Layer-by-Layer Molecular Assemblies for Dye-Sensitized

15 Aug 2017 - Degao Wang , Seth L. Marquard , Ludovic Troian-Gautier , Matthew V. Sheridan ... Charles E. Creissen , Julien Warnan , Erwin Reisner...
0 downloads 0 Views 2MB Size
Download PDF
Article Cite This: J. Am. Chem. Soc. 2017, 139, 14518-14525

pubs.acs.org/JACS

Layer-by-Layer Molecular Assemblies for Dye-Sensitized Photoelectrosynthesis Cells Prepared by Atomic Layer Deposition Degao Wang,† Matthew V. Sheridan,† Bing Shan,† Byron H. Farnum,†,⊥ Seth L. Marquard,† Benjamin D. Sherman,†,∥ Michael S. Eberhart,† Animesh Nayak,† Christopher J. Dares,‡ Atanu K. Das,§ R. Morris Bullock,§ and Thomas J. Meyer*,† †

Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, United States Department of Chemistry and Biochemistry, Florida International University, 11200 SW 8th St, Miami, Florida 33199, United States § Center for Molecular Electrocatalysis, Physical Sciences Division, Pacific Northwest National Laboratory, P.O. Box 999, K2-12, Richland, Washington 99352, United States ‡

S Supporting Information *

ABSTRACT: In a dye sensitized photoelectrosynthesis cell (DSPEC), the relative orientation of the catalyst and chromophore plays an important role in determining the device efficiency. Here we introduce a new, robust atomic layer deposition (ALD) procedure for the preparation of molecular chromophore−catalyst assemblies on wide bandgap semiconductors. In this procedure, solution deposited, phosphonate derivatized metal complexes on metal oxide surfaces are treated with reactive metal reagents in the gas phase by ALD to form an outer metal ion bridging group, which can bind a second phosphonate containing species from solution to establish a R1-PO2-O-M-O-PO2-R2 type surface assembly. With the ALD procedure, assemblies bridged by Al(III), Sn(IV), Ti(IV), or Zr(IV) metal oxide units have been prepared. To evaluate the performance of this new type of surface assembly, intra-assembly electron transfer was investigated by transient absorption spectroscopy, and light-driven water splitting experiments under steady-state illumination were conducted. A SnO2 bridged assembly on SnO2/TiO2 core/shell electrodes undergoes light-driven water oxidation with an incident photon to current efficiency (IPCE) of 17.1% at 440 nm. Light-driven water reduction with a ruthenium trisbipyridine chromophore and molecular Ni(II) catalyst on NiO films was also used to produce H2. Compared to conventional solution-based procedures, the ALD approach offers significant advantages in scope and flexibility for the preparation of stable surface structures.

■

INTRODUCTION Solar fuels production is a promising approach to solar utilization either by reduction of CO2 to hydrocarbon products or by production of hydrogen from water splitting. In a dyesensitized photoelectrosynthesis cell (DSPEC), molecular chromophores and catalysts are combined at semiconductor surfaces to carry out light driven water splitting to form hydrogen and oxygen, 2H2O + 4hν → O2 + 2H2.1,2 The most promising DSPEC approaches make use of TiO2-based semiconductor photoanodes to assist in charge separation, ruthenium complexes for light absorption and water oxidation catalysis, and a dark cathode, such as Pt, to produce H2.3−8 Currently, significant effort in this area focuses on improving charge separation lifetimes at the metal oxide(s) semiconductor,9,10 developing faster catalysts requiring lower overpotentials,11−14 and improving the energy output of the photoanode to allow for unassisted hydrogen production.15−17 An underdeveloped concept of these devices is the relative configuration of the chromophore and catalyst. In Nature, the relative orientation of the proteins and cofactors involved in photosynthesis are critical in determining the efficiency of light © 2017 American Chemical Society

harvesting, charge separation, catalysis, and stability, and these same factors are essential in DSPEC devices. Similar efforts in the related field of dye-sensitized solar cells (DSSCs) have led to improved chromophore designs capable of higher injection efficiencies, longer lived charge separation, and more efficient outer-sphere electron transfer catalysis with a solution mediator.18 The problem is more complicated in a multicomponent device such as a DSPEC where additional challenges exist including the need for four oxidative equivalents to build-up at the catalyst, greater instability of organic and metal−organic chromophores in aqueous solutions, and a much slower inner sphere oxygen-atom transfer reaction involved in water oxidation to form the O···O bond in O2. Progress has been made in this latter area through the advent of molecular water oxidation catalysts based on a family of Ru(II) 2,2′-bipyridine-6,6′-dicarboxylate complexes first described by Sun and co-workers, initiating a step forward in the performance of DSPEC water splitting.19 Received: July 11, 2017 Published: August 15, 2017 14518

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525

Article

Journal of the American Chemical Society The interfacial structure and stability of surface-bound molecular assemblies for light absorption and catalysis remains an important issue for the development of DSPECs. As shown in eq 1, light absorption and subsequent sensitization of the oxide by the excited-state chromophore (Chrom) initiates the process leading to water oxidation catalysis. Following injection, the oxidized chromophore can undergo electron transfer from the catalyst (Cat) (eq 2) or BET (back electron transfer) with the reduced metal oxide, MOx(e−) (eq 3).

