www.acsnano.org
Two-Dimensional MoS2 Confined Co(OH)2 Electrocatalysts for Hydrogen Evolution in Alkaline Electrolytes Yuting Luo,† Xu Li,‡ Xingke Cai,† Xiaolong Zou,† Feiyu Kang,†,‡ Hui-Ming Cheng,†,§ and Bilu Liu*,† †
Shenzhen Geim Graphene Center (SGC), Tsinghua-Berkeley Shenzhen Institute (TBSI), Tsinghua University, Shenzhen 518055, PR China ‡ Graduate School at Shenzhen, Tsinghua University, Shenzhen 518055, PR China § Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, Shenyang, Liaoning 110016, PR China S Supporting Information *
ABSTRACT: The development of abundant and cheap electrocatalysts for the hydrogen evolution reaction (HER) has attracted increasing attention over recent years. However, to achieve low-cost HER electrocatalysis, especially in alkaline media, is still a big challenge due to the sluggish water dissociation kinetics as well as the poor long-term stability of catalysts. In this paper we report the design and synthesis of a two-dimensional (2D) MoS2 confined Co(OH)2 nanoparticle electrocatalyst, which accelerates water dissociation and exhibits good durability in alkaline solutions, leading to significant improvement in HER performance. A two-step method was used to synthesize the electrocatalyst, starting with the lithium intercalation of exfoliated MoS2 nanosheets followed by Co2+ exchange in alkaline media to form MoS2 intercalated with Co(OH)2 nanoparticles (denoted Co-Ex-MoS2), which was fully characterized by spectroscopic studies. Electrochemical tests indicated that the electrocatalyst exhibits superior HER activity and excellent stability, with an onset overpotential and Tafel slope as low as 15 mV and 53 mV dec−1, respectively, which are among the best values reported so far for the Pt-free HER in alkaline media. Furthermore, density functional theory calculations show that the cojoint roles of Co(OH)2 nanoparticles and MoS2 nanosheets result in the excellent activity of the Co-Ex-MoS2 electrocatalyst, and the good stability is attributed to the confinement of the Co(OH)2 nanoparticles. This work provides an imporant strategy for designing HER electrocatalysts in alkaline solutions, and can, in principle, be expanded to other materials besides the Co(OH)2 and MoS2 used here. KEYWORDS: hydrogen evolution reaction, alkaline media, nanoconfinement, two-dimensional materials, MoS2, Co(OH)2, density functional theory
T
Unfortunately, the large-scale use of Pt-based catalysts is hampered by its scarcity and high cost. In recent years, tremendous efforts have been made in exploring efficient and abundant Pt-free catalysts to accelerate the sluggish HER kinetics. Indeed, many durable HER electrocatalysts with good HER activity have been reported, mainly in acidic electrolytes, such as molybdenum disulfide (MoS2),6,9−11 tantalum sulfide (TaS2),12,13 and their heterostructures.7,14−16 It is known that the scalable production of hydrogen is more cost-effective and the half reaction, i.e., the oxygen evolution reaction (OER), on the counter electrode is more efficient in alkaline media than in
he excessive use of fossil fuels has resulted in an increasing energy crisis and severe global pollution, and this has provoked great attention to the development of highly efficient and safe ways of producing renewable energy.1−5 Because hydrogen is a clean and high density energy carrier, the scalable and sustainable production of hydrogen by electrochemical water splitting is considered a promising route to meet renewable energy requirements.6,7 Generally, the hydrogen evolution reaction (HER) is assumed to initiate from the formation of hydrogen intermediates (the Volmer step), followed by the generation of hydrogen molecules by either a recombination step (the Tafel step) or an electrochemical desorption step (the Heyrovsky step). Due to its fast kinetics in these steps and good electrical conductivity, Pt has been acknowledged as the most efficient electrocatalyst for HER.8 © XXXX American Chemical Society
Received: February 5, 2018 Accepted: April 3, 2018
A
DOI: 10.1021/acsnano.8b00942 ACS Nano XXXX, XXX, XXX−XXX
Article
Cite This: ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
Figure 1. Synthesis of electrocatalysts for HER in alkaline media, consisting of Co(OH)2 confined in MoS2 and their Raman characterization. (a) Schematic of the synthesis. (b,c) Raman spectra of the products at different stages during the syntheses of (b) Co-Ex-MoS2 and (c) CoMoS2.
