Microbial-Phosphorus-Enabled Synthesis of Phosphide

Jul 28, 2017 - Add to Favorites · Download Citation · Email a Colleague · Order Reprints · Rights & Permissions · Citation Alerts · Add to ACS ChemWor...
6 downloads 12 Views 4MB Size
Download PDF
Article pubs.acs.org/JACS

Microbial-Phosphorus-Enabled Synthesis of Phosphide Nanocomposites for Efficient Electrocatalysts Tian-Qi Zhang,†,‡ Jian Liu,†,‡ Lin-Bo Huang,†,‡ Xu-Dong Zhang,†,‡ Yong-Gang Sun,†,‡ Xiao-Chan Liu,†,‡ De-Shan Bin,†,‡ Xi Chen,§ An-Min Cao,*,†,‡ Jin-Song Hu,*,†,‡ and Li-Jun Wan*,†,‡ †

CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences (ICCAS), Beijing 100190, China ‡ University of Chinese Academy of Sciences (UCAS), Beijing 100049, China § University of Science and Technology of China (USTC), Hefei 230026, China S Supporting Information *

ABSTRACT: Transition-metal phosphides have recently been identified as low-cost and efficient electrocatalysts that are highly active for the hydrogen evolution reaction. Unfortunately, to achieve a controlled phosphidation of nonprecious metals toward a desired nanostructure of metal phosphides, the synthetic processes usually turned complicated, high-cost, and even dangerous due to the reaction chemistry related to different phosphorus sources. It becomes even more challenging when considering the integration of those active metal phosphides with the structural engineering of their conductive matrix toward a favorable architecture for optimized catalytic performance. Herein, we identified that the biomass itself could act as an effective synthetic platform for the construction of supported metal phosphides by recovering its inner phosphorus upon reacting with transition-metals ions, forming well-dispersed, highly active nanoparticles of metal phosphides incorporated in the nanoporous carbon matrix, which promised high catalytic activity in the hydrogen evolution reaction. Our synthetic protocol not only provides a simple and effective strategy for the construction of a large variety of highly active nanoparticles of metal phosphides but also envisions new perspectives on an integrated utilization of the essential ingredients, particularly phosphorus, together with the innate architecture of the existing biomass for the creation of functional nanomaterials toward sustainable energy development.

■

INTRODUCTION The increasing demand on energy has raised serious concerns on the development of clean and sustainable energy sources, among which the production of H2 from the water-splitting process becomes an appealing solution to ensure an affordable energy supply.1−3 To facilitate the hydrogen evolution, different catalysts, typically noble metals such as Pt and Ir,4,5 have been identified to promote the electrochemical water splitting with high activity and low overpotentials.6,7 Unfortunately, due to the high cost and scarcity of the noble metals, it has become a persistence pursuit but a formidable challenge to develop highly efficient catalysts that are favorably composed of earthabundant elements.8 Recently, transition-metal phosphides have been identified as a family of nonprecious metal catalysts that were able to accelerate the hydrogen evolution reaction (HER).9−11 Particularly, well-dispersed nanoparticles of phosphides of transition metals, including Co, Ni, and Fe, could exhibit high HER activity while an extraordinary mechanical strength and chemical stability are maintained, showing promise toward a possible alternative for non-noblemetal catalysts.12−14 Accordingly, the construction of different transition-metal phosphides (TMPs) with designated nanostructures and functions has been under intensive pursuit so as © 2017 American Chemical Society

to fully exploit the potential of TMPs for clean energy production. Despite the low cost of the compositional elements in TMPs, it remains an open challenge to ensure an economic and environmentally friendly synthesis of the TMP nanocatalysts considering the fact that the current phosphorus donors are usually related to expensive and dangerous precursors.15,16 For example, to achieve a successful phosphidation of transition metals, reactants such as trinoctyl phosphine (TOP) or PH3 have been widely used as effective phosphorus sources so as to achieve a desired structural and compositional control, which makes the synthetic process expensive, complicated, and even dangerous.17,18 It has been highly demanded that a new phosphidation strategy with highly active alternatives of these precursors be identified for the preparation of TMP catalysts. Equally importantly, the structural engineering of the nanosized TMP particles should be favorably coupled with the architectural design of their conductive supports, typically carbon, so that the electrical conductivity and dispersity of the active components could be optimized to ensure an improved Received: June 13, 2017 Published: July 28, 2017 11248

