Research Article www.acsami.org
Cite This: ACS Appl. Mater. Interfaces 2017, 9, 40752-40759
Conductive Copper Benzenehexathiol Coordination Polymer as a Hydrogen Evolution Catalyst Xing Huang,†,∥ Huiying Yao,‡ Yutao Cui,†,∥ Wei Hao,§ Jia Zhu,*,‡ Wei Xu,*,†,∥ and Daoben Zhu†,∥ †
Beijing National Laboratory for Molecular Sciences CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China ‡ Key Laboratory of Theoretical and Computational Photochemistry of Ministry of Education, Department of Chemistry, Beijing Normal University, Beijing 100875, China § School of Materials Sciences & Engineering, Nanyang Technological University, Singapore 637459, Singapore ∥ University of Chinese Academy of Sciences, Beijing 100049, China S Supporting Information *
ABSTRACT: A graphene-like coordination polymer based on copper(II) benzenehexathiol (Cu-BHT, 1) with high electric conductivity (103 S·cm−1) was prepared recently. The high conductivity makes this material a good candidate for electrocatalysis, and here its catalytic activity toward hydrogen evolution reaction (HER) was evaluated. Cu-BHT shows good activity and stability for HER in acidic solutions under high current densities. By changing the preparation conditions, the morphology of Cu-BHT materials was controlled at the mesoscale, which allows the preparation of a thin film (TF-1), nanocrystal (NC-1), and amorphous nanoparticle (NP-1) of Cu-BHT. The overpotential of Cu-BHT toward HER shows an improved activity from 760 mV (NC-1) to 450 mV (NP-1) at a current density of 10 mA·cm−2. A Tafel slope of ∼95 mV·dec−1 and an exchange current density of 10−3 mA·cm−2 were achieved under optimized conditions. Density functional theory calculations suggest that the “Cu-edge site” on the (100) surface plays an important role in improving the HER catalytic performance of Cu-BHT nanoparticles. KEYWORDS: coordination polymers, metal−dithiolene, hydrogen evolution reaction, morphology control, nanoparticles
1. INTRODUCTION Metal−dithiolene is one of the most promising systems used to construct a non-noble metal hydrogen evolution reaction (HER) catalyst because of its redox flexibility and chemical selectivity.1−7 Several metal−dithiolene complexes show high catalytic activities toward HER. For example, the complex [Co(bdt)2] (where bdt = 1,2-benzenedithiolate) is one of the most active molecular catalysts for both electrochemical and photoelectrochemical hydrogen production from water.1 Furthermore, several copper−dithiolene complexes have also been reported to exhibit modest performance with photocatalytic water reduction activity.6 However, copper−dithiolene-based electrocatalysts for HER have not yet been investigated. Recently, the integration of metal−dithiolene motifs into coordination polymers has been reported.3−5,8 Unlike discrete coordination complexes that work as homogenous catalysts only in organic media,1,2 coordination polymers show catalytic activity in various media with improved stability.3−5,8 For example, the cobalt complex employing the hexadentate dithiolene ligands, such as 1,2,5,6,9,10-triphenylenehexathiol (THT) and benzenehexathiol (BHT), show a superior catalytic performance compared to those of small-molecule metal− dithiolene species.4 Yet, poor conductivity is still one of the © 2017 American Chemical Society
critical restrictions for coordination polymers to act as highperformance electrocatalysts.8,9 A popular idea to solve this problem is employing them as ultrathin films instead of bulk materials.3,10−12 For example, thin-film Ni-THT samples with thicknesses of several nanometers were prepared through an air−liquid interface process (the Langmuir−Blodgett method), which exhibited improved catalytic effects compared to that of the bulk material.3 An overpotential of 333 mV is needed for the Ni-THT thin film to achieve a current density of 10 mA· cm−2 for HER at pH 0.3 However, the synthesis of Ni-THT thin films is a complex and time-consuming procedure not suitable for mass production with limited practical applications; the intrinsically conductive, active, and stable HER electrocatalysts represent an exciting development. We have previously reported a two-dimensional coordination polymer Cu(II) benzenehexathiol (Cu-BHT, 1) which displays the highest electrical conductivity (thin-film sample, ∼1580 S· cm−1) among all the coordination polymers reported to date.13 The high electrical conductivity has inspired us to explore its electrochemical catalytic activity. However, unlike the porous Received: September 25, 2017 Accepted: October 31, 2017 Published: October 31, 2017 40752
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
Research Article
ACS Applied Materials & Interfaces
Figure 1. (a) Three different methods for synthesizing Cu-BHT with corresponding morphologies. SEM images of (b) TF-1, (c) NC-1, and (d) NP1 (scale bars are 3 μm, 1 μm, and 200 nm, respectively).
Figure 2. TEM images of (a) NC-1 and (b) NP-1; the inset image is high-resolution TEM image of a single aggregate. SAED image of (c) NC-1 and (d) NP-1. (e) PXRD pattern of TF-1, NC-1, and NP-1.
