TiO2 Water Splitting

Feb 27, 2017 - All-in-One Derivatized Tandem p+n-Silicon–SnO2/TiO2 Water Splitting .... Energy Diagram for the p/n-GaAs and p-GaInP2-Based Devices f...
18 downloads 0 Views 3MB Size
Download PDF
Letter pubs.acs.org/NanoLett

All-in-One Derivatized Tandem p+n‑Silicon−SnO2/TiO2 Water Splitting Photoelectrochemical Cell Matthew V. Sheridan, David J. Hill, Benjamin D. Sherman, Degao Wang, Seth L. Marquard, Kyung-Ryang Wee, James F. Cahoon, and Thomas J. Meyer* Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, United States S Supporting Information *

ABSTRACT: Mesoporous metal oxide film electrodes consisting of derivatized 5.5 μm thick SnO2 films with an outer 4.3 nm shell of TiO2 added by atomic layer deposition (ALD) have been investigated to explore unbiased water splitting on p, n, and p+n type silicon substrates. Modified electrodes were derivatized by addition of the water oxidation catalyst, [Ru(bda)(4O(CH2)3PO3H2)-pyr)2], 1, (pyr = pyridine; bda = 2,2′-bipyridine-6,6′-dicarboxylate), and chromophore, [Ru(4,4′-PO3H2-bpy) (bpy)2]2+, RuP2+, (bpy = 2,2′-bipyridine), which form 2:1 RuP2+/1 assemblies on the surface. At pH 5.7 in 0.1 M acetate buffer, these electrodes with a fluorine-doped tin oxide (FTO) back contact under ∼1 sun illumination (100 mW/cm2; white light source) perform efficient water oxidation with a photocurrent of 1.5 mA/cm2 with an 88% Faradaic efficiency (FE) for O2 production at an applied bias of 600 mV versus RHE (ACS Energy Lett., 2016, 1, 231−236). The SnO2/TiO2−chromophore− catalyst assembly was integrated with the Si electrodes by a thin layer of titanium followed by an amorphous TiO2 (Ti/a-TiO2) coating as an interconnect. In the integrated electrode, p+n-Si−Ti/a-TiO2−SnO2/TiO2|-2RuP2+/1, the p+n-Si junction provided about 350 mV in added potential to the half cell. In photolysis experiments at pH 5.7 in 0.1 M acetate buffer, bias-free photocurrents approaching 100 μA/cm2 were obtained for water splitting, 2H2O → 2H2 + O2. The FE for water oxidation was 79% with a hydrogen efficiency of ∼100% at the Pt cathode. KEYWORDS: Water splitting, silicon, titanium dioxide, ruthenium, core−shell, photoelectrochemistry Fuel production based on water splitting (2H2O → O2 + 2H2) is a key element in many schemes for diminishing society’s dependence on fossil fuels.1 Silicon photovoltaics have emerged as a common power source for driving water splitting devices by exploiting solar energy.2 Currently, silicon photoanodes are used frequently to drive transparent conducting metal oxide water oxidation catalysts.3−5 In some cases, an additional lightabsorbing element is added with the Si photovoltaic for water splitting.6−9 In these cells, TiO2 acts as a second light absorber of ultraviolet (UV) light and as the catalyst. Direct band gap excitation of TiO2 by UV light can lead to sustained water oxidation, 2H2O → O2 + 4H+ + 4e− but with low efficiency. As a way to improve cell design by using more promising light-absorbing chromophores and molecular water oxidation catalysts, mesoporous metal oxide films have been prepared on silicon semiconductors (Si) with appropriate interconnects or buried heterojunctions. Various p, n, and p+n type silicon photovoltaics have been explored for their ability to enhance © XXXX American Chemical Society

the photovoltages (ΔV) of injected electrons. Here we describe integrated, Si-based semiconductor photovoltaic layers with a chromophore-catalyst based metal oxide photoanode for water oxidation. In this approach, the semiconductor properties of the Si electrodes are integrated with mesoporous, metal oxide films. The films consist of 5.5 μm thick, nanoporous SnO2 film with an outer 4.3 nm shell of TiO2 added by atomic layer deposition (ALD).10,11 Deposition of a chromophore−catalyst layer on the surface of the oxide, combined with the core/shell structure to provide a basis for light-driven water splitting.12 An important element in the design of Si-based devices is protection of the outer Si layer at the solution/interface with a conductive layer.13−17 We describe here integration of the Received: January 9, 2017 Revised: February 3, 2017 Published: February 27, 2017 A