Scheme 1. Stepwise ALD Assemblies of Nanoparticle SnO2/ TiO2 or NiO Films of -RuP22+-ALD MOx-Cat, with Water Oxidation Catalyst 1 or Water Reduction Catalyst 2a

MOx |Chrom‐Cat + hν → MOx (e−)|Chrom+‐Cat injection

(1)

MOx (e−)|Chrom+‐Cat → MOx (e−)|Chrom‐Cat+ hole transfer

(2)

MOx (e−)|Chrom+‐Cat → MOx (0)|Chrom 0‐Cat back electron transfer

(3)

In a DSPEC device, undesirable BET can also occur from the reduced oxide support to the oxidized catalyst (eq 4). These recombination pathways are in competition with water oxidation catalysis, compete with the buildup of multiple oxidative equivalents needed at the catalyst, and limit the photocatalytic turnover especially given the overall slow rates of water oxidation observed for catalysts at surfaces.20 A full description of the dynamics of chromophore−catalyst assemblies has been recently reviewed.3

a

shell surfaces for exploring intra-assembly electron transfer and water oxidation, as well as water reduction on NiO nanoparticle films, with suitable molecular catalysts. The structures of the chromophores [Ru(bpy)2(4,4′-PO3H2-bpy)]2+ (RuP2+, bpy = 2,2′-bipyridine) and [Ru(bpy)(4,4′-PO3H2-bpy)2]2+ (RuP22+), water oxidation catalyst [Ru(bda)(4-O(CH2)3P(O3H2)-pyr)2] (1, pyr = pyridine; bda = 2,2′-bipyridine-6,6′-dicarboxylate)), and the water reduction catalyst [NiII(PPh2NPhP2)2]2+ (2, PPh2NPhP2 = bis-(1,5-di(-(C6H4)-CH2-PO3H2))-3,7-diphenyl1,5-diaza-3,7-diphosphacyclooctane)) are shown in the bottom of Scheme 1.32−34

MOx (e−)|Chrom‐Cat+ → MOx (0)|Chrom‐Cat0 back electron transfer

Relevant molecular structures are shown below.

(4)

This work explores a new synthetic strategy for preparing chromophore−catalyst assemblies on oxide surfaces with the goal of preventing BET pathways. Specifically, atomic layer deposition (ALD) is used to deposit a bridging layer between the molecular surface components. Multiple other strategies have been explored for assembling chromophores and catalysts on metal oxide surfaces including coloading, presynthesized assemblies, and electropolymerization.7,9,21−28 These strategies have various strengths and weaknesses. For instance, coloading is a simple technique that provides the most efficient hole transfer process, but it also positions the catalyst in close proximity to the reduced surface enabling more rapid BET. On the other hand, synthesis of molecular assemblies is attractive as it positions the catalyst away from the surface but requires tedious and demanding synthetic procedures. In previous reports, our laboratory has employed “layer-bylayer” approaches for preparing assemblies with multiple redox or light absorbing units on nanocrystalline metal oxide surfaces based on phosphonate-ZrIV ester bridging.4,29−31 This selflimiting procedure employs zirconyl chloride (ZrOCl2) in acidic solution for adding Zr(IV) ester bonds at free phosphonates on the surface. Here we describe a novel layerby-layer procedure based on a similar concept but with ALD for deposition of the metal bridge precursor (Scheme 1). ALD has the advantage of controlling both the choice of metal and thickness or length of the metal bridge. It also allows for the introduction of metals without suitable solution-based precursors for use as bridging linkers. We describe here the preparation and characterization of chromophore−catalyst layer-by-layer assemblies on nanoparticle SnO2/TiO2 core/

■

EXPERIMENTAL SECTION

General. All chemicals were purchased from Sigma-Aldrich and used as received. The molecular complexes were prepared as previously reported.32−34 Fluorine-doped tin oxide (FTO) was purchased from Hartford Glass with a sheet resistance of 15 Ω/sq. TiO2 films and nano-ITO films were made following the reported recipes.35−37 SnO2/TiO2 core/shell structures and NiO films were fabricated as reported in the literature.9,38 Metal precursors in the ALD experiments are shown in Scheme 2. Electrochemical and photoelectrochemical experiments were performed using either a CH Instruments 660D potentiostat or a CH Instruments 760E bipotentiostat. A THORLABS HPLS 30-04 light source was used to provide white light illumination. For all indicated experiments using 100 mW cm−2 white light illumination, the electrochemical cell was positioned an appropriate distance from

Scheme 2. Metal Precursors Used for ALD: (left) Tetrakis(Dimethylamino)M (M = Sn, Ti, or Zr) and (right) Trimethylaluminum