2D metal−dithiolene complexes previously reported3,4,7,14 (Figure S1), copper ions and BHT ligands are connected in an extremely dense fashion without porous structures in a Kagome lattice.13 It implies that the Cu-BHT film may have a small specific surface area and lacks active sites, which would severely limit its efficiency as an electrocatalyst. To improve this situation, we have developed a modulation strategy and prepared Cu-BHT nanocrystals and nanoparticles to accomplish the morphology control. These species were synthesized through the homogeneous reaction in much larger quantities than that is possible for the production of thin films through the interface reaction. Despite their different preparation
methods, they have compositions similar to that of the CuBHT thin film and exhibit conductivities at a comparatively high level. Our investigations show that the nanoparticles exhibit much better HER catalytic performance than the nanocrystals, and the Cu-BHT nanoparticle catalyst retains nearly 100% of its current density during prolonged electrochemical cycling and long-time electrolysis. In combination with density functional theory (DFT) calculations, we further investigated the nature of active sites and the reason why CuBHT nanoparticles have superior performance for HER compared to that of Cu-BHT nanocrystals. 40753
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
Research Article
ACS Applied Materials & Interfaces
Figure 3. Electrochemical studies of NC-1 and NP-1. (a,b) CV curves of GCE/NC-1 and GCE/NP-1 in buffer solution with pH 10.0 and scan rate 10 mV/s. (c) CV curves of GCE/NC-1 (black line) and GCE/NP-1 (red line) in H2SO4 solutions at pH 0.0 and scan rate 10 mV/s. (d) Tafel plot based on the CV curves in (c), green line is the linear fitting line.
2. RESULTS AND DISCUSSION 2.1. Morphology of the Cu-BHT Materials. Thin films, nanocrystals, and nanoparticles of Cu-BHT were obtained by various synthesis procedures (as shown in Figure 1a). The thin film of Cu-BHT (TF-1) was synthesized via an interface reaction according to our previous work.13 To obtain smallsized and discrete Cu-BHT materials, we adopted a homogeneous solution method where isolated crystals are formed owing to the absence of interface-confining effects. Furthermore, considering that the degree of deprotonation of BHT affects the coordination reaction speed and thus the morphology of the product, we added CH3ONa to control the deprotonation level. The scanning electron microscopy (SEM) images in Figure 1c,d show the morphological differences in the materials prepared. Without CH3ONa, the reaction between CuCl2 and BHT in ethanol produced a dark purple precipitate. The SEM image in Figure 1c shows that the resulting Cu-BHT solids are nanocrystals (denoted as NC-1). The nanoparticles (Figure 1d) of Cu-BHT (denoted as NP-1) were prepared by a reaction combining CuCl2 and BHT with CH3ONa. A black precipitate is formed immediately upon mixing the reactants. To avoid the formation of large agglomerates, the solution of CuCl2 was spotted slowly into the ethanol solution of deprotonated BHT with vigorous stirring. Unlike the interfacial reaction method used for the syntheses of Co-BTT,7 Co-BHT,4 and Ni-THT3 such as the Langmuir−Blodgett method, the selfassembly reaction in homogeneous solution we employed here requires no special instruments. This method is straightforward and time-saving because the as-synthesized materials directly precipitate from the solution in several minutes. This facile synthetic method is thus highly suitable for mass production. As shown in Figure 2a, NC-1 exists as prism-shaped nanocrystals with several tens of nanometers in diameter and several hundreds of nanometers in length, whereas NP-1 comprises irregular nanoparticles with several nanometers in diameter (Figure 2b). Elemental mapping with energydispersive spectroscopy (EDS) indicates uniform distribution of copper and sulfur (Figure S2). The selected area electron diffraction (SAED) results for NC-1 and NP-1 shown in Figure
2c,d reveal that they share similar diffraction patterns. These observations indicate that the complexes have an identical crystal structure, except the diffraction rings of NP-1 are weaker and more diffusive owing to its smaller particle size and poorer crystallinity. This was further verified by powder X-ray diffraction (PXRD) characterization as shown in Figure 2e; the diffraction peak positions are identical for all three Cu-BHT samples, but the peaks of NP-1 are much broader, indicating its lower crystallinity. The full spectra obtained by the X-ray photon spectroscopy (XPS) instrument of Cu-BHT samples are presented in Figure S3, which verify the absence of sodium and chlorine which are present in the starting reagents. Figure S4 shows the XPS results focused on the Cu 2p and S 2p regions, respectively. The peaks for NC-1 and NP-1 are consistent with those of TF1. Inductively coupled plasma optical emission spectrometry (ICP−OES) analysis revealed that the weight percentage values of copper are 41.1, 41.2, and 38.9% for TF-1, NC-1, and NP-1, respectively. As shown in Table S1, these values are all close to the theoretical value for the proposed chemical structure of [Cu3C6S6]n. The sulfur content and crystallinity show a strong correlation. NP-1 has a smaller size, poorer crystallinity, and larger sulfur content, which means that more dangling sulfur sites exist on the additional surface and in defects. These changes of the surface state and defect concentration would significantly improve its catalytic activity.15,16 Despite the poorer crystallinity of NC-1 and NP-1 compared to that of TF1, they still have excellent electrical conductivity. Using a fourelectrode configuration, the electrical conductivities of compressed pellets of the nanocrystals and nanoparticles of CuBHT were measured to be 280 and 48 S·cm−1, respectively. Although they are 1 to 2 orders of magnitude lower than that of the Cu-BHT thin film (∼1580 S·cm−1),13 they are still higher than those of most other coordination polymers.14,17,18 This high electrical conductivity would promote electron transfer in redox reactions and thus improve electrocatalytic efficiency.19 2.2. Electrochemistry. The electrochemical properties of Cu-BHT were investigated using cyclic voltammetry (CV) with a typical three-electrode system (Figure 3a,b). NC-1 and NP-1 40754
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
Research Article
ACS Applied Materials & Interfaces
durability of NP-1, potential sweeps were conducted in buffer solution with pH 0.0 from 0 to −0.75 V (vs SHE) over 2000 cycles. After cycling, the catalyst exhibited a polarization curve similar to that presented in the initial test (Figure 4), indicating