acidic media.17,18 Therefore, the development of Pt-free electrocatalysts with high activity and good stability for HER in alkaline media is very important. In this regard, Geng et al. have developed an efficient electrocatalyst made of threedimensional MoS2 nanostructures on graphene-covered Ni foam, which shows good HER activity in alkaline media with a Tafel slope of ∼98 mV dec−1.19 Moreover, Danilovic et al. have reported that by modifying Ni(OH)2 nanoclusters on a Ni electrode, the current density under an identical overpotential of 0.3 V increased by 4-fold compared to an unmodified Ni electrode in alkaline media, and showed an overpotential of ∼200 mV at 5 mA cm−2.20 Further, Jiang et al. have developed Ni(OH)2/MoS221 and Co3O4/MoS222 composites and heterostructures that show good performance. Nevertheless, despite great effort in recent years, the HER activity of Pt-free catalysts in alkaline media is still low due to the sluggish water dissociation kinetics (the Volmer step). In addition, the longterm stability of electrocatalysts in alkaline media is also of great concern due to their easy aggregation and peeling during the HER process. In recent years, confined catalysis have attracted increased interest because they may show new behavior due to the socalled nanoconfinement effect, which can be further used to tune and improve their performance as has been demonstrated in several reactions.23−30 Among the host materials used to confine catalysts, two-dimensional (2D) materials are of particular importance for the following two reasons. First, as layers of 2D materials are held together by weak interactions such as van der Waals forces, the layer−layer distances (i.e., van der Waals gaps) are widely tunable from tenths to several nanometers, which make them excellent hosts for a large variety of small species such as alkali metal ions, metal nanoparticles, and water molecules.31−35 To date, a few confined electro-
catalysts have been made and used for different reactions.23−25 For example, Chen et al. have prepared an HER catalyst where Pt nanoparticles (∼2 nm) are confined in the van der Waals gaps of bulk MoS2.24 The results show that such confinement not only suppresses the aggregation of the small Pt nanoparticles but also ensures easy transfer of H3O+ inside the gaps, which leads to their having a comparable HER performance to commercial Pt/C in acidic media, but with much lower Pt loading ratio (10 wt % in MoS2 confined Pt versus 40 wt % in Pt/C). In another study, Lei et al. demonstrated that tin quantum sheets confined in few layer graphene exhibit higher activity and stability than bulk tin in the carbon dioxide electroreduction reaction.25 This improved performance has been partially attributed to the graphene shell that protects active tin from oxidation, as well as the easy diffusion of electrolyte and products in the interlayer spaces of graphene. Second, many 2D materials, such as MoS2 and TaS2, are themselves good electrocatalysts for HER6,12,13 and it is therefore possible that by a rational design and choice of 2D nanoconfined catalysts, the host and intercalant will work cojointly to produce catalysts with both high efficiency and good stability in alkaline media, which could solve the longstanding need for high efficiency Pt-free HER electrocatalysts in such media. To design such an electrocatalyst for alkaline media, mass transfer as well as water dissociation both have to be considered. First, it has been reported that MoS2 exhibits a high HER activity in acidic media due to a moderate hydrogen absorption energy at its edges,11 but low activity in alkaline media due to the sluggish water dissociation kinetics.36 Second, Ni and Co species like Ni(OH)2 and Co(OH)2 are widely recognized as excellent water dissociation centers, because they can absorb oxygen, which decreases the energy needed to break B
DOI: 10.1021/acsnano.8b00942 ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
Figure 2. Structural characterization of the Co-Ex-MoS2 samples. (a) AFM image (top) and its height profile (bottom) along the white line. (b) TEM image, (c) HRTEM image and (d) the corresponding FFT pattern. (e,f) HRTEM images of (e) the side view of the (002) planes and (f) an enlarged image. I, II, III, IV, and V in (f) indicate MoS2 (002) planes with obviously different spacings. (g) HADDF STEM image and elemental maps showing the uniform distribution of S, Mo, and Co.