DOI: 10.1021/jacs.7b06123 J. Am. Chem. Soc. 2017, 139, 11248−11253

Article

Journal of the American Chemical Society electrochemical performance.19 Accordingly, it becomes appealing to develop a synthetic strategy to integrate the formation of TMPs together with the architectural tailoring of their supports, forming nanocatalysts with active nanoparticles of TMPs anchored in a highly porous carbon matrix. It is also a prerequisite that the new synthetic protocol be low cost and environmentally benign in terms of both the synthesis process and the phosphorus precursors. Biomass has long been qualified as raw carbon precursors for the production of different functional materials with designated structures and has found broad applications in different areas, including catalysts, sorption materials, and biochemicals.20−22 The biomass-based activated carbon showed attractive characteristics, including but not limited to the following: (1) The biomass production is usually related to environmentally benign processes with high quantity and quality, which guarantees a reliable feedstock of carbon production through a controlled carbonization process.23 (2) The transformation of the biomass precursors to active carbon showed a high flexibility in the structural design and the architectural control of key morphological parameters, such as the shape, size, and porosities.23,24 (3) The rich chemistry of the biomass leaves enormous room for the further functionalization and compositional control of the carbonized species. The abundant surface functional groups, such as C−O, COOH−, and OH, as well as the rich compositional element nitrogen, have endowed the biomass with further possibilities to tune the physicochemical properties of activated carbons.22,25 In our research, we have a special interest in yeast, a microorganism widely used in the ethanol and baking industry, mainly due to its high content of phosphorus. As a main nutrient required for yeast growth, phosphorus accounts for almost 0.3 wt % of the total weight.26 Interestingly, we found that the yeast itself can be developed into an ideal synthetic platform for supporting nanocatalysts with highly dispersed metal phosphide nanoparticles embedded in a mesoporous carbon matrix. We demonstrated that an in situ phosphidation of the metal species could be well-integrated with the structure engineering of the carbon matrix, resulting in favorable architectures of supported catalysts applicable to a large variety of metal phosphides, such as Co2P, Mn2P, Ni2P, and Zn3P2. We showed that the prepared nanocomposite could be used as a promising catalyst that was able to accelerate the hydrogen evolution reaction (HER) with high activity and stability. Our synthetic protocol not only provides a simple and effective strategy for the construction of a large variety of metal phosphides for morphological control and performance optimization but also demonstrates the possibility of a full utilization of the essential ingredients, as well as the innate architecture of the existing biomass, in the creation of functional nanomaterials toward sustainable energy development.

Scheme 1. Schematic Illustration of the Synthesis of Metal Phosphide−C Nanocompositea

a

Yeast particles (in light yellow) are the biomass precursor. NaCl and glutaraldehyde can induce different pores in the formed carbonaceous matrix (in blue) during the hydrothermal reaction. Metal ions can be trapped inside the porous matrix and produce metal oxide nanoparticles (in red). The high-temperature heating results in a conductive substrate (in deep blue) and a simultaneous formation of metal phosphide nanoparticles (in yellow) accomplished by the biomass phosphorous.

a hollow and porous architecture, making its inner structure easily accessible for the acquisition of the coexistent metal ions. The carbonaceous product after the hydrothermal reaction was then treated at a high temperature of 900 °C under an inert atmosphere, which simultaneously initiated a carbonization process of the organic species together with an efficient phosphidation of the inner metal ions by the constitutional phosphorus of yeast, forming a composite structure with uniform nanoparticles of metal phosphides dispersed across the highly porous carbon matrix. Figure 1a shows a scanning electron microscopy (SEM) image of the cobalt-containing sample after the high temper-

Figure 1. SEM (a) and TEM (b) images of the Co2P−C sample prepared with 0.09 M Co(NO3)2 used in the hydrothermal reaction. The inset in part a showed a high-magnification SEM image. (c) High resolution TEM image of randomly selected particles in part b. TEM images of Co2P−C nanocomposites synthesized by controlling the Co(NO3)2 concentrations at (d) 0.18 M, (e) 0.36 M, and (f) 0.54 M, respectively.