were suspended in Nafion solution to afford uniformly dispersed suspensions and then loaded on glass carbon electrodes (GCEs) with a loading amount of 84 μg/cm2 or 0.52 × 10−6 molCu/cm2. It was difficult for TF-1 to be fully dispersed because of its condensed nature, so it was directly transferred onto the GCE with a loading amount of 35 μg/cm2 or 0.21 × 10−6 molCu/cm2. Measurements were conducted in a buffer solution of pH 10.0 to avoid the current corresponding to proton reduction. TF-1/GCE showed negligible response in the CV test. This is mainly attributed to the lack of surface area for the nonporous film. As such, we will focus further discussions on NC-1 and NP-1. For NC-1, reductions took place at −320 and −700 mV (vs standard hydrogen electrode, SHE). The corresponding oxidations took place at −20 and −400 mV (vs SHE). The profile of NP-1 looked similar to that of NC-1, except that the peak current density was much smaller and the peak−peak separation between the redox peak and the oxidation peak was smaller. The peak-to-peak separation (ΔEp) reflects the electrochemical rate constant (ko). NC-1 contains larger particles, forms a thicker modified layer on the GCE, and thus has a lower electrochemical reaction rate, as reflected in the smaller ko and larger ΔEp. These differences have also been found for Ni-BHT with differing film thicknesses.14 In our study, different scan rates (from 1 to 1000 mV/s) were also employed (Figure S5). The ΔEp and the peak current exhibited a good linear relationship with the scan rate (Figure S6), indicating a surface-confined quasireversible redox couple. These CV features can be attributed to the transition from [Cu(S−S)2]0 to [Cu(S−S)2]− (I) and then [Cu(S−S)2]2− (II) (S−S = dithiolene) and are consistent with the common electrochemical behavior of metal−dithiolene-based complexes.20 To evaluate the electrocatalytic activity of NC-1 and NP-1 for hydrogen evolution, CV tests were performed in buffer solutions with pH values ranging from 7.0 to 1.0 (Figure S7). For both catalysts, an increase in current was observed when pH was decreased (Figure S8), indicating that a catalytic reaction occurred. In H2SO4 solutions with pH 0.0, NC-1 displayed a high onset potential of 570 mV for HER and an overpotential of 760 mV was required for NC-1 to achieve a current density of 10 mA·cm−2 (Figure 3c). These values are significantly smaller for NP-1, indicating dramatically improved HER activity. The onset potential was 200 mV, and an overpotential of only 450 mV was needed for NP-1 (Figure 3c) to achieve a current density of 10 mA·cm−2 at pH 0.0. In addition, the current density at −600 mV (vs SHE) for NP-1 was 100-fold larger than that for NC-1. For a more complete assessment of the electrode activity parameters, Tafel plots were constructed based on the polarization curves presented in Figure 3c. The Tafel slope is determined by fitting polarization data to the Tafel equation η = a + b log |J|, where η is the overpotential, b is the Tafel slope, and J is the current density. The green line in Figure 3d illustrates the linear fitting results of the Tafel plot. The morphology-dependent optimization is also confirmed by the change in the Tafel slope as it decreases from 120 mV·dec−1 for NC-1 to 95 mV·dec−1 for NP-1. The exchange current density for NP-1 is ∼10−3 mA·cm−2 which is more than 2 orders of magnitude above the value for NC-1 (∼10−5 mA·cm−2). These values are comparable to most of the reported metal−dithiolene-based small molecules and coordination polymer HER catalysts (Table S2).1−5,7 Durability is another important requirement for the practical application of HER catalysts.21 To assess the long-term
Figure 4. A durability test: the black line represents the first CV curves of GCE/NP-1 in 0.5 M H2SO4 solutions at pH 0 and scan rate 100 mV/s, and the red line represents the 2000th circle. The inset image in (d) is the time dependence of the current density of GCE/NP-1 under a static overpotential of 700 mV.
that NP-1 maintained its catalytic performance. In the industrial production of hydrogen, current densities in the order of 103 mA·cm−2 are required.22 As shown in the inset of Figure 4, the NP-1 modified GCE can work under a current density of ∼260 mA·cm−2 without significant loss (less than 1%) beyond 20 h of continuing operation in a highly acid solution (pH 0.0) under an applied potential of −0.7 V (vs SHE). The negligible decay of the cathodic current demonstrates the durability of NP-1 for long-term intensive electrochemical processes. Additionally, inductively coupled plasma−mass spectrometry (ICP−MS) measurements indicate that the amount of copper present in solution is negligible which means Cu-BHT does not decompose after the prolonged electrolysis of water. This remarkable stability in acid solution and under harsh conditions makes NP-1 an excellent electrochemical catalyst for practical applications. The different electrochemical catalysis performances of NP-1 and NC-1 for HER can be attributed first to different morphologies. NP-1 has a much smaller particle size according to the TEM image and dynamic light scattering (DLS) results (Figure S9). The dimensional difference between nanoparticles and nanocrystals will in turn affect the morphologies of the asmodified GCEs, which were characterized using SEM (Figure 5a,b). It can be seen that the surface of GCE/NP-1 was covered by a continuous film with NP-1 dispersed uniformly. Conversely, NC-1 was deposited on the GCE surface as isolated microcrystals rather than a uniform film (Figure 5a). Compared to the separate islands of NC-1, the highly dispersed nanoparticles of NP-1 obviously have a higher interfacial contact area with the electrode and the electrolyte. The change in the interfacial contact area (or electrochemical chemical area) can be roughly quantified by the change in double-layer capacitance (Cdl) using the CV method.19,23 As shown in Figure S10, NP-1 has 3-fold larger Cdl than NC-1. This larger contact interface helps to decrease the contact resistance or faradic impedance (Zf) and offer more accessible active catalytic sites for HER. To clarify this effect, we performed impedance 40755
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
Research Article
ACS Applied Materials & Interfaces
Figure 5. SEM images of the surface of (a) GCE/NC-1 and (b) GCE/ NP-1 (the scale bar is 2.5 μm for (a) and (b) and 500 μm for the inset image in (a) and (b)). (c) Electrochemical impedance spectra of GCE/NC-1 (green line) and GCE/NP-1 (red line) from 10 MHz to 10 mHz are contrasted at the same overpotential (600 mV).