the H−O bond which is beneficial for water dissociation in alkaline media.17,20,37 In addition, we note that the van der Waals gap of MoS2 is large (∼0.6 nm), which allows easy intercalation/deintercalation of small species such as lithium ions.38 On the basis of these considerations, we have designed an electrocatalyst consisting of Co(OH)2 nanoparticles confined in few-layer MoS2 nanosheets, which shows excellent HER performance in alkaline media. The catalyst has a low onset overpotential of ∼15 mV, a small Tafel slope of 53 mV dec−1, and a good durability after 20 h test. These values are much better than those of most Pt-free HER catalysts in alkaline media developed so far, indicating that our rationale for the design and preparation of efficient electrocatalysts is valid. Experimental and theoretical investigations reveal that the reason for this excellent activity is attributed to cojoint effect of the confined Co(OH)2 nanoparticles in accelerating the water dissociation kinetics and of the MoS2 nanosheets in speeding up the hydrogen generation kinetics, and the good stability stems from the confinement of the Co(OH)2 nanoparticles in 2D MoS2 nanosheets to avoid their aggregation or loss during the reaction.
nanosheets (denote Ex-MoS2, Figures S1 and S2), by dispersing the nanosheets in a solution of n-butyllithium (n-BuLi) in hexane, and the sample obtained is denoted Ex-LixMoS2. The second step was to exchange Li+ with Co2+ by adding CoCl2· 6H2O to the Ex-LixMoS2 in an n-methyl-pyrrolidone (NMP) solution. We found that the pH value of the solution increases to 9.88 after this step, which is the result of the reactions between lithium and water that is simultaneously slowly released from CoCl2·6H2O upon heating. As a result, the cobalt species were converted into Co(OH)2 nanoparticles in the alkaline environment and remained in the interlayer spaces of the Ex-MoS2 nanosheets. The synthesized sample was denoted Co-Ex-MoS2. The synthesis process was studied by Raman characterization of the products at different reaction stages (Figure 1b). We started with the 2H-phase MoS2 which has two characteristic Raman peaks at 379 and 405 cm−1, corresponding to the A1g and E2g1 peaks, respectively. After lithium intercalation, the 2H-phase MoS2 was converted into the 1T′ phase, which has three characteristic Raman peaks at 158 (J1), 218 (J2), and 334 cm−1 (J3).39 These results are in accordance with previously report that alkali metals like Li can donate electrons to MoS2, making it negatively charged and convert to 1T′ phase which is stabilized by electrons.39−41 In contract, transitional metals like Co cannot donate electrons and thus cannot induce phase change of MoS2. As a result, after further exchange of Li+ with Co2+, the MoS2 was converted back to the 2H phase, as evidenced by the disappearance of the
RESULTS AND DISCUSSION A two-step method was developed to synthesize electrocatalysts made of Co(OH)2 nanoparticles confined by 2D MoS2 nanosheets (Figure 1a). The first step was to intercalate lithium into the van der Waals gaps of exfoliated MoS2 C
DOI: 10.1021/acsnano.8b00942 ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
Figure 3. Structural and chemical characterization of the Co-Ex-MoS2 samples. (a) XRD pattern compared with that of pristine MoS2 powder. (b−d) XPS spectra of Mo 3d (b), Co 2p (c), and O 1s (d). The blue curve in (c) is the XPS spectrum of pure Co(OH)2, which is used as a reference.
retain electron transport along basal surfaces, which in turn may give good HER performance.9 It can be also seen in the sideview HRTEM images that the spacings of the (002) lattice planes are uniform in Ex-MoS2 nanosheets (0.62 ± 0.02 nm, Figure S2), while there is a wide range of spacings (0.64 ± 0.20 nm) for the Co-Ex-MoS2 nanosheets (Figures 2e and 2f), indicating that intercalation causes expansion and contraction of the interlayer spacings in MoS2. The uniform interlayer spacing of MoS2 may have positive effect on HER activity.9 To provide additional information on the Co-Ex-MoS2, high-angle annular dark-field scanning TEM (HAADF STEM) characterization