■

RESULTS AND DISCUSSION Scheme 1 shows the synthesis route we adopted for the construction of metal−phosphide-based nanocomposites. First, the yeast was treated under a hydrothermal condition, which is a process widely applied to transform biomasses to carbonaceous species with rich surface functional groups.22 To ensure a satisfactory utilization of the whole carbon matrix for further functionalization, we designed an etching process initiated by the additives NaCl and glutaraldehyde (GA) in the hydrothermal process. Accordingly, the yeast can be transformed into

ature treatment. The final product exists as nearly spherical microparticles with a size around 2 μm. A close look at the particle surface (inset in Figure 1a) reveals the existence of holes with different sizes on the wall. A transmission electron microscopy (TEM) analysis was then carried out to look inside the particles. Figure 1b shows a dark-field TEM image of a randomly selected particle, which details a porous structure with different kinds of cavities distributed across the microparticle. Meanwhile, a careful TEM observation is able to locate 11249

DOI: 10.1021/jacs.7b06123 J. Am. Chem. Soc. 2017, 139, 11248−11253

Article

Journal of the American Chemical Society

observe an obvious change in the shape and size for either the Co2P particles or the carbon matrix, showing a reliable capability for the structural control of the key components in the Co2P−C nanocomposite. The actual weight loading of the Co2P particles in each sample is listed in Table S1 (SI). Further investigation was carried out on the Co2P−C nanocomposite with a focus on its compositional distribution. Figure S2 (SI) shows the elemental mapping of a representative Co2P−C particle that was randomly picked for compositional analysis. The existence of three concerned elements, namely, C, Co, P, is verified and their distributions are in good agreement with the shape character of the microsized particle. A detailed elemental analysis was then carried out on single Co2P nanoparticles with the help of the line-scan technique of energy-dispersive X-ray spectroscopy (EDX). As shown in Figure 2b, the profiles of Co and P fit well to the outlines of the nanoparticle, confirming the formation of Co2P nanoparticles from our synthetic approach. The highly porous structure of Co2P−C is further determined by the nitrogen adsorption− desorption isotherms (Figure 2c), which shows a characteristic hysteresis loop in the curves and revealed the mesoporous nature of the carbon matrix as a result of the high-temperature treatment. The result from the Barrett−Joyner−Halenda (BJH) pore size and volume analysis shows a relatively broad distribution of mesopores, which is in good agreement with the TEM observation. Furthermore, the specific surface area of the nanocomposite is measured to be as high as 632.8 m2/g, which is highly desirable for its further applications when mass transport is a concern, typically catalysis with a need for more accessible active sites. The Raman spectrum of this sample (Figure 2d) shows a high intensity ratio of D-band over Gband, which indicates a relatively low graphitic degree of the formed carbon, a phenomenon that agrees well with the HRTEM observation of the carbon matrix. Further characterization is then carried out with the help of X-ray photoelectron spectroscopy (XPS), which provides detailed information on the surface property of the Co2P−C sample. As shown in Figure S3 (SI), the signal of nitrogen is detected, indicating the existence of nitrogen doping in the carbon matrix after the high-temperature treatment, as a result of the residue from the constituent nitrogen in yeast. Intrigued by the effectiveness and simplicity of our synthetic route in forming Co2P−C, we took a close examination of the key steps so as to track the formation mechanism of this nanocomposite. Figure S4a,b (SI) shows the TEM and SEM images of the yeast particles used for the reaction, which exist as solid particles with a larger particle size of around 4 μm. For the precursor sample collected after the hydrothermal reaction (denoted as p-Co 2 P−C), the TEM image shows the incorporation of tiny nanoparticles inside the biomass-derived carbonaceous matrix. Elemental analysis (inset in Figure S5a, SI) discloses the existence of both cobalt and oxygen inside. Its XRD curve in Figure S6 showed tiny but discernible peaks of Co3O4 (PDF #43-1003). The HRTEM analysis on these cobalt-containing nanoparticles confirms that there exist poorly defined but clearly discernible lattice fringes. A fast Fourier transformation (FFT) of the HRTEM image (Figure S5b, SI) shows weak FFT points in the pattern, and their interplanar spacing are calculated to be 0.285 and 0.243 nm, which are in good agreement with the distances from the (220) and (311) planes of Co3O4, respectively. It is therefore expected that a hydrothermal treatment transforms the cobalt species in solution into poorly crystallized Co3O4 nanoparticles, which

highly dispersed nanoparticles embedded inside the porous matrix. The X-ray diffraction (XRD) pattern of this cobaltcontaining sample identifies the emergence of Co2P (PDF #893030) with no extra impurities (Figure 2a). Meanwhile, a blank

Figure 2. (a) XRD pattern of the prepared Co2P−C product. (b) Line-scan analysis of a single Co2P particle (red, Co; yellow, P). (c) Nitrogen adsorption−desorption isotherms of Co2P−C; the inset shows the mesopore size distributions in the sample. (d) Raman spectrum of the Co2P−C sample.