14519

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525

Article

Journal of the American Chemical Society the light source to receive the indicated light intensity as measured with a photodiode (Newport), and a 400 nm cutoff filter (Newport) was used to prevent direct bandgap excitation of the semiconductor layer. X-ray photoelectron spectroscopy (XPS) studies were performed with a Kratos Axis Ultra DLD X-ray photoelectron spectrometer. Detailed experimental procedures for the ALD film preparation, transient absorption, incident photon to current efficiency, and collector−generator measurements are described in the Supporting Information. Assembly Formation. The structures of the molecular assemblies are shown in Figure S1. Bilayers were prepared by immersing metal oxide mesoporous films in solutions of the dyes at pH 1, 0.5 mM in 0.1 M HClO4. The RuP2+ and RuP22+ dyes adsorb readily on metal oxide films to form closely packed dye monolayers.29,34 In the experiments described here, assemblies were prepared on fluorine-doped tin oxide (FTO) slides with 2.3 μm nanoparticle films of Sn(IV)-doped indium tin oxide (nano-ITO), TiO2, or SnO2/TiO2 core/shell electrodes. Following addition of the chromophore, the modified metal oxide slides were placed in the ALD chamber for addition of the metal oxide precursor. An atomic layer deposition system (Savannah S200, Cambridge Nanotech) was used for preparation of all samples in this study. The ALD metal precursors are shown above. After ALD deposition, the slides were immersed in solutions containing chromophore, WOC (1 mM in MeOH), or hydrogen evolution catalyst (0.5 mM in MeOH) for 16 h to complete the assembly formation. Solution deposition of Zr (denoted ZrIV in the results here) was accomplished by replacing the ALD step with immersion of the slide in 2 mM Zr(O)Cl2 solution in 0.1 M HClO4 for 16 h. ALD experiments were performed at 150 °C with the preformed complexes used known to be stable at this temperature.39 During the exposure mode, the metal−amide precursor was dispersed in the chamber headspace above the chromophore-bound films under vacuum. The free phosphonic acid groups at the chromophore react with the metal-amide precursor (M(NR2)n) releasing an amine (HNR2) to the headspace to form a R-PO2-O-M-(NR2)n−x bond. The phosphonic acid groups are reactive due to the acidity of the −PO3H2 groups (pKa ≈ 2.15); the amide precursors are known to react readily with hydroxylated surfaces and water.40 In the ALD reactor, the reactions are terminated following purging of the headspace to remove the metal precursor and free amines and subsequent addition of water to form new −OH bonds at the phosphonate−M ester. The ALD process is described in eqs 5 and 6 for RuP22+ at the surface (Ru is the chromophore fragment attached to the surface: [Ru(4,4′-OP(O)2-MOx-bpy)(bpy)]2+).

Figure 1. Absorption spectra of dry 2.3 μm films following sequential loading of the assemblies to give, nano-ITO|-RuP22+-ALD MOxRuP22+.

TiO2|-RuP22+-ALD ZrO2, compared to the solution prepared, TiO2|-RuP2+-ZrIV, gave a Zr/Ru ratio of ca. 1:1, consistent with formation of the metal bridge (after correcting for ZrO2 deposited at free hydroxyl sites at the surface). The elemental compositions of Ru and P in the assembly TiO2|-RuP22+-ALD ZrO2-RuP22+ were also double those for TiO2|-RuP22+ with a constant P and Ru ratio of 4:1 The results of UV−visible spectrophotometric studies gave additional evidence for sequential layer-by-layer formation at the oxide surface. The spectra in Figure 1 show that addition of the second MLCT chromophore results in a doubling of the MLCT absorption at 450 nm as expected for formation for the surface-bound, metal oxide-bridged assemblies. Results for the analogous Al2O3 assembly are not shown because of the instability of the assembly under acidic conditions. As a control, when nano-ITO|-RuP2+ was treated in a single ALD deposition cycle and immersed in RuP22+ in 0.1 M HClO4 for extended periods, there was no evidence of further addition of chromophore to the surface. Spectrophotometric measurements on the prepared films revealed that a single ALD cycle was sufficient to provide complete ALD coverage. Nevertheless, the ALD procedure allows for control of the thickness of the MOx linkage by employing multiple pump-purge cycles of the metal precursor and water into the headspace. The UV−visible results of a series of experiments are presented in Figure S4 in which different slides were cycled through a series of sequential ALD steps to give the extended assemblies, nano-ITO|-RuP22+{(ALD) MOx)}y-RuP22+ (y = 1−5). However, there are limitations in the number of cycles; increasing the bridge length of the {(ALD) MOx)}y unit from 1 to 5 cycles resulted in nearly constant doubling of the UV−visible spectrum, however, increasing the number of cycles to 10 results in a decrease in absorbance of ∼20% for the second layer of chromophore. The decrease in the MLCT band with 10 cycles compared to 1−5 cycles is presumably caused by occlusion of the internal cavities of the film and thereby an effective decrease in the surface area for deposition of the outer dye layer. Assembly Dynamics. Photoinduced electron transfer dynamics have been reported in related, solution-prepared bridged chromophore−catalyst assemblies using ZrIV on nano-

Ru(4, 4′‐OP(OH)2 ‐bpy) + M(NMe2)n → Ru(4, 4′‐OP(O)2 ‐M(NMe2)n − x ‐bpy) + x HNMe2

(5)

Ru(4, 4′‐OP(O)2 ‐M(NMe2)n − x ‐bpy) + y H 2O → Ru(4, 4′‐OP(O2 )‐M‐(OH)y −‐ bpy) + y HNMe2

(6)

Surface Properties. Evaluation of surface structures was performed using attenuated total reflectance-infrared measurements (ATR-IR) using a BRUKER ALPHA. Atom composition and atomic ratios were evaluated by X-ray photoelectron spectra (XPS). Results for the assemblies are shown in Figure S3. UV−visible measurements were also used to monitor assembly formation by using an Agilent 8453 UV/visible photodiode array spectrophotometer. Results are shown in Figure 1 for nano-ITO|-RuP22+-ALD MOx-RuP22+ (MOx is TiO2, ZrO2, or SnO2).