Figure 6. Top and side views of the optimized geometries of the H atoms adsorbed on the (001) and (100) surfaces of Cu-BHT. The H atom adsorbed in the “S top” (a), “C top” (b), “Cu top” (c), “top of the “S3Cu3 ring” (d), “top of the center of the benzene ring” (e) on the (001) surface and “Cu edge” (g),“S edge” (h) on the (100) surface which have been demonstrated along the 001 and 100 directions, respectively. The side views of all H atoms abosorbed on the (001) surface are similar, which is shown in (f). The side views of H atoms adsorbed at the “Cu edge” and “S edge” on the (100) surface are shown in (i) and (j), respectively.
measurements on both electrodes (Figure S11). The GCE/NP1 electrode exhibited much lower impedance (110 Ω) than that of the GCE/NC-1 electrode (1300 Ω) at a overpotential of 600 mV (Figure 5c). The significantly reduced Zf endows the NP-1 catalyst significantly with faster HER kinetics. However, the increase of the surface area is not enough to explain the significant change of catalytic performance, such as the much negative overpotential for NP-1. A deep investigation of the HER activity of NP-1 through DFT calculation is needed. 2.3. DFT Investigation over HER Active Sites. The identification of active sites is of fundamental importance to understand the source of enhanced catalytic activity for NP-1.24 To determine the active site of Cu-BHT for HER, ΔGH* for H* adsorption of different sites on the optimized (001) and (100) surfaces (as shown in Figure S12), which are the representative surfaces according to the PXRD results of Cu-BHT, have been calculated using DFT simulations. The noninnocent nature of the dithiolene ligand provides the possibility that protonation can take place at either metal or sulfur sites.25,26 Figure 6a−f shows that on the (001) surface H atoms reached their minimal ground state at the “S top”, “C top”, “Cu top”, the “top of the “S3Cu3 ring” (the ring is composed of three S atoms and three Cu atoms in the periodic structure), and the “top of the center of the “benzene ring”. Figure 6g−j shows that on the (100) surface H atoms are located at positions of the “S edge”and “Cu edge”. The corresponding adsorption distance and ΔGH* are listed in Table S3. Comparing all the obtained ΔGH* energies, the “Cu-edge” site on the (100) surface is found to be the most active site. This finding is similar to the theory and experiments reported in MoS2 that the S-edge site is crucial for the optimized HER performance.24,27−29 According to the TEM image, NC-1 has a regular shape which means limited “Cuedge” sites are exposed. On the contrary, NP-1 has a more
disordered structure because the formation process of NP-1 was much quicker than that of NC-1, thus smaller domains with additional grain boundaries are generated. These boundaries can not only increase the surface area but also expose many edge sites. It can explain why the HER performance of NP-1 has been dramatically enhanced (300 mV lower potential and 100-fold increase in exchange current density), although the Cdl was just marginally improved (3fold) compared to that of NC-1. However, more characterizations about the amounts of “Cu edge” and other potential active sites, such as S or Cu vacancy, of NC-1 and NP-1 should be performed to construct a more precise description of their effect in catalytic activities.