test created by excluding the Co2+ species produced only porous carbon [Figure S1a,b, Supporting Information (SI)] that is amorphous in nature, as seen from its XRD pattern (Figure S1c, SI). Therefore, it is reasonable to expect that these emerged nanoparticles are crystalline Co2P rooted from the Co2+ species. As a matter of fact, we expected to confirm the existence of Co2P, since its formation usually necessitates a phosphidation process, which is achieved by the designated phosphorus reactants.27−30 The emergence of Co2P in our synthesis inspired us to revisit the yeast itself, which is a widely researched biomass with a well-known, but easily ignored, high content of phosphorus. Clearly, the phosphidation of cobalt was no doubt accomplished by the constitutional phosphorus of yeast during the heating, which enlightened us of an interesting tool to prepare metal phosphide without the involvement of those expensive and dangerous precursors discussed above. High-resolution TEM (HRTEM) analysis was carried out on these embedded nanoparticles of Co2P. Figure 1c shows the HRTEM image of a randomly selected Co2P particle, which is crystalline, as revealed by the lattice fringes. The lattice spacings are measured to be 0.272 nm, corresponding to the (111) planes of the orthorhombic Co2P. We found that our synthesis route showed a flexible capability to form and host an increased amount of Co2P particles by simply using higher concentrations of the cobalt precursor, namely Co(NO3)2, in the hydrothermal reaction. Figure 1d−f show the TEM images of different Co2P−carbon composites (denoted as Co2P−C) with Co2+ contents controlled at different levels. It is clear that more Co2P nanoparticles formed inside as we doubled the Co 2+ concentration from 0.09 M (Figure 1b) to 0.18 M (Figure 1d). An increase to 0.36 M of Co(NO3)2 could further manifest such a trend (Figure 1e), and finally, the carbon matrix would be swamped with densely populated Co2P particles, as shown in Figure 1f, with 0.54 M Co(NO3)2. Interestingly, we did not 11250

DOI: 10.1021/jacs.7b06123 J. Am. Chem. Soc. 2017, 139, 11248−11253

Article

Journal of the American Chemical Society is actually not surprising when considering the fact that hydrothermal processes have been widely adopted as a reliable protocol for the preparation of Co3O4 nanocrystals.31 Our control experiments highlight the importance of NaCl as an effective cavity creator during the hydrothermal treatment. Without NaCl, the formed microparticles would be mostly solid in shape with no obvious inner cavities observed (Figure S7a). The selection of NaCl in our synthesis is based on the fact that NaCl is able to damage the cell walls as a result of the salt stress.32 In this way, we are able to channel into the inner particle to form different holes inside and, accordingly, make the whole particle accessible for mass transport of the cobalt species. To prevent the collapse of the yeast during the cavitation, GA is used to strengthen the skeleton of the yeast, which is a widely used additive in yeast treatment due to its capability to improve the strength of polysaccharide networks.22,33 For comparison, in the synthesis without GA, yeast particles will collapse into sticky and solid ones (Figure S7b, SI). We consider that a successful entrapment of cobalt species inside the carbonaceous matrix is probably benefiting from the following two facts: First, the porous and open structure can facilitate the transportation of the Co2+ species into its inner part. Otherwise only a limited amount of nanoparticles can be found inside the solid microparticles, showing the inadequate capability of the untreated matrix to acquire the coexistent metal species (Figure S7c, SI) as compared to the Co2P−C sample we produced in Figure S7d (SI). Second, the rich surface groups can provide rich functionality to anchor Co2+. The Fourier transform infrared spectroscopy (FTIR) curve shown in Figure S8 (SI) confirms the existence of different amphiphilic groups, including hydroxyl, carbonyl, and carboxyl. It has been expected that such a hydrothermally treated biomass surface is suitable for the adsorption of metal ions due to the electrostatic attraction,22,34 which explains well our observation that an efficient entrapment of cobalt species was achieved inside the carbonaceous matrix without forming a large amount of separate Co3O4 particles. Most importantly, we are able to achieve an in situ phosphidation of the anchored metal species during the high-temperature treatment, which is no doubt ascribed to the rich phosphorus constituent in the carbon matrix, since yeast itself is the only reactant source we have. Favorably, the high-temperature treatment designated for the carbonization process also ensures a suitable inert environment for the phosphidation of the metals. The synthetic protocol demonstrated here successfully circumvents the challenge, as discussed above, of the preparation of metal phosphides, making the preparation of highly dispersed nanoparticles simple and effective through a revisit of the compositional chemistry of the microbial biomass. Considering that there is no specific requirements necessitated on the metal ions for the preparation of metal phosphides, we expect that such a synthetic route could be readily applicable for other metals ions, as long as they can be absorbed and dispersed inside the carbonaceous matrix. Accordingly, a large variety of metal ions have been tested by following the same recipe. Figure 3 shows the TEM (Figure 3a−c) and XRD (Figure 4d) results of three representative samples when nitrates of manganese, nickel, and zinc are used instead of cobalt. Similar to the TEM observations on the Co2P−C sample, all these samples show similar structural characteristics, where highly dispersed nanoparticles are embedded inside the porous carbon matrix. The inset in each