■

RESULTS AND DISCUSSION Assembly Formation. In comparing ALD assembly ATRIR spectra with films of TiO2|-RuP2+, the discernible, nonsurface-bound P−OH stretch disappears at 935 cm−1 consistent with phosphonate coordination and ZrO2, TiO2, or SnO2 bridges (Figure S2).4,31 From XPS measurements of the layer-by-layer surfaces, formation of the ALD ZrO2 bridge in 14520

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525

Article

Journal of the American Chemical Society ITO and related oxide surfaces.29,41−43 Here, we have explored the photophysical properties of the assemblies -RuP22+-ALD MOx-DA on 2.3 μm thick, 15 nm particle nano-ITO film electrodes on FTO with nanosecond transient absorption measurements. In these experiments, the dianiline derivative, N,N,N′,N′-((CH2)3PO3H2)4-4,4′-dianiline (DA), with E(DA+/0) = 0.49 V vs Ag/AgCl, was added to the assembly as a one-electron donor.10 The oxidized form of the donor, -DA+•, features an intense absorption maximum at 450 nm, which was monitored in the transient absorption experiments.29 The sequence of reactions that occur in the assemblies following MLCT excitation of the -RuP22+-chromophore are shown in Scheme 3. Kinetic traces following 488 nm MLCT

citation of nano-ITO|-RuP22+-ALD MOx-DA, is followed by rapid injection, eq 7 in Scheme 3. However, in the assemblies, the TA results show that BET to nano-ITO, eq 9, also occurs on the sub-microsecond time scale. In the transient kinetics, both reactions occur on the approximately microsecond time scale and contribute to the initial kinetics at 450 nm shown in Figure 2. Under our conditions, kinetic analysis of the initial data set in Figure 2 is kinetically complex because of kinetic overlap between back electron transfer from nano-ITO(e−) to -RuP23+ or to DA+•. It was possible to analyze the second part of the decay traces in Figure 2, in which back electron transfer occurs from nano-ITO(e−) to -DA+•, eq 10, by application of Kolrausch−Williams−Watts (KWW) kinetics, eq 11.29,45 In this analysis, which assumes a distribution of reacting sites, an average rate constant for the distribution, ⟨k⟩, is defined by eq 12, with β the width of the distribution.

Scheme 3. Light-Induced Electron Transfer in the Assembly Nano-ITO|-RuP22+-ALD MOx-DA

ΔA(t ) = ΔA exp( −(kt )β )

(11)

⟨k⟩ = kβ /Γ(1/β)

(12) +•

Fitting the decay traces for DA , formed during the initial part of the decay traces in Figure 2, to data acquired past ∼5 μs gave lifetimes for the interfacial redox separated states, nanoITO(e−)|-RuP22+-ALD MOx-DA+•. These results are summarized in Table S1.29,45 Based on the data, average rates for back electron transfer from the reduced electrode surface to -DA+• did vary with the bridge in the sequence: 0.027 s−1 (ZrO2) < 0.094 s−1 (ZrIV) < 0.12 s−1 (TiO2) < 0.34 s−1 (SnO2). The extended lifetimes in the assemblies and the variations in dynamic behavior through the series are both notable results of the analysis. Comparison of half-lives for the DA absorance decay in the series is in good agreement with the kinetic analysis: ZrIV (1.9 ms), ALD ZrO2 (1.8 ms), ALD TiO2 (2.5 ms), ALD SnO2 (4.8 ms). From these results, it appears that the solution and ALD processes with Zr as the bridge give similar results suggesting a chemical similarity between the ALD procedure and the solution deposition technique. Additionally, improved BET lifetimes are observed with TiO2 and SnO2 bridges. The effect of increasing the number of ALD cycles to increase the bridge length between the chromophore and catalyst was also investigated with ALD TiO2 assemblies. The results showed that an increasing number of ALD cycles led to decreased rates for BET between oxidized DA and nano-ITO (Figure S6 and Table S2). However, the increase in separation distance leads to lower injection and hole transfer efficiencies as the number of ALD cycles was increased. The results with ALD TiO2 and ALD SnO2 emphasize the advantages of the ALD strategy in tuning and controlling interfacial recombination kinetics and allowing for metal bridges to be used that otherwise could not be as there is no solution counterpart for introducing TiO2 or SnO2 as the bridge. The results obtained in this study highlight a key point in the assembly strategy: with systematic variations in the bridge, it is possible to significantly influence and exploit the internal electronic properties of the assemblies with variations in the time scale for back electron transfer by a factor of ∼10 between ALD ZrO2 and ALD SnO2. Light-Driven Water Oxidation. Water oxidation was also explored in ALD assemblies built atop SnO2/TiO2 core/shell electrodes. The value of core/shell electrodes, based on thin outer layers of TiO2, in promoting long-lived electron−hole

excitation in pH 4.65 acetate buffer solutions with monitoring at 450 nm are shown in Figure S5, and time-dependent spectra are in Figure 2. Analysis of the kinetics data for the series of assemblies reveals an important role for the ALD metal oxide bridge in the assembled structures that affects the rate of BET. Consistent with earlier observations on the same time scale for the assembly, nano-ITO|-RuP2+,41,44 initial MLCT ex-

Figure 2. Time-dependent transient absorption spectral changes at 450 nm following excitation of the assemblies nano-ITO|-RuP22+-ALD MOx-DA (M is Zr, Sn, Ti, or ZrIV) at 488 nm in degassed pH 4.65 acetate buffer with 0.4 M NaClO4 supporting electrolyte under a bias of 0.2 V vs Ag/AgCl. The optical density changes were normalized relative to the absorption maxima. 14521