DOI: 10.1021/acs.nanolett.7b00105 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Scheme 2. Proposed Energy Diagram for the Si−Ti/a-TiO2− SnO2/TiO2 Electrode

oxide-based photoanode with a Si photovoltaic based on a titanium-amorphous TiO2 (a-TiO2) overlayer as an interconnect between the Si and mesoporous film. Junctions of this type have been used previously in related applications with the ability of a-TiO2 as a protective heterojunction widely recognized.15 In a seminal study, Khaselev and Turner described a monolithic dual-light absorbing device composed of p/nGaAs and p-GaInP2 semiconductors for water splitting. The device gave H2 as a photoproduct with O2 produced at a dark anode.1 A schematic diagram is shown at the top of Scheme 1. Scheme 1. Energy Diagram for the p/n-GaAs and p-GaInP2Based Devices from Reference 1 (top) and an Illustration of the Monolithic PEC/PV Device (bottom)

interfacial dynamics of molecular chromophores and catalysts at metal oxide electrodes have been recently reviewed.20 The molecular catalyst and chromophore used here were the known water oxidation catalyst (WOC) [Ru(bda)(4-O(CH2)3PO3H2)-pyr)2], (1, pyr = pyridine; bda = 2,2′bipyridine-6,6′-dicarboxylate) and the metal-to-ligand charge transfer (MLCT) chromophore, [Ru(4,4′-PO 3 H 2 -bpy) (bpy)2]2+ (RuP2+; bpy = 2,2′-bipyridine). As shown in Scheme 3, they are known to assemble in a 2:1 ratio on the surface of Scheme 3. Structures of 1 and RuP2+

the core/shell.21 Assemblies of this type on SnO2/TiO2 core/ shell electrodes under white light illumination with 100 mW/ cm2 intensity at pH 5.7 are active toward water splitting with photocurrents of 1.5 mA/cm2 (88% Faradaic efficiency, FE) at an applied bias (∼600 mV vs RHE).22 Catalysts of the type [Ru(bda)(L)2], originally reported by Licheng Sun and coworkers,23 have low overpotentials making them ideal for lightdriven water oxidation.24 Because of the low overpotential for catalyst 1 toward water oxidation, about 1 V versus NHE at pH 7,25 electron transfer from the oxidized form of the dye, RuP3+, with E1/2 = 1.28 V versus NHE for the RuP3+/2+ couple, is favored by >0.2 V.26 The phosphonated versions of the [Ru(bpy)3]2+ dyes have been used extensively as sensitizers in water splitting devices due to their intense MLCT absorptions at about 458 nm for RuP2+ and excited-state potentials that are sufficient for injecting into the conduction band of TiO2.20 Experimental Section. General. All chemicals were purchased from Sigma-Aldrich or Alpha Aesar and used as received unless otherwise noted. The fluorine-doped tin oxide (FTO) electrodes (TEC 15) were purchased from Hartford glass (Hartford, IN). Nafion membranes were purchased from FuelCellsEtc (College Station, TX). Hydrogen measurements

Monolithic devices such as these have shown the highest solarto-hydrogen (STH) efficiencies to date.18,19 The basis for our integrated Si−Ti/a-TiO2−SnO2/TiO2 device is shown in Scheme 2. The diagram shows an energy scheme illustrating the microscopic origins for the operating cell. In the current device, light absorption excites a surface-bound chromophore (A), which injects into the conduction band (CB) of TiO2 (B). Injected electrons are collected at the Ti/aTiO2 back contact where they recombine with holes in Si (C). In a second light-driven step, electrons in the Si are excited by excess light passing through the Ti/a-TiO2−SnO2/TiO2 layers (D). The excited electron is subsequently transferred (E) to a cathode for proton reduction and hydrogen (H2) evolution (F). At the SnO2/TiO2 semiconductor, the ground state of the oxidized chromophore, formed by injection into the TiO2 conduction band, transfers its hole into the lower-lying oxidation states of the catalyst (G) where four oxidative equivalents accumulate for water oxidation catalysis (H). The B