Figure 3. Dark-field TEM images for the samples of (a) Mn2P−C, (b) Ni2P−C, and (c) Zn3P2−C, respectively. The inset in each TEM image showed the line scan result of a randomly selected particle from the corresponding sample. The XRD patterns of all of these three different samples are shown in part d.

Figure 4. (a) Polarization curves of three different samples, including Co2P−C, blank C, control sample, and 20 wt % Pt−C (tested in 0.5 M H2SO4 at a scan rate of 5 mV s−1). (b) Tafel plots of both Co2P−C and 20 wt % Pt−C. (c) Durability test of Co2P−C sample tested in 0.5 M H2SO4. (d) Capacitive currents of Co2P−C tested at 0.164 V as a function of scan rates. Inset: CV curves at different scan rates from 4 to 16 mV s−1.

TEM image shows a corresponding elemental analysis on single nanoparticles. The line scan results clearly confirm the formation of phosphides for different metals, including manganese (Figure 3a), nickel (Figure 3b), and zinc (Figure 3c), respectively. The XRD characterization confirms the formation of their crystalline phases, with no impurity phases observed. The HRTEM analyses were also carried out on nanoparticles of different metal phosphides, and the results are 11251

DOI: 10.1021/jacs.7b06123 J. Am. Chem. Soc. 2017, 139, 11248−11253

Journal of the American Chemical Society

Article

■

CONCLUSION In conclusion, we demonstrated that phosphorus-containing yeast can act as an effective synthetic platform for the preparation of active nanocatalysts with metal phosphide nanoparticles dispersed across the porous carbon matrix. We verified that NaCl can act as an effective etching agent for the yeast, which could be engineered into a highly porous structure toward an effective entrapment of metal species inside by utilizing the rich surface functional groups. We demonstrated that an in situ phosphidation of the metal species could be wellintegrated with the structure control of the carbon matrix, forming favorable architectures of supported catalysts applicable to a large variety of metal phosphides, such as Co2P, Mn2P, Ni2P, and Zn3P2. Our preliminary test confirmed that the prepared nanocomposite could be promising to accelerate the hydrogen evolution reaction with high activity and stability. Such a synthetic protocol not only provided a simple and effective strategy for the construction of a large variety of metal phosphide nanocomposites but also demonstrated the possibility of a full utilization of the essential ingredients as well as the innate architecture of the existing biomass in the creation of functional nanomaterials toward sustainable energy.