3. CONCLUSIONS In conclusion, we have demonstrated that a highly conductive copper−dithiolene-based coordination polymer Cu-BHT can work as an HER catalyst. The modification of the synthesis procedure of Cu-BHT affects its morphology and creates more active catalytic sites. Through this morphology control, the overpotential to achieve the current density of 10 mA·cm−2 was decreased from 760 to 450 mV. The economical and benign synthesis process and the remarkable stability of the products have shown the great potential of the coordination polymer for practical applications. In addition, by using DFT calculations, we identify that the “Cu edge” is the most active site in CuBHT. The enhanced performance of NP-1 should partly be attributed to the exposed “Cu edge”. The in-depth study of the 40756
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
Research Article
ACS Applied Materials & Interfaces
from the baseline for 2θ > 35°. Before PXRD testing, the samples should be well-grounded. 4.3.5. Electrical Behavior Characterization. The Cu-BHT powders were manufactured into a compressed cuboid (2 mm × 5 mm × 1 mm) under a pressure of 30 MPa for the characterization of electrical behaviors. The electrical conductivity was measured by a fourelectrode setup equipped with KEITHLEY 2002 Multimeter (Keithley Instrument Inc.). 4.3.6. Electrochemistry of Cu-BHT. Saturated calomel electrodes, glassy carbon working electrodes, and graphite bar electrode were purchased from Tianjin Ada Heng Sheng Technology Development Co., Ltd. The auxiliary graphite electrode was cleaned by submersion in concentrated HCl for 12 h, followed by washing with Milli-Q water and drying in atmosphere before each experiment. Working electrodes were cleaned by submersion in concentrated HCl, followed by sonication for 15 min, washing with Milli-Q water, and also drying in atmosphere. The working electrodes were then sequentially polished with alumina powder of diameters 1, 0.3, and 0.05 from CH Instruments Inc. All electrochemical experiments except electrochemical impedance spectroscopy (EIS) were executed with a CH660 electrochemical workstation (CH Instruments Inc.) in a custom threecompartment electrochemical cell. EIS experiments were performed with ZENNIUM pro (ZAHNER-elektrik GmbH & Co. KG) in the same cell. A graphite rod, placed in a separate compartment connected by porous glass, was used as the counter electrode. The reference electrode, placed in a separate compartment and connected by a Vycor tip, was based on a saturated calomel electrode. The reference electrode in aqueous media was calibrated externally relative to ferrocenecarboxylic acid (Fc-COOH) at pH 7.0, with the Fe 3+/2+ couple at 0.23 V versus Hg/HgO. All potentials reported in this paper were converted to a SHE by adding a value of 0.25 V. The aqueous solutions with different pH (from 0.0 to 10.0) values used in the electrochemical experiments were prepared as follows: for the pH 0.0 solution, 13 mL of concentrated sulfuric acid (about 18 M) was first added to 500 mL of water and then sulfuric acid was continually added dropwise till the pH value reached 0.0 according to the pH meter. For the pH 1.0 solution, 1 mL of concentrated sulfuric acid (about 18 M) was first added to 500 mL of water and then sulfuric acid was continually added dropwise till the pH value reached 1.0 according to the pH meter. For the pH 2.7 solution, citric acid (8.645 g) and Na2HPO4 (3.763 g) were dissolved in water (500 mL). For the pH 4.6 solution, NaOAc (4.013 g) and acetic acid (3 mL) were dissolved in water (500 mL). For the pH 7.0 solution, NaH2PO4 (2.382 g) and Na2HPO4 (8.182 g) were dissolved in water (500 mL). For the pH 10.0 solution, NaHCO3 (1.695 g) and Na2CO3 (3.162 g) were dissolved in water (500 mL). The pH values of the solutions were measured with a Thermo Fisher pH meter. All solutions were degassed and purged with nitrogen. The Cu-BHT samples were well-grounded, and then 2 mg of them was mixed with 100 μL of Nafion solution (0.5 wt %), 450 μL of water and 450 μL of ethanol followed by sonication for 2 h to afford uniform dispersed suspensions. These suspensions (3 μL) were then loaded on the polished round GCE (d = 3 mm, in which the loading amount of the corresponding catalyst is measured to be 84 μg/cm2 or 0.52 × 10−6 molCu/cm2). These modified electrodes were then dried in atmosphere and heated in vacuum at 70 °C for 12 h to remove any residual solvent. 4.4. Theoretical Calculation. All calculations were performed using DFT implemented in the Vienna Ab Initio Simulation Package (VASP) code.31 The optimization of the structures was carried out using the generalized gradient approximation in the Perdew−Burke− Ernzerhof32 form. The dispersion corrections for van der Waals interactions were described by the DFT-D2 method.33 The ion− electron interactions were represented by the projector augmented wave approach.34 A cutoff energy of 500 eV for the plane-wave basis set was employed. The vacuum layer is thick enough to safely ignore the interactions between slabs. Because the two processes on the electrode of adsorption of hydrogen atoms and H2 evolution are competitive in HER, it is essential for a good catalyst to provide H atoms with adequate
active sites of Cu-BHT help us understand the origin of the HER activity of metal−dithiolene-based coordination polymers and provide new insights for guiding the development of this family of HER catalysts.