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525

Article

Journal of the American Chemical Society separation critical in water oxidation is well documented.22 In fabricating the electrodes, core/shell structures consisting of 15 nm nanoparticles and 4 μm films of SnO2 were coated by 4.5 nm layers of TiO2 by ALD. In the final electrodes, FTO|SnO2/ TiO2|-RuP22+-ALD MOx-1, the water oxidation catalyst [Ru(bda)(4-O(CH2)3P(O3H2)-pyr)2], 1, replaced DA used in the photophysical experiments as the electron donor. Surface loaded versions of related assemblies have been studied previously in water oxidation catalysis.34 The surface content of the electrodes was evaluated first by electrochemical monitoring following the ALD procedure and soaking of the catalyst. From the data in Figure S7, a reversible Ru(III/II) wave appears for the surface-bound RuP3+/2+ couple ca. 1.2 V versus Ag/AgCl. For the catalyst, a -RuIII/II couple appears at ∼0.5 V and a kinetically distorted wave is apparent at ∼0.8 V for the -RuIV/III couple. Further oxidation to Ru(V) and water oxidation occur at potentials past 1.0 V in good agreement with previous results.19,20,46 Based on peak current comparisons between the RuP3+/2+ wave and the Ru(III/II) catalyst wave in the assembly, the relative concentrations of chromophore and catalyst in the assemblies were ca. 1:2 for chromophore to catalyst suggesting each of the free phosphonate sites at RuP22+ is occupied by a catalyst. In the water oxidation experiments, a two compartment cell with a Nafion membrane was used in a three electrode configuration with Ag/AgCl as the reference electrode and a Pt mesh counter electrode. The experiments were carried out under N2 at pH = 4.65 in a 0.1 M aqueous acetic acid/sodium acetate buffers in 0.4 M NaClO4 with a 100 mW/cm2 white light source (400 nm cutoff filter) to simulate the intensity of 1 sun. Based on the photocurrent responses in Figure 3, significant variations between the ALD metal oxide bridges are observed in

procedure by dipping slides into a zirconyl chloride, Zr(O)Cl2, solutions (2 mM in pH 1, 0.1 M HClO4). As shown by the data in Figure S10a, the results obtained were comparable to the ALD assembly with ZrO2 as the bridge. Incident photon to current efficiencies (IPCE) were also evaluated with the results shown in Figure 4. As expected, the

Figure 4. IPCE results for FTO|SnO2/TiO2|-RuP22+-ALD MOx-1, at an applied bias of 0.2 V versus Ag/AgCl at pH = 4.65 0.1 M in acetate, 0.4 M NaClO4. A 400 nm cutoff filter was used to mimic the conditions shown in Figure 2.

IPCEs closely overlap the spectral distribution of the MLCT absorption of RuP22+. The maximum IPCE value observed was for the electrode, FTO|SnO2/TiO2|-RuP22+-ALD SnO2-1, which reached 17.1% at the MLCT wavelength maximum at 440 nm. Also included in the comparison are data for the chemically derivatized assembly, FTO|SnO2/TiO2|-RuP22+ZrIV-1. The latter closely matched the ALD assembly with the ZrO2 metal precursor. These results are in excellent agreement with photophysical data reported earlier for assemblies with DA as the electron donor. Critical in the light-driven water oxidation scheme is the appearance of long-lived oxidative equivalents at the catalyst for initiating rate-limiting O···O bond formation. The influence of longer lived redox separation leading to improved performance is clearly visible in the DSPEC data evaluated here. The impact of bridge length on the assemblies FTO|SnO2/ TiO2|-RuP22+-{(ALD) MOx}y-1 was also investigated. Based on the results in Figure S9, a single ALD cycle on the assembly, FTO|SnO2/TiO2|-RuP22+-(ALD) MOx-1, maximized photocurrents. Although, a further increase in bridge length resulted in increased BET lifetimes (Figure S6), there was an efficiency decrease observed with increasing ALD cycles likely due to a more challenging barrier to hole transfer.24 To test the stability and efficiency of the electrodes for O2 generation, collector−generator (C-G) experiments were carried out by using a dual electrode apparatus that has been described earlier.39,47,48 In these experiments, O2 generated at a DSPEC photoanode was monitored at a fluorine-doped tin oxide (FTO) collector electrode with the collector electrode separated by 1 mm from the generator. Using a bipotentiostat allows the photoanode and oxygen reduction electrode to be under separate applied biases.

Figure 3. Current−time (I−t) traces over 10 s dark−light cycles for water oxidation by the electrodes, FTO|SnO2/TiO2|-RuP22+-ALD MOx-1, at an applied bias of 0.2 V versus Ag/AgCl; pH = 4.65, 0.1 M acetate, 0.4 M NaClO4. 1 is the catalyst [Ru(bda)(4-O(CH2)3P(O3H2)-pyr)2].

the assemblies FTO|SnO2/TiO2|-RuP22+-ALD MOx-1 with maximum currents observed for -TiO2- and -SnO2- as bridges. For the other oxides, the maximum photocurrent decreased by ∼40% for -ZrO2- and by ∼90% for -Al2O3- compared to the maximum value at -SnO2-. As an internal reference to ALD ZrO2, the same assembly was prepared by the solution 14522