shown in Figure S9 (SI), which shows the crystalline feature of the supported particles. It is also noted that these nanoparticles of metal phosphides are mostly incorporated into the carbon structures, showing the effectiveness of the adsorption and phosphidation strategy in the construction of the metal phosphide−C (MP−C) nanocomposite. The consistency in the structural and shape control also reveals the versatility and reliability of our synthetic protocol, which is critical for the further evaluation of their performance in different application areas. The highly dispersed nanoparticles of metal phosphides incorporated in a high-surface-area support have attracted increased attention as active catalysts for different reactions, such as hydrogenation, desulfurization,35 and water electrolysis.36 The last of these potentials is demonstrated here by using the Co2P−C nanocomposite as a HER electrocatalyst. Figure 4a shows the HER polarization curves of different samples when tested in 0.5 M H2SO4. For comparison, the catalytic performance for three different samples, including the biomass-derived carbon, 20 wt % Pt−C, and a Co2P control sample prepared by a conventional wet-chemistry protocol (Figure S10, SI; see the Experimental Section for details) are also evaluated. As expected, the pure carbon itself without Co2P nanoparticles inside shows barely any HER activity. On the contrary, the existence of Co2P in the carbon matrix shows an obvious boost of the catalytic activity. The overpotential of this Co2P−C composite is measured to be 96 mV at catholic current densities of 10 mA/cm2 as compared to 178 mV of the control sample, showing that the formed Co2P−C nanocomposite is a highly active catalyst (a detailed comparison with the reported results on Co2P is listed in Table S2, SI). Meanwhile, we also observe a rapid increase of the cathode current as the potential turned more negative, revealing the favorable capability of our Co2P−C nanocomposite to facilitate the hydrogen evolution from water.12,37 Figure 4b displays the Tafel plots, which are the graphs of the overpotential against the logarithm of the current density. The Tafel slope of Co2P− C is 68 mV dec−1, indicating that the hydrogen evolution on Co2P−C should follow the Volmer−Heyrovsky mechanism.38,39 To have a better understanding of the durability of the Co2P−C sample as a reliable catalyst, we have carried out a further stability study through accelerated degradation. The time-dependent current density curve under a static overpotential of 50 mV (Figure 4c) showed that Co2P−C would maintain its catalytic activity for at least 8 h. The inset in Figure 4c recorded the overpotential values after 1000 continuous CV scans at a scan rate of 50 mV/s. The sample showed almost negligible difference in the overpotential upon long cycles, showing the high stability of the Co2P−C nanocomposite under the acid condition. To have a better understanding of the electrochemical surface area (ECSA) of the Co2P−C nanocomposite, the double-layer capacitance of the material was tested, due to its positive correlation to the ECSA value.40,41 As shown in Figure 4d, the calculated value was 24.75 mF, corresponding to 124.1 mF cm−2 normalized by the geometric area of the electrode, which was much higher than the value of 6.03 mF from the control sample (Figure S11, SI). A higher ECSA is known to indicate the capacity to provide enough active sites for the HER, which is expected to be a major factor to achieve a high HER performance, as observed for our Co2P− C nanocomposite.42

■

EXPERIMENTAL SECTION

Preparation of Carbon Microspheres. One gram of yeast cells prewashed with acetone were dissolved in 30 mL of pure water with magnetic stirring. Subsequently, 5 mL of sodium chloride solution (100 mg/mL) and 100 μL of glutaraldehyde aqueous solution were added to the yeast suspensions, which was then placed in a 50 mL Teflon-sealed autoclave and maintained at 190 °C for 8 h. The solid products with a purplish brown color were centrifuged, washed by three cycles of centrifugation/washing/redispersion in pure water, and oven-dried at 80 °C for 10 h. Finally, the obtained product was prepared via further carbonization at a temperature of 900 °C for 6 h with a tubular reactor in a flow of argon. After naturally cooling to room temperature, black carbonized products were obtained. Preparation of Co2P−C, Mn2P−C, Zn3P2−C, and Ni2P−C Microspheres. One gram of yeast cells prewashed with acetone were dissolved in of 30 mL pure water with magnetic stirring. Subsequently, 5 mL of sodium chloride solution (100 mg/mL), 100 μL of glutaraldehyde aqueous solution, and 0.18 mol/L transition-metal nitrate [cobaltous nitrate hexahydrate, manganese nitrate, zinc nitrate hexahydrate, or nickel(II) nitrate hexahydrate] were added to the yeast suspensions, which were then placed in a 50 mL Teflon-sealed autoclave and maintained at 190 °C for 8 h. The solid products with a purplish brown color were centrifugally separated, washed by three cycles of centrifugation/washing/redispersion in pure water, and ovendried at 80 °C for 10 h. Finally, the obtained product was prepared via further carbonization at 900 °C for 6 h, respectively, with a tubular reactor in a flow of argon. After naturally cooling to room temperature, black carbonized products were obtained. Preparation of the Co2P Control Sample. Hollow microspheres were prepared by following the protocol reported in ref 43. A 0.5 g portion of CoCl2 and 1.2 g of yellow phosphorus were dispersed in 30 mL of ethanol. The suspension was then added into the 40 mL Teflonlined autoclave. The autoclave was heated to 200 °C for 30 h and the formed solid product was collected for further characterizations. Electrochemical Measurement. All of the electrochemical measurements were performed in a three-electrode system on an electrochemical workstation (CHI 660D) in 0.5 M H2SO4 electrolyte. Typically, 3 mg of catalyst and 30 μL of Nafion solution (SigmaAldrich, 5 wt %) were dispersed in 1.2 mL of ethanol solution by sonicating for 1 h to form a homogeneous ink. Then 30 μL of the dispersion (containing 75 μg of catalyst) was loaded onto a glassy carbon electrode with 5 mm diameter (loading 0.382 mg cm−2). A Hg/HgCl2 (in 4.2 mol KCl solution) electrode and a graphite rod served as the reference electrode and counter electrodes, respectively. 11252

DOI: 10.1021/jacs.7b06123 J. Am. Chem. Soc. 2017, 139, 11248−11253

Article

Journal of the American Chemical Society All of the potentials were calibrated to a reversible hydrogen electrode (RHE). The working electrode was polished with Al2O3 powders with size down to 0.05 μm.