DOI: 10.1021/acs.nanolett.7b00105 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters were performed with an electrochemical sensor with analyte specific response (Unisense, Denmark). Oxygen was monitored with a collector−generator dual electrode technique as described elsewhere.27 A Thor Laboratories HPLC-30-04 plasma light source was used to provide white light illumination (100 mW/cm2) with a 400 nm long pass filter used to avoid any direct bandgap illumination of TiO2. This irradiation level was chosen to approximate 1 sun illumination. A CH Instruments 760E bipotentiostat was used for the electrochemical and photochemical measurements. An electrochemical sensor for H2 measurements (Unisense) was placed in the sealed head space of the cathodic chamber containing the Pt cathode in a custom-built cell.28 For the scanning electron microscopy (SEM), a Hitachi S-4700 Cold Cathode FESEM was used to take the picture of the samples. Film Preparation. p-type and n-type Si ⟨111⟩ wafers (University Wafer, resistivity 1−10 Ω-cm, 381 μm thick) were etched in buffered hydrofluoric acid (BHF Improved, Transene) to remove the native oxide. Metallic back contacts (3 nm Ti/300 nm Pd) were deposited with an electron beam evaporator (Thermionics VE-100) at a base pressure of ∼9 × 10−8 Torr. In order to form the p−n junctions, an n-type wafer with a back contact was etched in BHF and inserted into a home-build chemical vapor deposition system. The substrate was contained in a 1 inch fused silica tube in a hot-wall tube furnace (Lindberg Blue M). The system was evacuated to ∼3 × 10−3 Torr and heated to 800 °C with a flow of 20 sccm H2 (Matheson TriGas, 5N semiconductor grade). Gas flow was changed to 20 sccm of B2H6 (1000 ppm in He, Voltaix) and the tube was pressurized to 20 Torr. The wafer was removed from the reactor after 2 h. On all wafers, a 5 nm Ti top contact was deposited with the electron beam evaporator after oxide etching. Four hundred cycles of TiO2 were deposited with tetrakis(dimethylamido)titanium(IV) (TDMAT) and water to the top contact by atomic layer deposition in an Ultratech Savannah S200 system (200 °C; 500 ms TDMAT pulse, 15 ms water pulse; 10 s hold). Preparation of the DSPEC. Core/shell SnO2/TiO2 electrodes were prepared following earlier procedures.12 Briefly, locally prepared SnO2 paste was doctor bladed onto FTO or silicon electrodes and sintered at 450 °C for 45 min. This resulted in 5.5 μm thick mesoporous electrodes. To form the TiO2 shell, the electrodes underwent 75 cycles of Ti atomic layer deposition (ALD) (Cambridge instruments) with a tetra(dimethlyamido)titanium(IV) (TDMAT) precursor forming ∼4.3 nm thick surface films. The SnO2/TiO2 electrodes were sintered at 450 °C for 30 min following the ALD treatment. An illustration of the Si−SnO2/TiO2 device with a molecular chromophore and catalyst is shown in Scheme 4. Band energy diagrams for the silicon/SnO2/TiO2 electrodes are shown in Scheme 5 under dark and illuminated conditions.1 In the dark, the interconnect allows the Fermi levels of the silicon and the SnO 2 /TiO 2 to equilibrate. 29,30 Under illumination, the majority carrier Fermi level in the Si is pinned to the back contact. In the p-Si, photogenerated electrons migrate to the interface, equilibrating the Si electron quasi Fermi level with the SnO2/TiO2 Fermi level and flattening the Si bands. This results in a loss in potential equal to the difference between the Si electron and hole quasi Fermi levels. In the n-Si, photogenerated holes migrate to the interconnect and recombine with electrons injected into the TiO2, increasing the potential by the difference between the Si electron and hole quasi Fermi levels. In the p+n-Si case,