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525

Article

Journal of the American Chemical Society

preparation of the electrodes, 20 nm nanostructured 2 μm thick films of NiO on FTO50,51 were modified by adding the proton reduction catalyst [NiII(PPh2NPhP2)2]2+, 2, (Figure S1). The ALD prepared electrodes, FTO|NiO|-RuP22+-ALD MOx-2, were prepared by initially coating the NiO surface with RuP22+ followed by stepwise ALD addition of a MOx linker and addition of the Ni(II) complex from solution. Photoelectrochemical measurements were carried out under the same conditions as for water oxidation at the photoanode but at an applied bias of −0.2 V vs Ag/AgCl. As shown by the photocurrent traces in Figure 6, the photocurrents are relatively low because of competitive back electron transfer at the interface.52

Short 10 min illumination periods with a white light source (100 mW cm−2 and 400 nm filter) resulted in the photocurrent responses shown in Figure 5. At the end of a photolysis cycle,

Figure 6. Current−time (I−t) traces over 10 s dark−light cycles for FTO|NiO-RuP22+-ALD MOx-2 electrodes at an applied bias of −0.2 V versus Ag/AgCl: pH = 4.65 0.1 M acetate, 0.4 M NaClO4 under a 100 mW cm−2 white light source with a 400 nm filter. The ZrIV assembly was prepared from Zr(O)Cl2 solutions 2 mM in 0.1 M HClO4.

Figure 5. (a) O2 measurements for water oxidation from (black) FTO| SnO2/TiO2|-RuP22+-ALD SnO2-1 on 0.5 cm2 slides illuminated with 100 mW cm−2 white light with a 400 nm cutoff filter from 30 to 630 s at a bias of 0.2 V versus Ag/AgCl. The current−time response in red is for an O2 sensing electrode, 1 mm from the photoanode biased at −0.85 V versus Ag/AgCl; the experiment was performed in 0.1 M acetic acid/acetate buffer at pH 4.65 in 0.4 M NaClO4. (b) The integrated current over time (charge) obtained in the experiment used with eq 13 to calculate FE.

Based on the data in Figure 6, the highest photocurrents in the series of ALD metal oxide linked assemblies occurred for the assembly NiO|-RuP22+-ALD TiO2-2. Photocurrents reached 40 μA cm−2 at the end of three on−off cycles. A comparable value was obtained for -ALD SnO2- as the bridge in NiO|RuP22+-ALD SnO2-2 with a photocurrent density of 35 μA cm−2. The overall trend in photocurrents with the ALD metal oxide bridge is comparable to those for water oxidation as shown in Figure 3. The short-term stabilities of the photocathodes were also investigated under electrocatalytic conditions with a summary given in Figure S11. In all of the electrodes, the photocurrent response was relatively stable over a photolysis period of 10 min.

the generator current decayed instantaneously while the collector current gradually decays as the remaining O2 between the two electrodes is reductively consumed. The Faradaic efficiencies (FE) for O2 production were calculated from eq 13 with Qcollector and Qgenerator being the total charge passed at the collector and generator electrodes. The constant 0.7 is the experimentally determined collection efficiency for the cell.16,47 (Q collector /Q generator)/0.70 × 100% = FE(%)

■

(13)

CONCLUSION The results presented here extend the work of earlier studies on the development of procedures for the preparation and stabilization of functional assemblies on conducting oxide substrates for use in a DSPEC device. The focus here has been on the synthesis and characterization of layer-by-layer assemblies with an added synthetic feature, the use of atomic layer deposition to form M(IV) or Al(III) metal oxide bridges. The use of ALD opens a new dimension in this area. It offers a controlled environment for preparing complex assemblies with a systematic basis for adding functional and bridging groups in controlled ratios. As a procedure, it offers new and exploitable

As shown by the data in Figure 5, over a 10 min illumination cycle, only a slight decrease in photocurrent was observed due to partial loss of the chromophore as observed in similar systems in earlier reports.49 Based on the results in Figure 5 and Figures S8−S10, the collector−generator experiments show that O2 was produced with Faradaic efficiencies of 80−90% in all of the ALD assemblies consistent with near quantitative production of O2.34 Light-Driven Hydrogen Evolution. ALD assemblies for water reduction to H2 were also explored in mixed assemblies of the phosphonate-derivatized catalyst 2 and RuP22+. In the 14523