■

(13) Andreiadis, E. S.; Jacques, P.-A.; Tran, P. D.; Leyris, A.; Chavarot-Kerlidou, M.; Jousselme, B.; Matheron, M.; Pécaut, J.; Palacin, S.; Fontecave, M.; et al. Nat. Chem. 2013, 5, 48. (14) Miao, J.; Xiao, F.-X.; Yang, H. B.; Khoo, S. Y.; Chen, J.; Fan, Z.; Hsu, Y.-Y.; Chen, H. M.; Zhang, H.; Liu, B. Sci. Adv. 2015, 1, e1500259. (15) Callejas, J. F.; Read, C. G.; Roske, C. W.; Lewis, N. S.; Schaak, R. E. Chem. Mater. 2016, 28, 6017. (16) Prins, R.; Bussell, M. E. Catal. Lett. 2012, 142, 1413. (17) Park, J.; Koo, B.; Yoon, K. Y.; Hwang, Y.; Kang, M.; Park, J. G.; Hyeon, T. J. Am. Chem. Soc. 2005, 127, 8433. (18) Li, Y.; Malik, M. A.; O’Brien, P. J. Am. Chem. Soc. 2005, 127, 16020. (19) Wang, S.; Xiao, C.; Xing, Y.; Xu, H.; Zhang, S. J. Mater. Chem. A 2015, 3, 15591. (20) Wu, Z.-Y.; Liang, H.-W.; Chen, L.-F.; Hu, B.-C.; Yu, S.-H. Acc. Chem. Res. 2016, 49, 96. (21) Gong, X.; Liu, S.; Ouyang, C.; Strasser, P.; Yang, R. ACS Catal. 2015, 5, 920. (22) Guan, Z.; Liu, L.; He, L.; Yang, S. J. Hazard. Mater. 2011, 196, 270. (23) White, R. J.; Budarin, V.; Luque, R.; Clark, J. H.; Macquarrie, D. J. Chem. Soc. Rev. 2009, 38, 3401. (24) Dutta, S.; Bhaumik, A.; Wu, K. C.-W. Energy Environ. Sci. 2014, 7, 3574. (25) Liu, W.-J.; Jiang, H.; Yu, H.-Q. Chem. Rev. 2015, 115, 12251. (26) Northcote, D.; Horne, R. Biochem. J. 1952, 51, 232. (27) Hu, B.; Wang, K.; Wu, L.; Yu, S. H.; Antonietti, M.; Titirici, M. M. Adv. Mater. 2010, 22, 813. (28) Almaguer, C.; Cheng, W.; Nolder, C.; Patton-Vogt, J. J. Biol. Chem. 2004, 279, 31937. (29) Sun, H.; He, W.; Zong, C.; Lu, L. ACS Appl. Mater. Interfaces 2013, 5, 2261. (30) Wu, H.; Geng, J.; Ge, H.; Guo, Z.; Wang, Y.; Zheng, G. Adv. Energy Mater. 2016, 6, 1600794. (31) Titirici, M.-M.; Antonietti, M.; Thomas, A. Chem. Mater. 2006, 18, 3808. (32) Virto, R.; Manas, P.; Alvarez, I.; Condon, S.; Raso, J. Appl. Environ. Microbiol. 2005, 71, 5022. (33) Ni, D.; Wang, L.; Sun, Y.; Guan, Z.; Yang, S.; Zhou, K. Angew. Chem., Int. Ed. 2010, 49, 4223. (34) Chen, B.; Zhou, D.; Zhu, L. Environ. Sci. Technol. 2008, 42, 5137. (35) Yu, C. C.; Ramanathan, S.; Oyama, S. T. J. Catal. 1998, 173, 1. (36) Popczun, E. J.; McKone, J. R.; Read, C. G.; Biacchi, A. J.; Wiltrout, A. M.; Lewis, N. S.; Schaak, R. E. J. Am. Chem. Soc. 2013, 135, 9267. (37) Callejas, J. F.; Read, C. G.; Popczun, E. J.; McEnaney, J. M.; Schaak, R. E. Chem. Mater. 2015, 27, 3769. (38) Morales-Guio, C. G.; Stern, L.-A.; Hu, X. Chem. Soc. Rev. 2014, 43, 6555. (39) Zou, X.; Zhang, Y. Chem. Soc. Rev. 2015, 44, 5148. (40) McCrory, C. C. L.; Jung, S.; Ferrer, I. M.; Chatman, S. M.; Peters, J. C.; Jaramillo, T. F. J. Am. Chem. Soc. 2015, 137, 4347. (41) Long, X.; Li, G.; Wang, Z.; Zhu, H.; Zhang, T.; Xiao, S.; Guo, W.; Yang, S. J. Am. Chem. Soc. 2015, 137, 11900. (42) Yu, L.; Zhou, H.; Sun, J.; Qin, F.; Yu, F.; Bao, J.; Yu, Y.; Chen, S.; Ren, Z. Energy Environ. Sci. 2017, DOI: 10.1039/C7EE01571B. (43) Hou, H.; Peng, Q.; Zhang, S.; Guo, Q.; Xie, Y. Eur. J. Inorg. Chem. 2005, 2005, 2625.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b06123. Experimental details; TEM and SEM images and XRD patterns of the yeast cells; STEM image and XPS of Co2P-C; TEM, HRTEM, and SEM images and XRD of p-Co2P-C, FTIR of the carbon microspheres; HRTEM of the metal phosphides; and tables summarizing the Co content of each Co2P-C sample and the Co-based electrocatalysts for the HER (Figures S1−S11 and Tables S1 and S2) (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*[email protected] *[email protected] *[email protected] ORCID