Scheme 4. Device Illustration and SEM Image of the Device

majority Fermi levels equilibrate in a conventional photophysical fashion,31 leading to the largest increase in potential. Here, the depletion width in the p+-Si is narrow, allowing for the tunneling of holes to the interconnect. Results and Discussion. Dye-Sensitized Solar Cell (DSSC). To evaluate the effect of the Si junctions on SnO2/ TiO2 overlayer films, they were first investigated in 0.2 M LiClO4 acetonitrile solutions in a pseudo-DSSC arrangement. Because of the physical restrictions with Si as the back contact, the DSSC cells, which are commonly back lit, were front lit with the two redox couples physically separated in a two compartment cell with a porous polyethylene frit. The redox mediator couple, I2/I−, was employed as a sacrificial reductant with 0.5 M I− in the photoanode compartment and 0.1 M I2 in the cathode compartment to avoid competitive light absorption by I2/I3−. In an initial series of experiments, the overlayer films of the core−shell were derivatized with the chromophore RuP2+ (Scheme 3). The cells were used to monitor current, i, and, more importantly, the change in the open-circuit potential, Voc, for comparisons across the series of Si junctions. The cell used is illustrated in Scheme 6. A Pt wire was used as the reference and counter electrode with the cell potential determined by the two-electron reduction of iodine/triiodide to iodide with potentials referenced versus the I2 + 2e− → 2I− couple at Pt. In these experiments, the Si photoanodes were overlaid with titanium (Ti) and amorphous TiO2 (a-TiO2). The SnO2/TiO2 (5.5 μm SnO2 layer; 4.3 nm TiO2 shell) core/shell electrode was derivatized with RuP2+ loaded from methanol solutions at C

DOI: 10.1021/acs.nanolett.7b00105 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters Scheme 5. Energy diagram for p-, n-, and p+n-Silicon Electrodes with an Overlayer of SnO2/TiO2 in the Dark and under Illumination

Figure 1. The 1.0 cm2 slides of a SnO2/TiO2 core/shell on p+n-Si (red line (0.01 V/s) and red circles (0.1 V/s)), n-Si (green, 0.1 V/s), and pSi (blue, 0.1 V/s) in 0.5 M LiI with 0.2 M LiClO4 in acetonitrile, potential range: 0 → −1.4 V, Pt counter/reference electrode, V versus Pt/(I2/I−) in 0.1 M I2 in 0.2 M LiClO4 in acetonitrile; 1 sun illumination.

previously.3,4 The p-doped Si sample has the lowest Voc value followed by n-Si and p+n-Si. Notable in the comparison is that FTO as the back contact is more efficient than n-Si. In earlier studies, n-Si has been shown to give appreciable photovoltages (>500 mV).13,15 Dye-Sensitized Photoelectrosynthesis Cell (DSPEC). In the preparation of water splitting electrodes, derivatized oxide film electrodes were prepared by combining of surface-bound catalyst, 1, and chromophore RuP2+. The structures are shown in Scheme 3. Films were coloaded from methanol solutions 1 mM in RuP2+ (24 h) and then soaking in a methanol/1% acetonitrile solution of 0.5 mM 1 (24 h). As observed earlier, under these conditions the surface loading ratio is 2:1 RuP2+/1 with 1 occupying sites that remain open after full loading of the chromophore.21,22 Photoelectrochemical experiments were conducted on the coloaded assemblies in pH 5.7 solutions in 0.5 M NaClO4 and 0.1 M CH3CO2H/ NaCH3CO2 buffer under 1 sun illumination. As in the DSSC experiments in Figure 1, the magnitudes of Voc in Figure 2 decrease in the order: p+n-Si > n-Si > p-Si. In these measurements, the trap states inherent to the semiconductor electrode appear mimicking the energy shifts in Voc.32 Variations in open-circuit potentials were similar to those in the pseudo-DSSC experiments (Figure S2). In Figure 3, a LSV (linear scan voltammogram) of the most promising photoanode, p + n-Si−Ti/a-TiO 2 −SnO 2 /TiO 2 |2RuP2+/1, under 1 sun illumination from Figure 2 is compared to a Pt wire under the same conditions under N2. Using the absolute values of the current at Pt, the intersection between the current levels can be used to infer the maximum output of the cell with 0.13 mA/cm2 based on the data in Figure 3. In Figure 3, a comparison to an FTO back contact with the same assembly illustrates the need for added potential in order to produce current at the Pt electrode. A single Pt electrode was used as both the counter electrode for hydrogen production and the reference electrode to complete the unbiased cell. A current−time trace is shown at the top of Figure 4 under 1 sun illumination. The current density of about 100 μA/cm2 shows that successful unbiased water splitting can occur in the cell. The corresponding H2 trace, obtained with a Unisense Clark-type electrode in the

Scheme 6. Cell Arrangement for Voc Measurements at the Si Junctions for Iodide Oxidation

