future.

CONCLUSIONS In summary, we report ultrathin NiPS3-G composites as highperformance electrocatalysts. The NiPS3 nanosheets are synthesized by exfoliating bulk NiPS3 crystals in water, free of surfactant contaminations, which are more compatibile to the OER application with overpotential comparable to that of IrO2 and RuO2 catalysts. More importantly, the composites show excellent catalytic performance, far better than that of the reported 2D materials and even better than that of many traditional materials, even at a much lower mass loading of NiPS3 nanosheets than those reported electrocatalysts, due to the greatly increased active surface areas and significantly decreased electron transfer resistance caused by the synergistic effect. Such excellent overall performance highlights the potential application of NiPS3 nanosheets and graphene composites as highly efficient OER electrocatalysts. METHODS Preparation of Bulk NiPS3 Crystals. Nickel powder (634 mg) (Alfa Aesar, 99.8+%, metals basis,