Jin-Song Hu: 0000-0002-6268-0959 Li-Jun Wan: 0000-0002-0656-0936 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors acknowledge funding support from the National Natural Science Foundation of China (Grant No. 51672282 21373238), the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA09010101), and the major State Basic Research Program of China (973 program: 2013CB934000).

■

REFERENCES

(1) Luo, J. S.; Im, J. H.; Mayer, M. T.; Schreier, M.; Nazeeruddin, M. K.; Park, N. G.; Tilley, S. D.; Fan, H. J.; Gratzel, M. Science 2014, 345, 1593. (2) Turner, J. A. Science 2004, 305, 972. (3) Lu, Q.; Yu, Y.; Ma, Q.; Chen, B.; Zhang, H. Adv. Mater. 2016, 28, 1917. (4) Ma, G.; Li, S.; Zhang, W.; Yang, Z.; Liu, S.; Fan, X.; Chen, F.; Tian, Y.; Zhang, W.; Yang, S.; Li, M. Angew. Chem., Int. Ed. 2016, 55, 3667. (5) Durst, J.; Simon, C.; Hasché, F.; Gasteiger, H. A. J. Electrochem. Soc. 2015, 162, F190. (6) Zalineeva, A.; Baranton, S.; Coutanceau, C.; Jerkiewicz, G. Sci. Adv. 2017, 3, e1600542. (7) Sheng, W.; Zhuang, Z.; Gao, M.; Zheng, J.; Chen, J. G.; Yan, Y. Nat. Commun. 2015, 6, 5848. (8) Kyriakou, G.; Boucher, M. B.; Jewell, A. D.; Lewis, E. A.; Lawton, T. J.; Baber, A. E.; Tierney, H. L.; Flytzani-Stephanopoulos, M.; Sykes, E. H. Science 2012, 335, 1209. (9) Shi, Y.; Zhang, B. Chem. Soc. Rev. 2016, 45, 1529. (10) Zeng, M.; Li, Y. J. Mater. Chem. A 2015, 3, 14942. (11) Aetukuri, N. B.; Kitajima, S.; Jung, E.; Thompson, L. E.; Virwani, K.; Reich, M.-L.; Kunze, M.; Schneider, M.; Schmidbauer, W.; Wilcke, W. W.; Bethune, D. S.; Scott, J. C.; Miller, R. D.; Kim, H.-C. Adv. Energy Mater. 2015, 5, 1500265. (12) Zhang, Q.; Lu, H.; Zhong, H.; Yan, X.; Ouyang, C.; Zhang, L. J. Mater. Chem. A 2015, 3, 13706. 11253

DOI: 10.1021/jacs.7b06123 J. Am. Chem. Soc. 2017, 139, 11248−11253