and energy dispersive spectroscopy (EDS) elemental mapping were performed (Figure 2g). It can be seen that Mo, S, and Co exist in the Co-Ex-MoS2 samples and the Co is uniformly distributed. A comparison of the AFM and STEM characterization of the Co-Ex-MoS2 samples shows that many nanoparticles are observed under STEM, while they are not seen under AFM, revealing that most of the nanoparticles are in the interlayer spaces of the MoS2 (Figure S5). The above AFM, HRTEM, STEM, and EDS analyses suggest that nanoscale Co species have been intercalated into the interlayer spaces of the MoS2 nanosheets in the Co-Ex-MoS2 samples. The next step is to determine the structure and chemical state of Co species intercalated in the MoS2. Figure 3a shows the X-ray diffraction (XRD) pattern of the Co-Ex-MoS2 sample, and the peak at 14.3° can be assigned to the (002) plane of MoS2. For the pristine MoS2 powders, this peak is very sharp and strong due to the good crystallinity and bulk nature. In sharp contrast, in Co-Ex-MoS2 this (002) peak becomes blunt and very weak, because of the thinning of the MoS2 flakes and a reduction in the long-range order, in agreement with the AFM and HRTEM results shown in Figure 2. Note that no obvious peaks originating from Co species were observed, indicating that they are very likely to be amorphous or too weak to be detected. To further study the surface chemistry of the Co-Ex-
J1, J2, and J3 peaks in Co-Ex-MoS2 sample (the blue spectrum in Figure 1b). As a comparison, we intercalated lithium into bulk MoS2 powder, instead of exfoliated MoS2 nanosheets, to make bulk MoS2 confined Co(OH)2 catalysts (denoted Co-MoS2, Figure 1c). We now focus on the structure of the as-synthesized Co-ExMoS2 sample. Figure 2a shows an atomic force microscopy (AFM) image and the corresponding height profile. Statistical results show that these Co-Ex-MoS2 nanosheets have an average thickness and lateral size of 8.2 and 85.5 nm, respectively, indicating that many edge sites may be exposed to the electrolyte in the HER process (Figure S3). As a comparison, the Ex-LixMoS2 flakes have an average thickness of 15.2 nm (Figure S4). Transmission electron microscopy (TEM) confirms the thin sheet-like morphology and small lateral size of Co-Ex-MoS2 (Figure 2b), consistent with the AFM results. High-resolution TEM (HRTEM) image shows that the basal planes of MoS2 in the Co-Ex-MoS2 nanosheets have obviously become disordered, and only short-range order of the nanoparticles was observed (Figures 2c−f). To determine whether this disorder is caused by exfoliation of bulk MoS2 or by lithium intercalation and subsequent Co2+ exchange, we also characterized the Ex-MoS2 nanosheets (without lithium intercalation) by HRTEM. In sharp contrast to the Co-Ex-MoS2 samples, the Ex-MoS2 samples are highly crystalline with a uniform spacing (Figures S1 and S2), indicating that the disorder of Co-Ex-MoS2 samples is mainly due to the lithium intercalation and subsequent Co2+ exchange. A comparison of the fast Fourier transform (FFT) patterns from Co-Ex-MoS2 (Figure 2d) and Ex-MoS2 (Figure S1b) samples also suggests that a certain degree of disorder has been introduced in the Co-Ex-MoS2 sample, as its (002) planes have different spacings. According to previous studies, an optimum degree of disorder in the MoS2 may not only provide more active atoms that are accessible and more active sites, and also D
DOI: 10.1021/acsnano.8b00942 ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
Figure 4. HER electrocatalytic activity of different samples. (a) Polarization curves after iR compensation and (b) Tafel plots of the bulk MoS2, Ex-MoS2, Co-MoS2, Co-Ex-MoS2, and Pt/C in KOH (1.0 M) at a scan rate of 5 mV s−1. (c) Polarization curves of the initial Co-ExMoS2 and after 1000 cyclic voltammetry scans at a scan rate of 50 mV s−1. Inset shows the chronoamperometric responses (i−t) recorded on Co-Ex-MoS2 for 20 h. (d) Comparison of the overpotentials at 10 mA cm−2 and the Tafel slopes of recently reported MoS2-based electrocatalysts.46−54 The yellow shadow denotes ink samples, while the rest are powder samples.