4. EXPERIMENTAL SECTION 4.1. Materials. CuCl2, NaBr, CH3ONa, citric acid, Na2HPO4, NaH2PO4, NaOAc, NaHCO3, Na2CO3, acetic acid, and Nafion were purchased from Alfa Aesar China (Tianjin) Co., Ltd. Concentrated sulfuric acid was purchased from Sinopharm Chemical Reagent Co., Ltd. Dichloromethane and ethanol were purchased from Acros Organics Co. Water was purified using the Milli-Q purification system. Both solvents were degassed by the freeze−thaw method within pretreatment. BHT was synthesized according to the literature.30 4.2. Synthesis of Cu-BHT. 4.2.1. Film of Cu-BHT (TF-1). The thin film of the Cu-BHT-film was prepared via a reaction between Cu(II) nitrate and BHT at the interface of dichloromethane−water according to our previous work.13 The formed film was filtered from the solution and cleaned with ethanol and acetone. 4.2.2. Nanocrystals of Cu-BHT (NC-1). Under argon atmosphere, 1.5 mM of CuCl2 was first dissolved in 40 mL of degassed ethanol. The fine ground white powder of BHT (135 mg, 0.5 mM) was slowly added to the solution and then reacted for 1 h under room temperature. A dark purple precipitate was formed at room temperature. Also, the formed product was filtered from the solution and cleaned with water, ethanol, and acetone. This prepared solid was heated to 120 °C under dynamic vacuum for 24 h. Yield of dark purple Cu3C6S6: 0.135 g, 91%. Anal. Calcd for Cu3C6S6: Cu, 41.89; C, 15.84; S, 42.27. Found: Cu, 41.2; C, 16.2; S, 42.6. 4.2.3. Nanoparticles of Cu-BHT (NP-1). Under argon atmosphere, 0.5 mM BHT and 3 mM CH3ONa were first dissolved in 30 mL of degassed ethanol to afford deprotonation of BHT. A 10 mL ethanol solution of CuCl2 (1.5 mM) was then added slowly (0.5 mL·min−1) by using a syringe pump while the solution was vigorously stirred. A black precipitate (NP-1) was formed immediately. As NC-1, it was also filtered from the solution and cleaned with water, ethanol, and acetone. The as-prepared solids were heated to 120 °C under dynamic vacuum for 24 h. Yield of black Cu3C6S6: 0.139 g, 92%. Anal. Calcd for Cu3C6S6: Cu, 41.89; C, 15.84; S, 42.27. Found: Cu, 38.9; C, 16.8; S, 44.3. 4.3. Characterization Methods. 4.3.1. Component Analysis. The entire elemental analysis of Cu-BHT was performed with electron probe microanalysis (JEOL, JXA-8100). The content of carbon was analyzed using FlashEA 1112 (Thermo Fisher Scientific). The copper content in Cu-BHT was analyzed by ICP−OES (iCAP 6300 Radial, Thermo Scientific). For ICP−OES measurements, the samples were prepared by dissolving them into fuming nitric acid. Next, the upper solution was diluted to a known volume with Milli-Q water. The metal content in the electrolyte was determined by ICP−MS (Agilent 7800, Agilent Technologies). 4.3.2. SEM and TEM Characterizations. SEM images were obtained using a Hitachi S4800-SEM with an acceleration voltage of 5 kV. For the pretreatment, the samples were transferred to conductive silicon substrates and then covered with a platinum thin film with a thickness of several nanometers. TEM images were obtained by using a JOEL 2100F TEM with an accelerated voltage of 120 kV. 4.3.3. XPS Characterization. XPS results were characterized using an AXIS Ultra-DLD ultrahigh vacuum photoemission spectroscopy system (Kratos Co.). A monochromatic aluminum Kα source (1486.6 eV) was used for XPS. All the characterizations were performed at an initial pressure higher than 3 × 10−9 Torr. All XPS spectra were corrected using the C 1s line at 284.6 eV, and curve fitting and background subtraction were accomplished. 4.3.4. X-ray Characterization. PXRD was performed on a PANalytical Empyrean II X-ray diffractometer. The measurements were employed on D/max 2500 with a Cu Kα source (λ = 1.5406 Å). Samples were observed using a 0.023° 2θ step scan from 5.0 to 45.0° with an exposure time of 30 s per step. No peaks could be resolved 40757
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
Research Article
ACS Applied Materials & Interfaces attractions and meanwhile allow the adsorbed H atoms to form gaseous hydrogen which is released from the electrode according to the Sabatier principle.35 Considering the view of physical chemistry and the Sabatier principle, the HER will reach the maximum rate when the free energy change for H* adsorption (ΔGH*) equals zero, which is one of the key descriptors for determining the activity of HER36,37
Wei Xu: 0000-0003-1950-9037 Daoben Zhu: 0000-0002-6354-940X Author Contributions
J.Z., W.X., and D.Z. initiated and supervised the research program. X.H. was responsible for the material preparation and characterizations. H.Y. performed the DFT calculations. X. H., H.Y., W. X., and J.Z. wrote the manuscript. All the authors participated in the discussion and interpretation of data, read the manuscript, and provided input.
ΔG H * = ΔE H * + ΔEZPE − T ΔS where the adsorption energy ΔEH* is as follows ΔE H * = EM−H * − EM −
1 EH 2 2
Notes
The authors declare no competing financial interest.
where EH2 is the energy of gaseous hydrogen as the reference state. The energy of hydrogen was computed by placing a single H2 molecule in a cubic cell with a length of 10 Å. EM−H* and EM are the total energies of the catalyst surface with and without the H atom, respectively. The difference between the values of zero-point energy change (ΔEZPE) and TΔS (ΔS: the entropy change) is approximately 0.24 eV.36 To determine the active site of Cu-BHT for HER, we have calculated the ΔGH* for H* adsorption in different sites on the representative (001) and (100) surfaces of Cu-BHT, which were extracted from the PXRD results. Layered slab models of (001) and (100) surfaces are constructed with the AA stacking of Cu-BHT layers,13 which have both established a full-relaxed surface layer and a fixed bulk layer. The adsorption of H atoms occurs on the relaxed layer.
■
■
ACKNOWLEDGMENTS We gratefully acknowledge the financial support by the National Natural Science Foundation of China (21333011, 21372227, 21773016) and the Strategic Priority Research Program of the Chinese Academy of Sciences (grant no. XDB12010000).