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525

Article

Journal of the American Chemical Society

(2) Brennaman, M. K.; Dillon, R. J.; Alibabaei, L.; Gish, M. K.; Dares, C. J.; Ashford, D. L.; House, R. L.; Meyer, G. J.; Papanikolas, J. M.; Meyer, T. J. J. Am. Chem. Soc. 2016, 138, 13085−13102. (3) 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. (4) Hanson, K.; Torelli, D. A.; Vannucci, A. K.; Brennaman, M. K.; Luo, H.; Alibabaei, L.; Song, W.; Ashford, D. L.; Norris, M. R.; Glasson, C. R. K.; Concepcion, J. J.; Meyer, T. J. Angew. Chem., Int. Ed. 2012, 51, 12782−12785. (5) Ding, X.; Gao, Y.; Zhang, L.; Yu, Z.; Liu, J.; Sun, L. ACS Catal. 2014, 4, 2347−2350. (6) Duan, L.; Wang, L.; Li, F.; Li, F.; Sun, L. Acc. Chem. Res. 2015, 48, 2084−2096. (7) Youngblood, W. J.; Lee, S.-H. A.; Kobayashi, Y.; HernandezPagan, E. A.; Hoertz, P. G.; Moore, T. A.; Moore, A. L.; Gust, D.; Mallouk, T. E. J. Am. Chem. Soc. 2009, 131, 926−927. (8) Xu, P.; McCool, N. S.; Mallouk, T. E. Nano Today 2017, 14, 42− 58. (9) 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. (10) Farnum, B. H.; Wee, K.-R.; Meyer, T. J. Nat. Chem. 2016, 8, 845−852. (11) Concepcion, J. J.; Zhong, D. K.; Szalda, D. J.; Muckerman, J. T.; Fujita, E. Chem. Commun. 2015, 51, 4105−4108. (12) Matheu, R.; Ertem, M. Z.; Benet-Buchholz, J.; Coronado, E.; Batista, V. S.; Sala, X.; Llobet, A. J. Am. Chem. Soc. 2015, 137, 10786− 10795. (13) Wang, L.; Duan, L.; Wang, Y.; Ahlquist, M. S. G.; Sun, L. Chem. Commun. 2014, 50, 12947−12950. (14) 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. (15) Sheridan, M. V.; Hill, D. J.; Sherman, B. D.; Wang, D.; Marquard, S. L.; Wee, K.-R.; Cahoon, J. F.; Meyer, T. J. Nano Lett. 2017, 17, 2440−2446. (16) 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, 16745−16753. (17) Li, F.; Fan, K.; Xu, B.; Gabrielsson, E.; Daniel, Q.; Li, L.; Sun, L. J. Am. Chem. Soc. 2015, 137, 9153−9159. (18) Boschloo, G.; Hagfeldt, A. Acc. Chem. Res. 2009, 42, 1819−1826. (19) Duan, L.; Bozoglian, F.; Mandal, S.; Stewart, B.; Privalov, T.; Llobet, A.; Sun, L. Nat. Chem. 2012, 4, 418−423. (20) Sheridan, M. V.; Sherman, B. D.; Wee, K.-R.; Marquard, S. L.; Gold, A. S.; Meyer, T. J. Dalton Trans. 2016, 45, 6324−6328. (21) Wee, K.-R.; Sherman, B. D.; Brennaman, M. K.; Sheridan, M. V.; Nayak, A.; Alibabaei, L.; Meyer, T. J. J. Mater. Chem. A 2016, 4, 2969− 2975. (22) Alibabaei, L.; Sherman, B. D.; Norris, M. R.; Brennaman, M. K.; Meyer, T. J. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 5899−5902. (23) Ashford, D. L.; Lapides, A. M.; Vannucci, A. K.; Hanson, K.; Torelli, D. A.; Harrison, D. P.; Templeton, J. L.; Meyer, T. J. J. Am. Chem. Soc. 2014, 136, 6578−6581. (24) Ashford, D. L.; Song, W.; Concepcion, J. J.; Glasson, C. R. K.; Brennaman, M. K.; Norris, M. R.; Fang, Z.; Templeton, J. L.; Meyer, T. J. J. Am. Chem. Soc. 2012, 134, 19189−19198. (25) Duan, L.; Tong, L.; Xu, Y.; Sun, L. Energy Environ. Sci. 2011, 4, 3296−3313. (26) Gao, Y.; Ding, X.; Liu, J.; Wang, L.; Lu, Z.; Li, L.; Sun, L. J. Am. Chem. Soc. 2013, 135, 4219−4222. (27) Wang, D.; Farnum, B. H.; Sheridan, M. V.; Marquard, S. L.; Sherman, B. D.; Meyer, T. J. ACS Appl. Mater. Interfaces 2017, DOI: 10.1021/acsami.7b00225. (28) Moore, G. F.; Blakemore, J. D.; Milot, R. L.; Hull, J. F.; Song, H.-e.; Cai, L.; Schmuttenmaer, C. A.; Crabtree, R. H.; Brudvig, G. W. Energy Environ. Sci. 2011, 4, 2389−2392.

routes for the preparation of multifunctional molecular assemblies that are otherwise not available with solution assembly formation. Finally, the transient absorption and lightdriven water oxidation results are in excellent agreement for the direct correlation between longer BET lifetimes leading to greater photocurrent efficiency in the DSPEC.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b07216. Details of the preparation of the electrodes studied, UV− visible spectra and electrochemical experiments of the prepared surfaces, additional transient absorption studies, and further details on the O2 production and XPS measurements (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Benjamin D. Sherman: 0000-0001-9571-5065 Michael S. Eberhart: 0000-0002-6261-5727 R. Morris Bullock: 0000-0001-6306-4851 Thomas J. Meyer: 0000-0002-7006-2608 Present Addresses ⊥

B.H.F.: Department of Chemistry and Biochemistry, Auburn University, Auburn, Alabama 36849. ∥ B.D.S.: Department of Chemistry and Biochemistry, Texas Christian University, Fort Worth, Texas, 76129. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was primarily supported by the UNC EFRC Center for Solar Fuels, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Award No. DESC0001011, supporting studies on photoelectrochemical water splitting and atomic layer deposition (D.W, M.V.S, S.L.M, B.D.S, M.S.E, A.N.). Transient absorption experiments were provided by B.S., supported by the U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, under Award DE-SC0015739. Synthesis of the nickel catalyst by A.K.D. was supported as part of the Center for Molecular Electrocatalysis, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences. Pacific Northwest National Laboratory is operated by Battelle for the U.S. Department of Energy. 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 National Science Foundation (Grant ECCS-1542015) as part of the National Nanotechnology Coordinated Infrastructure (NNCI).