and is beneficial for water dissociation in alkaline media.17,20,44 Thus, an efficient and stable HER performance could be expected from electrocatalysts made of MoS2 nanosheets confined Co(OH)2 nanoparticles. To test our expectation, we conducted electrochemical measurements on the HER activity of the synthesized Co-ExMoS2 samples in alkaline media. For comparison, the HER performance of bulk MoS2, Ex-MoS2, bulk MoS2 confined Co(OH)2 (Co-MoS2), and commercial Pt/C (20 wt % Pt) was also measured under the same conditions, with all measurements carried out using the same catalyst loading of 0.2 mg cm−2. Because Li-MoS2 is unstable in water or air as reported before,10 here we did not test this sample. Figure 4a shows representative polarization curves of different samples in a KOH (1.0 M) solution. Notice that the Co-Ex-MoS 2 electrocatalyst shows a much smaller onset overpotential (η) of around 15 mV versus a reversible hydrogen electrode (RHE) than that of bulk MoS2 (∼230 mV), suggesting the superior HER activity of Co-Ex-MoS2. Interestingly, we found that after intercalation of Co(OH)2 nanoparticles, even the sluggish bulk MoS2 exhibited an impressive improvement in the HER activity (i.e., MoS2 vs Co-MoS2), in sharp contrast to the minor increase obtained using bulk MoS2 after exfoliation (i.e., MoS2 vs ExMoS2). These comparisons suggest that the nanoconfinement of Co(OH)2 in MoS2 plays a very important role in the improvement of the HER performance. Moreover, the Co-ExMoS2 exhibits the lowest overpotential of 89 mV at a current density (j) of 10 mA cm−2, while the Co-MoS2 and Ex-MoS2 have values of 179 mV and 227 mV, respectively. Due to its lack of active sites, bulk MoS2 has a current density smaller than 10 mA cm−2 even at overpotentials up to 400 mV. As a
MoS2, high-resolution X-ray photoelectron spectroscopy (XPS) measurements were carried out. As shown in Figure 3b, two major peaks located at 229.6 and 232.7 eV were observed, which arise from Mo 3d5/2 and Mo 3d3/2 in Mo(IV), suggesting the dominance of MoS2 in the sample. A weak peak at 235.7 eV, which originates from Mo(VI), is also observed, suggesting slight oxidation of the MoS2. Because the detection depth of XPS and the thickness of the Co-Ex-MoS2 sample are of the same order, i.e., several nanometers, the chemical state of the Co intercalant can be detected and analyzed by XPS. As shown in Figure 3c, the Co 2p spectrum shows two characteristic peaks with binding energies of 781.6 and 797.5 eV, corresponding to Co 2p3/2 and Co 2p1/2. The positions and distance between these two peaks (Δ = 15.9 eV) indicate that Co exists in the Co(OH)2 form, consistent with reports in the literature.42,43 We also collected an XPS spectrum from a pure Co(OH)2 reference sample (the blue spectrum in Figure 3c), and found that it shows exactly the same peak positions and peak distance as do the Co-Ex-MoS2 sample. The XPS results show that the majority of Co species are in Co(OH)2 form. In addition, the O 1s spectrum of Co-Ex-MoS2 shows three deconvoluted peaks centered at 532.7, 531.6, and 530.5 eV (Figure 3d). The main peak, located at 531.6 eV, is assigned to oxygen ions (Olattice) in Co(OH)2,42,43 and peaks at 532.7 and 530.5 eV are assigned to absorbed water and oxide species.42,43 The selective area electron diffraction (SAED) pattern of the Co-Ex-MoS2 also confirms the Co species are Co(OH)2 (Figure S6). All the above results confirm that the intercalated Co species are Co(OH)2. As reported in several previous studies, Co(OH)2 nanoclusters have the ability to absorb oxygen, which can decrease the energy to break H−O bonds E
DOI: 10.1021/acsnano.8b00942 ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
MoS2 has superior HER activity and stability in alkaline solutions (Table 1). Figure 4d summarizes the HER kinetic data of several MoS2-based electrocatalysts in alkaline solutions reported recently. It should be noted that most of these reports used nanostructured films or foams with larger surface areas and loading amounts than the Co-Ex-MoS2 reported here. In a few other reports (marked by pink shadow), the catalysts are in ink form with a loading weight comparable to the Co-Ex-MoS2 report here. By inspecting Figure 4d and Table S1, it is apparent that the Co-Ex-MoS2 samples show one of the best performances of these electrocatalysts in alkaline media, especially with an impressive low Tafel slope of 53 mV dec−1. The low Tafel slope suggests that the HER rate of Co-Ex-MoS2 will increase rapidly with increasing overpotential, leading to a competitive advantage in practical applications. First-principles density functional theory (DFT) calculations have been performed to gain a fundamental understanding of the high electrocatalytic activity of Co-Ex-MoS2 in alkaline media. The relaxed structures of MoS2 (top) and Co(OH)2 (down) for OH* + H* and H* adsorption are shown in Figures 5a and 5b, respectively. Figure 5c shows the free energy diagrams of the HER reactions on the optimized edges of MoS2, Co(OH)2, and Co-Ex-MoS2 samples. The binding free energies were calculated to determine the energy barriers of water dissociation for generating H* intermediates (the Volmer step), of the desorption of *OH intermediates from surface of catalysts, and of hydrogen generation (the Heyrovsky step in this case). The calculated energy barrier for water dissociation on MoS2 is rather high (1.15 eV), suggesting that it is the ratelimiting step. In contrast, the barrier for water dissociation on Co(OH)2 is much lower at 0.02 eV. However, the adsorption energy of H on Co(OH)2 is relatively high, which makes it tend to transfer to MoS2 with a lower energy, as indicated by the blue arrow in Figure 5c. The DFT calculations show that Co(OH)2 and MoS2 play cojoint roles in increasing the HER activity of the Co-Ex-MoS2 electrocatalyst. In addition to the improved activity, the Co(OH)2 “promoters” confined in the 2D MoS2 nanosheets are protected from aggregation or loss that could happen in free space, leading to the improved cycling stability of the Co-Ex-MoS2, as illustrated in Figure 5d.