■
(1) McNamara, W. R.; Han, Z.; Alperin, P. J.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R. A Cobalt−Dithiolene Complex for the Photocatalytic and Electrocatalytic Reduction of Protons. J. Am. Chem. Soc. 2011, 133, 15368−15371. (2) McNamara, W. R.; Han, Z.; Yin, C.-J.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R. Cobalt-dithiolene complexes for the photocatalytic and electrocatalytic reduction of protons in aqueous solutions. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 15594−15599. (3) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Zhang, J.; Feng, X. Large-area, free-standing, two-dimensional supramolecular polymer single-layer sheets for highly efficient electrocatalytic hydrogen evolution. Angew. Chem., Int. Ed. 2015, 54, 12058−12063. (4) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-dimensional metal−organic surfaces for efficient hydrogen evolution from water. J. Am. Chem. Soc. 2015, 137, 118−121. (5) Downes, C. A.; Marinescu, S. C. One dimensional metal dithiolene (M = Ni, Fe, Zn) coordination polymers for the hydrogen evolution reaction. Dalton Trans. 2016, 45, 19311−19321. (6) Zarkadoulas, A.; Koutsouri, E.; Mitsopoulou, C. A. A perspective on solar energy conversion and water photosplitting by dithiolene complexes. Coord. Chem. Rev. 2012, 256, 2424−2434. (7) Downes, C. A.; Marinescu, S. C. Efficient Electrochemical and Photoelectrochemical H2 Production from Water by a Cobalt Dithiolene One-Dimensional Metal−Organic Surface. J. Am. Chem. Soc. 2015, 137, 13740−13743. (8) Zhang, L.; Su, Z.; Jiang, F.; Yang, L.; Qian, J.; Zhou, Y.; Li, W.; Hong, M. Highly graphitized nitrogen-doped porous carbon nanopolyhedra derived from ZIF-8 nanocrystals as efficient electrocatalysts for oxygen reduction reactions. Nanoscale 2014, 6, 6590−6602. (9) Li, Q.; Xu, P.; Gao, W.; Ma, S.; Zhang, G.; Cao, R.; Cho, J.; Wang, H.-L.; Wu, G. Graphene/Graphene-Tube Nanocomposites Templated from Cage-Containing Metal-Organic Frameworks for Oxygen Reduction in Li-O2 Batteries. Adv. Mater. 2014, 26, 1378− 1386. (10) Zhao, S.; Wang, Y.; Dong, J.; He, C.-T.; Yin, H.; An, P.; Zhao, K.; Zhang, X.; Gao, C.; Zhang, L.; Lv, J.; Wang, J.; Zhang, J.; Khattak, A. M.; Khan, N. A.; Wei, Z.; Zhang, J.; Liu, S.; Zhao, H.; Tang, Z. Ultrathin metal−organic framework nanosheets for electrocatalytic oxygen evolution. Nat. Energy 2016, 1, 16184. (11) Yin, H.; Zhao, S.; Zhao, K.; Muqsit, A.; Tang, H.; Chang, L.; Zhao, H.; Gao, Y.; Tang, Z. Ultrathin platinum nanowires grown on single-layered nickel hydroxide with high hydrogen evolution activity. Nat. Commun. 2015, 6, 6430. (12) Sun, J.; Yin, H.; Liu, P.; Wang, Y.; Yao, X.; Tang, Z.; Zhao, H. Molecular engineering of Ni−/Co−porphyrin multilayers on reduced
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b14523. Chemical structure of two-dimensional lattices of Co, Ni, Pd-BHT, Co-THT, and Cu-BHT (Figure S1); the TEMEDS mapping results and corresponding TEM brightfield image for NC-1 and NP-1 (Figure S2); XPS full spectra (Figure S3) and XPS spectra focused on the Cu 2p and S 2p regions (Figure S4) of TF-1, NC-1, and NP1; scan rate-dependent experiments for NC-1 and NP-1 in the pH 10.0 buffer solution (Figure S5); peak-to-peak separation for NC-1 as a function of the scan rate (Figure S6); polarization curves in different solutions (Figure S7) and pH dependence of polarization current density (Figure S8) for NC-1 and NP-1; dynamic light scattering (DLS) results of NC-1 and NP-1 (Figure S9); cyclic voltammograms in the region of 0.50−0.58 V versus SHE for the bare GC, NC-1/GC, and NP-1/GC (Figure S10); electrochemical impedance spectra (EIS) of GCE/ NC-1 and GCE/NP-1 at different overpotentials versus SHE from 10 MHz to 10 mHz (Figure S11); top view and side view of the double-layer Cu-BHT (001) surface (Figure S12); elemental contents for the proposed structure of Cu-BHT ([Cu3(C6S6)]n), TF-1, NC-1, and NP-1 (Table S1); catalytic HER activity of metal− dithiolene coordination polymers (Table S2); and ΔGH* energies and absorption distance for the different sites of H atoms adsorbed on (001) and (100) surfaces of CuBHT (Table S3) (PDF)
■
REFERENCES
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected] (J.Z.). *E-mail:
[email protected] (W.X.). ORCID
Xing Huang: 0000-0001-6807-7440 40758
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
Research Article