■

REFERENCES

(1) Meyer, T. J. Nat. Chem. 2011, 3, 757−758. 14524

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525

Article

Journal of the American Chemical Society (29) Farnum, B. H.; Wee, K.-R.; Meyer, T. J. Nat. Chem. 2016, 8, 845−852. (30) Nayak, A.; Knauf, R. R.; Hanson, K.; Alibabaei, L.; Concepcion, J. J.; Ashford, D. L.; Dempsey, J. L.; Meyer, T. J. Chem. Sci. 2014, 5, 3115−3119. (31) Gross, M. A.; Creissen, C. E.; Orchard, K. L.; Reisner, E. Chem. Sci. 2016, 7, 5537−5546. (32) Kilgore, U. J.; Roberts, J. A. S.; Pool, D. H.; Appel, A. M.; Stewart, M. P.; DuBois, M. R.; Dougherty, W. G.; Kassel, W. S.; Bullock, R. M.; DuBois, D. L. J. Am. Chem. Soc. 2011, 133, 5861−5872. (33) Norris, M. R.; Concepcion, J. J.; Glasson, C. R. K.; Fang, Z.; Lapides, A. M.; Ashford, D. L.; Templeton, J. L.; Meyer, T. J. Inorg. Chem. 2013, 52, 12492−12501. (34) 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 Lett. 2016, 1, 231−236. (35) O’Regan, B.; Moser, J.; Anderson, M.; Graetzel, M. J. Phys. Chem. 1990, 94, 8720−8726. (36) Heimer, T. A.; D’Arcangelis, S. T.; Farzad, F.; Stipkala, J. M.; Meyer, G. J. Inorg. Chem. 1996, 35, 5319−5324. (37) Hoertz, P. G.; Chen, Z.; Kent, C. A.; Meyer, T. J. Inorg. Chem. 2010, 49, 8179−8181. (38) Flynn, C. J.; McCullough, S. M.; Oh, E.; Li, L.; Mercado, C. C.; Farnum, B. H.; Li, W.; Donley, C. L.; You, W.; Nozik, A. J.; McBride, J. R.; Meyer, T. J.; Kanai, Y.; Cahoon, J. F. ACS Appl. Mater. Interfaces 2016, 8, 4754−4761. (39) Lapides, A. M.; Sherman, B. D.; Brennaman, M. K.; Dares, C. J.; Skinner, K. R.; Templeton, J. L.; Meyer, T. J. Chem. Sci. 2015, 6, 6398−6406. (40) Hausmann, D. M.; Kim, E.; Becker, J.; Gordon, R. G. Chem. Mater. 2002, 14, 4350−4358. (41) Farnum, B. H.; Morseth, Z. A.; Lapides, A. M.; Rieth, A. J.; Hoertz, P. G.; Brennaman, M. K.; Papanikolas, J. M.; Meyer, T. J. J. Am. Chem. Soc. 2014, 136, 2208−2211. (42) Farnum, B. H.; Morseth, Z. A.; Brennaman, M. K.; Papanikolas, J. M.; Meyer, T. J. J. Am. Chem. Soc. 2014, 136, 15869−15872. (43) Shan, B.; Das, A. K.; Marquard, S.; Farnum, B. H.; Wang, D.; Bullock, R. M.; Meyer, T. J. Energy Environ. Sci. 2016, 9, 3693−3697. (44) Farnum, B. H.; Morseth, Z. A.; Brennaman, M. K.; Papanikolas, J. M.; Meyer, T. J. J. Phys. Chem. B 2015, 119, 7698−7711. (45) Ardo, S.; Meyer, G. J. Chem. Soc. Rev. 2009, 38, 115−164. (46) 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. (47) Sherman, B. D.; Sheridan, M. V.; Dares, C. J.; Meyer, T. J. Anal. Chem. 2016, 88, 7076−7082. (48) Leem, G.; Sherman, B. D.; Burnett, A. J.; Morseth, Z. A.; Wee, K.-R.; Papanikolas, J. M.; Meyer, T. J.; Schanze, K. S. ACS Energy Lett. 2016, 1, 339−343. (49) Hyde, J. T.; Hanson, K.; Vannucci, A. K.; Lapides, A. M.; Alibabaei, L.; Norris, M. R.; Meyer, T. J.; Harrison, D. P. ACS Appl. Mater. Interfaces 2015, 7, 9554−9562. (50) Natu, G.; Huang, Z.; Ji, Z.; Wu, Y. Langmuir 2012, 28, 950−956. (51) Li, L.; Gibson, E. A.; Qin, P.; Boschloo, G.; Gorlov, M.; Hagfeldt, A.; Sun, L. Adv. Mater. 2010, 22, 1759−1762. (52) Ji, Z.; He, M.; Huang, Z.; Ozkan, U.; Wu, Y. J. Am. Chem. Soc. 2013, 135, 11696−11699.

■

NOTE ADDED AFTER ASAP PUBLICATION This paper was published ASAP on August 30, 2017, with an error in Scheme 2. The corrected version was reposted on September 7, 2017.

14525

DOI: 10.1021/jacs.7b07216 J. Am. Chem. Soc. 2017, 139, 14518−14525