1 mM for 24 h. Linear sweep voltammograms (LSVs) of the resulting, Si−Ti/a-TiO2−SnO2/TiO2|-RuP2+, electrodes are shown in Figure 1 under 1 sun illumination. The qualitatively low fill factors can be attributed to the high cell resistance (>1000 Ω). An increased fill factor was obtained at slower scan rates; see Figure 1 for p+n-Si at different scan rates. Under the conditions of these experiments, comparable performances were obtained between a SnO2/TiO2 core−shell electrode and TiO2 with a FTO back contact (Figure S1).22 The Voc values for the four electrodes were 0.80 V (p-Si), 0.95 V (n-Si), 1.05 V (FTO-SnO2/TiO2), and 1.40 V (p+n-Si). The variations in Voc values are consistent with values reported D

DOI: 10.1021/acs.nanolett.7b00105 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 2. The 1.0 cm2 slides of Ti/a-TiO2−SnO2/TiO2|-2RuP2+/1 core/shell electrodes on p+n-Si (red), n-Si (green), and p-Si (blue) in 0.5 M NaClO4 with 0.1 M CH3CO2H/NaCH3CO2 buffer at pH 5.7, v = 0.1 V/s (0.5 V → negative potentials), V versus Ag/AgCl, 1 sun illumination.

Figure 3. LSVs of p+n-Si−Ti/a-TIO2−SnO2TiO2|-2RuP2+/1 under 1 sun illumination (red), FTO-SnO2TiO2|-2RuP2+/1 (blue), and a Pt electrode (absolute current, black) in 0.5 M NaClO4 with 0.1 M CH3CO2H/NaCH3CO2 buffer at pH 5.7, v = 0.1 V/s, V versus Ag/ AgCl, under N2.

headspace above the Pt electrode compartment, shown in the middle of the figure gave a Faradaic yield of ∼100% for H2 evolution. The yield for O2 generation was evaluated by using a collector−generator (C-G) cell, Scheme 7. The cell consisted of the photoanode and an FTO electrode facing each other, separated by 1 mm. The two electrodes were operated simultaneously with a bipotentiostat. The FTO collector electrode was used to sense for O2 generated at the photoanode.27 By applying a potential at the collector electrode selective for O2 reduction (i.e., −0.8 V vs Ag/AgCl; well below H+ reduction about −1.2 V vs Ag/AgCl), the FE of water oxidation at the photoanode was evaluated with results shown in Figure 4.33 In control experiments with the potential of the FTO collector maintained at −0.35 V vs Ag/AgCl (a potential more positive than the onset of O2 reduction at FTO), no evidence was observed for the reduction of other possible oxidation products. The red trace at the bottom of Figure 4 is the current measured for oxygen reduction at the collector electrode with

Figure 4. (A) The 1.0 cm2 p+n-Si−Ti/a-TiO2−SnO2/TiO2|-2RuP2+/ 1, electrode with 1 sun illumination from t = 0−30 min, Eapp = 0 V versus Pt. (B) H2 trace obtained during the electrolysis in the top figure. (C) C-G cell with 1 sun illumination from 60−660 s.

the photocurrent from water oxidation at the photoanode shown in the black trace. Integration of the current at the two electrodes with application of a 70% collection efficiency for the cell,28 provided an FE for O2 of 79%. For comparison, an FTOSnO2/TiO2|-2RuP2+/1 photoanode with a bias of 600 mV versus RHE under the same conditions produced O2 with an FE of 89% (Figure S3) and generated a photocurrent of about 1 mA/cm2. Photocurrent differences between the unbiased p+n-Si and biased FTO electrodes are attributed to less than optimal photovoltages from the underlying silicon photovoltaic. In Figure 3, an additional 250 mV is necessary to maximize the photocurrent density of the photoanode under our conditions. E

DOI: 10.1021/acs.nanolett.7b00105 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Award No. DE-SC0001011. This work made use of instrumentation at the Chapel Hill Analytical and Nanofabrication Laboratory (CHANL), a member of the North Carolina Research Triangle Nanotechnology Network (RTNN), which is supported by the NSF (Grant ECCS1542015) as part of the National Nanotechnology Coordinated Infrastructure (NNCI). D.J.H. acknowledges an NSF graduate research fellowship.Also, we express our thanks to CHANL staff, Dr. Amar Kumbhar and Dr. Carrie Donley, for their assistance in collecting SEM data.