comparison, we also tested the HER activity of pure Co(OH)2 nanosheets, which shows an overpotential of 240 mV at a current density of 10 mA cm−2, comparable to previous works (Figures S7 and S8). To further analyze the HER reaction pathways, Tafel plots of different samples are shown (Figure 4b). They show that the linear region of the Co-Ex-MoS2 sample has a slope of 53 mV dec−1, close to the theoretical value of 40 mV dec−1 at which the Volmer−Heyrovsky reaction pathway is operative with the electrochemical desorption of hydrogen being the rate limiting step.45 In contrast, the Tafel slopes of Co-MoS2, Ex-MoS2, and bulk MoS2 increase in turn from 62, 91, to 104 mV dec−1. For Tafel slopes close to 120 mV dec−1, the primary discharge step becomes the rate-determining step,45 suggesting that the reaction kinetics of bulk MoS2 is different with Co-Ex-MoS2. The lower Tafel slope of Co-ExMoS2 than other samples indicates a good HER kinetics, further supported by Nyquist plots, which shows that the CoEx-MoS2 has better electron transfer ability compared to other samples (Figure S9). Another important parameter to evaluate the HER activity of a sample is the exchange current density (j0). By extrapolating the Tafel plots, the j0 values were obtained (Table 1). Again, the Co-Ex-MoS 2 shows a Table 1. Summary of the Electrochemical Parameters of the Co-Ex-MoS2 Sample and Other Samples Studied in This Work onset overpotential (mV vs RHE) Co-ExMoS2 CoMoS2 ExMoS2 bulk MoS2
overpotential @ 10 mA cm−2 Tafel slope (mV vs RHE) (mV dec−1)
exchange current density (μA cm−2)
∼15
89
53
72.9
∼75
179
62
27.0
∼190
227
91
3.2
∼230
>400
104
1.6
remarkably large j0 of ∼72.9 μA cm−2, which is ∼2.7 and ∼22 times larger than the respective values obtained for CoMoS2 and Ex-MoS2, and ∼45 times larger than the value for bulk MoS2, indicating the excellent electrocatalytic activity of Co-Ex-MoS2. In addition to the overpotential, Tafel slope, and exchange current density, stability is another important parameter in evaluating an electrocatalyst. To investigate the electrochemical stability of Co-Ex-MoS2 in alkaline media, long-term cycling tests were conducted in a KOH (1.0 M) solution. For comparison, a Pt/C electrocatalyst was also tested under the same conditions. After 1000 times of voltammetry cycles, its polarization curve shifts in a negative direction by more than 25 mV at a current density of 10 mA cm−2 (red curve, Figure S10), compared with the first polarization curve (black curve, Figure S10). However, the polarization curve of Co-Ex-MoS2 shows a negligible shift after 1000 cycles, suggesting its excellent electrochemical stability in an alkaline environment (Figure 4c). In addition, the Co-Ex-MoS2 shows good activity after 20 h test (inset in Figure 4c). HRTEM image shows that the structure of Co-Ex-MoS2 is stable after testing (Figure S11). This good stability may originate from the “sandwich” structure of the Co-Ex-MoS2 sample which prevents the Co(OH)2 from aggregating or falling out, and easy mass transfer in this catalytic system. All of the electrochemical results show that the Co-Ex-
CONCLUSIONS In summary, we designed and synthesized HER electrocatalysts consisting of 2D MoS2 confined Co(OH)2 that have excellent activity and stability in alkaline solutions, and outperform most recently reported electrocatalysts (including powders and films) for alkaline HER. Our experimental and theoretical analyses suggest that the excellent HER activity mainly stems from the cojoint roles of confined Co(OH)2 nanoparticles for accelerating water dissociation kinetics and MoS2 nanosheets for speeding up hydrogen generation kinetics. In addition, the confinement of Co(OH)2 in the 2D MoS2 nanosheets prevents it from aggregating or being lost during HER, resulting in CoEx-MoS2 catalysts with good durability. Our work provides an effective method for the rational design of functional materials for improved electrocatalytic activity. This strategy could in principle be extended to other 2D materials (e.g., graphene, MXenes, black phosphorus, and transitional metal dichalcogenides) and intercalated materials (e.g., Ni(OH)2, as shown in Figure S12) to make efficient catalysts for other electrochemical reactions and energy-related applications. F