ACS Applied Materials & Interfaces graphene oxide sheets as bifunctional catalysts for oxygen evolution and oxygen reduction reactions. Chem. Sci. 2016, 7, 5640−5646. (13) Huang, X.; Sheng, P.; Tu, Z.; Zhang, F.; Wang, J.; Geng, H.; Zou, Y.; Di, C.-a.; Yi, Y.; Sun, Y.; Xu, W.; Zhu, D. A two-dimensional π−d conjugated coordination polymer with extremely high electrical conductivity and ambipolar transport behaviour. Nat. Commun. 2015, 6, 7408. (14) Kambe, T.; Sakamoto, R.; Hoshiko, K.; Takada, K.; Miyachi, M.; Ryu, J.-H.; Sasaki, S.; Kim, J.; Nakazato, K.; Takata, M.; Nishihara, H. π-Conjugated Nickel Bis(dithiolene) Complex Nanosheet. J. Am. Chem. Soc. 2013, 135, 2462−2465. (15) Merki, D.; Fierro, S.; Vrubel, H.; Hu, X. Amorphous molybdenum sulfide films as catalysts for electrochemical hydrogen production in water. Chem. Sci. 2011, 2, 1262−1267. (16) Vrubel, H.; Merki, D.; Hu, X. Hydrogen evolution catalyzed by MoS3 and MoS2 particles. Energy Environ. Sci. 2012, 5, 6136−6144. (17) Sheberla, D.; Sun, L.; Blood-Forsythe, M. A.; Er, S.; Wade, C. R.; Brozek, C. K.; Aspuru-Guzik, A.; Dincă, M. High Electrical Conductivity in Ni3(2,3,6,7,10,11-hexaiminotriphenylene)2, a Semiconducting Metal−Organic Graphene Analogue. J. Am. Chem. Soc. 2014, 136, 8859−8862. (18) Givaja, G.; Amo-Ochoa, P.; Gómez-García, C. J.; Zamora, F. Electrical conductive coordination polymers. Chem. Soc. Rev. 2012, 41, 115−147. (19) Lukowski, M. A.; Daniel, A. S.; Meng, F.; Forticaux, A.; Li, L.; Jin, S. Enhanced Hydrogen Evolution Catalysis from Chemically Exfoliated Metallic MoS2 Nanosheets. J. Am. Chem. Soc. 2013, 135, 10274−10277. (20) Bowmaker, G. A.; Boyd, P. D. W.; Campbell, G. K. Electrochemical and ESR studies of the redox reactions of nickel(II), palladium(II), and platinum(II) complexes of 1,2-diphenyl-1,2ethenedithiolate(2-)-S,S′. Inorg. Chem. 1983, 22, 1208−1213. (21) Turner, J. A. Sustainable Hydrogen Production. Science 2004, 305, 972−974. (22) Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. S. Solar Water Splitting Cells. Chem. Rev. 2010, 110, 6446−6473. (23) Da Silva, L. M.; De Faria, L. A.; Boodts, J. F. C. Determination of the morphology factor of oxide layers. Electrochim. Acta 2001, 47, 395−403. (24) Jaramillo, T. F.; Jorgensen, K. P.; Bonde, J.; Nielsen, J. H.; Horch, S.; Chorkendorff, I. Identification of Active Edge Sites for Electrochemical H2 Evolution from MoS2 Nanocatalysts. Science 2007, 317, 100. (25) Ward, M. D.; McCleverty, J. A. Non-innocent behaviour in mononuclear and polynuclear complexes: consequences for redox and electronic spectroscopic properties. J. Chem. Soc., Dalton Trans. 2002, 275−288. (26) Ray, K.; Petrenko, T.; Wieghardt, K.; Neese, F. Joint spectroscopic and theoretical investigations of transition metal complexes involving non-innocent ligands. Dalton Trans. 2007, 1552−1566. (27) Hinnemann, B.; Moses, P. G.; Bonde, J.; Joergensen, K. P.; Nielsen, J. H.; Horch, S.; Chorkendorff, I.; Noerskov, J. K. Biomimetic Hydrogen Evolution: MoS2 Nanoparticles as Catalyst for Hydrogen Evolution. J. Am. Chem. Soc. 2005, 127, 5308. (28) Bonde, J.; Moses, P. G.; Jaramillo, T. F.; Nørskov, J. K.; Chorkendorff, I. Hydrogen evolution on nano-particulate transition metal sulfides. Faraday Discuss. 2009, 140, 219−231. (29) Li, Y.; Wang, H.; Xie, L.; Liang, Y.; Hong, G.; Dai, H. MoS2 Nanoparticles Grown on Graphene: An Advanced Catalyst for the Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2011, 133, 7296− 7299. (30) Yip, H. K.; Schier, A.; Riede, J.; Schmidbaur, H. Benzenehexathiol as a template rim for a golden wheel: synthesis and structure of [{CSAu(PPh3)}6]. J. Chem. Soc., Dalton Trans. 1994, 2333−2334. (31) Kresse, G.; Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54, 11169.
(32) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865. (33) Grimme, S. Semiempirical GGA-type density functional constructed with a long-range dispersion correction. J. Comput. Chem. 2006, 27, 1787−1799. (34) Kresse, G.; Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59, 1758. (35) Sabatier, P. Hydrogénations et déshydrogénations par catalyse. Ber. Dtsch. Chem. Ges. 1911, 44, 1984−2001. (36) Nørskov, J. K.; Bligaard, T.; Logadottir, A.; Kitchin, J. R.; Chen, J. G.; Pandelov, S.; Stimming, U. Trends in the Exchange Current for Hydrogen Evolution. J. Electrochem. Soc. 2005, 152, J23. (37) Zheng, Y.; Jiao, Y.; Jaroniec, M.; Qiao, S. Z. Advancing the electrochemistry of the hydrogen-evolution reaction through combining experiment and theory. Angew. Chem., Int. Ed. 2015, 54, 52−65.
40759
DOI: 10.1021/acsami.7b14523 ACS Appl. Mater. Interfaces 2017, 9, 40752−40759