Scheme 7. O2 Monitoring in a C-G Cell

■

Conclusions. The monolithic tandem device reported here is an important addition to the field of water splitting solar cells.34 It combines a photoanode for water oxidation with a Sibased photovoltaic at a single electrode. With the p+n-Si−Ti/aTiO2−SnO2/TiO2 electrode described here, the p+n-Si junction adds about 350 mV/e−, sufficient to drive unbiased water splitting. At pH 5.7 (0.1 M acetate buffer), photocurrents approach 100 μA/cm2 with an FE of 79% for O2 and 100% for hydrogen generation at a Pt cathode. We anticipate that in future work, changes in the design of the device to improve the photovoltages at the silicon junctions should greatly enhance photocurrent efficiencies. Our results will hopefully help lead to efficient interfacing of silicon photovoltaics with mesoporous film electrodes for DSSC and DSPEC applications.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.7b00105. Linear sweep voltammetry (LSV), open-circuit potential (OCP), and current versus time (I−t) plots (PDF)

■

REFERENCES

(1) Khaselev, O.; Turner, J. A. Science 1998, 280, 425−427. (2) Sun, K.; Shen, S.; Liang, Y.; Burrows, P. E.; Mao, S. S.; Wang, D. Chem. Rev. 2014, 114, 8662−8719. (3) Chen, L.; Yang, J.; Klaus, S.; Lee, L. J.; Woods-Robinson, R.; Ma, J.; Lum, Y.; Cooper, J. K.; Toma, F. M.; Wang, L.-W.; Sharp, I. D.; Bell, A. T.; Ager, J. W. J. Am. Chem. Soc. 2015, 137, 9595−9603. (4) Sun, K.; McDowell, M. T.; Nielander, A. C.; Hu, S.; Shaner, M. R.; Yang, F.; Brunschwig, B. S.; Lewis, N. S. J. Phys. Chem. Lett. 2015, 6, 592−598. (5) Oh, S.; Oh, J. J. Phys. Chem. C 2016, 120, 133−141. (6) Santinacci, L.; Diouf, M. W.; Barr, M. K. S.; Fabre, B.; Joanny, L.; Gouttefangeas, F.; Loget, G. ACS Appl. Mater. Interfaces 2016, 8, 24810−24818. (7) Shi, J.; Hara, Y.; Sun, C.; Anderson, M. A.; Wang, X. Nano Lett. 2011, 11, 3413−3419. (8) Liu, C.; Tang, J.; Chen, H. M.; Liu, B.; Yang, P. Nano Lett. 2013, 13, 2989−2992. (9) Gao, X.; Wu, S.; Yan, J.; Zhai, X.; Li, X. ACS Appl. Mater. Interfaces 2016, 8, 30072−30078. (10) Knauf, R. R.; Kalanyan, B.; Parsons, G. N.; Dempsey, J. L. J. Phys. Chem. C 2015, 119, 28353−28360. (11) Swierk, J. R.; McCool, N. S.; Nemes, C. T.; Mallouk, T. E.; Schmuttenmaer, C. A. J. Phys. Chem. C 2016, 120, 5940−5948. (12) Sherman, B. D.; Ashford, D. L.; Lapides, A. M.; Sheridan, M. V.; Wee, K.-R.; Meyer, T. J. J. Phys. Chem. Lett. 2015, 6, 3213−3217. (13) Hendricks, O. L.; Scheuermann, A. G.; Schmidt, M.; Hurley, P. K.; McIntyre, P. C.; Chidsey, C. E. D. ACS Appl. Mater. Interfaces 2016, 8, 23763−23773. (14) Hu, S.; Richter, M. H.; Lichterman, M. F.; Beardslee, J.; Mayer, T.; Brunschwig, B. S.; Lewis, N. S. J. Phys. Chem. C 2016, 120, 3117− 3129. (15) McDowell, M. T.; Lichterman, M. F.; Carim, A. I.; Liu, R.; Hu, S.; Brunschwig, B. S.; Lewis, N. S. ACS Appl. Mater. Interfaces 2015, 7, 15189−15199. (16) Hu, S.; Lewis, N. S.; Ager, J. W.; Yang, J.; McKone, J. R.; Strandwitz, N. C. J. Phys. Chem. C 2015, 119, 24201−24228. (17) Yao, T.; Chen, R.; Li, J.; Han, J.; Qin, W.; Wang, H.; Shi, J.; Fan, F.; Li, C. J. Am. Chem. Soc. 2016, 138, 13664−13672. (18) Jia, J.; Seitz, L. C.; Benck, J. D.; Huo, Y.; Chen, Y.; Ng, J. W. D.; Bilir, T.; Harris, J. S.; Jaramillo, T. F. Nat. Commun. 2016, 7, 13237. (19) Kwon, J.; Im, M. J.; Kim, C. U.; Won, S. H.; Kang, S. B.; Kang, S. H.; Choi, I. T.; Kim, H. K.; Kim, I. H.; Park, J. H.; Choi, K. J. Energy Environ. Sci. 2016, 9, 3657−3665. (20) Ashford, D. L.; Gish, M. K.; Vannucci, A. K.; Brennaman, M. K.; Templeton, J. L.; Papanikolas, J. M.; Meyer, T. J. Chem. Rev. 2015, 115, 13006−13049. (21) Sheridan, M. V.; Sherman, B. D.; Marquard, S. L.; Fang, Z.; Ashford, D. L.; Wee, K.-R.; Gold, A. S.; Alibabaei, L.; Rudd, J. A.; Coggins, M. K.; Meyer, T. J. J. Phys. Chem. C 2015, 119, 25420− 25428. (22) Sheridan, M. V.; Sherman, B. D.; Coppo, R. L.; Wang, D.; Marquard, S. L.; Wee, K.-R.; Murakami Iha, N. Y.; Meyer, T. J. ACS Energy Letters 2016, 1, 231−236. (23) Duan, L.; Bozoglian, F.; Mandal, S.; Stewart, B.; Privalov, T.; Llobet, A.; Sun, L. Nat. Chem. 2012, 4, 418−423.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Benjamin D. Sherman: 0000-0001-9571-5065 Thomas J. Meyer: 0000-0002-7006-2608 Present Address †