DOI: 10.1021/acsnano.8b00942 ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
Figure 5. Theoretical analysis of the mechanism for Co-Ex-MoS2 catalyzed HER in alkaline media. (a) Top and side views of relaxed MoS2 (top) and Co(OH)2 (bottom) edges with *OH and *H adsorption. (b) Top and side views of relaxed MoS2 (top) and Co(OH)2 (bottom) edges with *H adsorption. Dark green, yellow, blue, red and white spheres represent Mo, S, Co, O and H atoms, respectively. The *OH and *H on the edges of Co(OH)2 are highlighted in green. (c) Free energy diagrams of HER on the edges of MoS2, Co(OH)2, and Co-Ex-MoS2 sample. (d) Diagram showing HER catalytic reaction in Co-Ex-MoS2. with a beam size of ∼1 μm (Horiba LabRAB HR800, Japan). Structural and chemical analyses of the samples were performed by powder XRD (Cu Kα radiation, λ = 0.15418 nm, Bruker D8 Advance, Germany) as well XPS (monochromatic Al Kα X-rays, 1486.6 eV, PHI VersaProbe II, Japan). Electrochemical Measurements. To prepare a homogeneous ink of the catalysts, catalyst powder (4 mg) was added to a mixture of ethanol (1750 μL), water (200 μL), black carbon (CB, 0.5 mg), and Nafion solution (50 μL, 5 wt %), followed by bath sonication to produce a homogeneous dispersion. Subsequently, the catalyst ink (7 μL) was loaded onto a glass carbon electrode with a diameter of 3 mm (catalyst loading weight ∼0.2 mg cm−2). A standard three-electrode electrolyzer with KOH solution (1.0 M) was used in all tests, with a glassy carbon electrode, a saturated calomel electrode (SCE), and a graphite rod as the working, reference, and counter electrodes, respectively. The scan rates were 5 mV s−1 for linear sweep voltammetry (LSV) tests and 50 mV s−1 for long-term cyclic voltammetry tests. Before each test, the electrolyte was bubbled with argon for 15 min to remove dissolved oxygen in the KOH solution. The onset overpotentials were the potentials where the current starts increasing. The exchange current densities were obtained by extrapolating the corresponding Tafel plots. DFT Simulations. All calculations were carried out using the Vienna Ab Initio Simulation Package (VASP)55,56 adopting Perdew− Burke−Ernzerhof (PBE)57 parametrization of generalized gradient approximation with projector-augmented wave potentials.58,59 Nanoribbons with widths of four (for MoS2) and five (for Co(OH)2) zigzag chains and periodic length of four primitive units were chosen as our models. A vacuum layer thickness larger than 10 Å was selected to ensure spurious interactions were negligible. All structures were fully optimized until the force on each atom was less than 0.01 eV/Å. According to the computational hydrogen electrode models proposed by Nørskov et al.,60 the free energy of OH− was obtained as G(OH¯) = G(H2O) − G(H+) assuming H+ + OH− → H2O was in equilibrium. G(H+), was equal to half the free energy of a hydrogen molecule. When the pH was different from 0, a concentration dependence of the entropy was added to the energy of the proton, i.e., G(pH) = kT ln[H+] = −kT ln10 × pH. Thermodynamic free energies of all states were defined as G = EDFT + EZPE − TS, where EDFT and EZPE are DFT
EXPERIMENTAL SECTION Materials and Chemicals. MoS2 powder (