(K.-R.W.) Department of Chemistry, Daegu University, Gyeongsan 38453, Republic of Korea.

Author Contributions

The manuscript was written through contributions of all authors. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was wholly supported by the UNC EFRC Center for Solar Fuels, an Energy Frontier Research Center funded by F

DOI: 10.1021/acs.nanolett.7b00105 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters (24) Gao, Y.; Zhang, L.; Ding, X.; Sun, L. Phys. Chem. Chem. Phys. 2014, 16, 12008−12013. (25) Song, N.; Concepcion, J. J.; Binstead, R. A.; Rudd, J. A.; Vannucci, A. K.; Dares, C. J.; Coggins, M. K.; Meyer, T. J. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 4935−4940. (26) Ashford, D. L.; Brennaman, M. K.; Brown, R. J.; Keinan, S.; Concepcion, J. J.; Papanikolas, J. M.; Templeton, J. L.; Meyer, T. J. Inorg. Chem. 2015, 54, 460−469. (27) Sherman, B. D.; Sheridan, M. V.; Dares, C. J.; Meyer, T. J. Anal. Chem. 2016, 88, 7076−7082. (28) Ashford, D. L.; Sherman, B. D.; Binstead, R. A.; Templeton, J. L.; Meyer, T. J. Angew. Chem., Int. Ed. 2015, 54, 4778−4781. (29) Lewis, N. S. J. Electrochem. Soc. 1984, 131, 2496−2503. (30) Hu, S.; Shaner, M. R.; Beardslee, J. A.; Lichterman, M.; Brunschwig, B. S.; Lewis, N. S. Science 2014, 344, 1005−1009. (31) Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. S. Chem. Rev. 2010, 110, 6446−6473. (32) Nunzi, F.; Mosconi, E.; Storchi, L.; Ronca, E.; Selloni, A.; Gratzel, M.; De Angelis, F. Energy Environ. Sci. 2013, 6, 1221−1229. (33) Sherman, B. D.; Sheridan, M. V.; Wee, K.-R.; Song, N.; Dares, C. J.; Fang, Z.; Tamaki, Y.; Nayak, A.; Meyer, T. J. Inorg. Chem. 2016, 55, 512−517. (34) Sherman, B. D.; Sheridan, M. V.; Wee, K.-R.; Marquard, S. L.; Wang, D.; Alibabaei, L.; Ashford, D. L.; Meyer, T. J. J. Am. Chem. Soc. 2016, 138, 16745−167453.

G

DOI: 10.1021/acs.nanolett.7b00105 Nano Lett. XXXX, XXX, XXX−XXX