The direct preparation of crystalline indium antimonide (InSb) by the electrodeposition of antimony (Sb) onto indium (In) working electrodes has been demonstrated. When Sb is electrodeposited from dilute aqueous electrolytes containing dissolved Sb2O3, an alloying reaction is possible between Sb and In if any surface oxide films are first thoroughly removed from the electrode. The presented Raman spectra detail the interplay between the formation of crystalline InSb and the accumulation of Sb as either amorphous or crystalline aggregates on the electrode surface as a function of time, temperature, potential, and electrolyte composition. Electron and optical microscopies confirm that under a range of conditions, the preparation of a uniform and phase-pure InSb film is possible. The cumulative results highlight this methodology as a simple yet potent strategy for the synthesis of intermetallic compounds of interest.

Deposition of epitaxial germanium (Ge) thin films on silicon (Si) wafers has been achieved over large areas with aqueous feedstock solutions using electrochemical liquid phase epitaxy (ec-LPE) at low temperatures (T ≤ 90 °C). The ec-LPE method uniquely blends the simplicity and control of traditional electrodeposition with the material quality of melt growth. A new electrochemical cell design based on the compression of a liquid metal electrode into a thin cavity that enables ec-LPE is described. The epitaxial nature, low strain character, and crystallographic defect content of the resultant solid Ge films were analyzed by electron backscatter diffraction, scanning transmission electron microscopy, high resolution X-ray diffraction, and electron channeling contrast imaging. The results here show the first step toward a manufacturing infrastructure for traditional crystalline inorganic semiconductor epifilms that does not require high temperature, gaseous precursors, or complex apparatus.

Macroporous GaP photoelectrodes with wall thicknesses of approximately 500 nm and pore depths ranging from 0 to 45 μm have been prepared from nondegenerately doped single-crystalline n-GaP(100) with a minority-carrier diffusion length of only 110 nm. The photoelectrochemical behaviors of planar and macroporous photoelectrodes have been assessed in nonaqueous regenerative photoelectrochemical cells operated under potentiostatic control and employing dry acetonitrile containing ferrocene/ferrocenium. Enhancements in the short-circuit photocurrents tracked increases in internal quantum yield measurements at long wavelengths for macroporous n-GaP. The observed photocurrent densities and spectral response measurements were consistent with values expected from GaP/liquid heterojunctions controlled by faradaic charge transfer to an outer-sphere, dissolved redox couple. Lower open-circuit photovoltages were observed with macroporous electrodes with increasing pore depth, consistent with distribution of the photocurrent across the entire macropore/solution interface. Fill factors observed under these conditions did not systematically track pore depth in macroporous photoelectrodes. The noted changes in short-circuit photocurrents, open-circuit photovoltages, and fill factors with increasing porosity resulted in more than an order of magnitude improvement in the photoelectrode efficiency of macroporous n-GaP with 45 μm deep pores at 100 mW cm−2 illumination. The presented data show the level and type of enhancement in energy conversion efficiency that high-aspect-ratio electrode architectures can provide carrier-collection-limited materials such as GaP.

High-performance Li ion battery anodes have been made using Ge microwire films containing high levels of residual gallium (Ga). These materials were prepared by the electrochemical liquid–liquid–solid (ec-LLS) process with liquid metal alloy droplets containing Ga, indium (In), and copper (Cu) at T = 80 °C on Cu foil. The as-prepared Ge microwires yielded an initial discharge capacity of 1350 mA h g–1 and retained more than 80% of their original capacity after 80 cycles when subject to discharge–charge cycles at 0.1 C. Ge microwires where the residual metal content had been lowered still showed unusually large capacities and decent capacity retention, albeit less than those of the as-prepared materials. The cumulative data point to the premises that ec-LLS is amenable to making Ge microwires that are innately active as Li+ battery anodes and that Ga incorporation in Ge is beneficial to countering the material stress incurred during Li+ insertion.

P-type macroporous gallium phosphide (GaP) photoelectrodes have been prepared by anodic etching of an undoped, intrinsically n-type GaP(100) wafer and followed by drive-in doping with Zn from conformal ZnO films prepared by atomic layer deposition (ALD). Specifically, 30 nm ALD ZnO films were coated on GaP macroporous films and then annealed at T = 650 °C for various times to diffuse Zn in GaP. Under 100 mW cm–2 white light illumination, the resulting Zn-doped macroporous GaP consistently exhibit strong cathodic photocurrent when measured in aqueous electrolyte containing methyl viologen. Wavelength-dependent photoresponse measurements of the Zn-doped macroporous GaP revealed enhanced collection efficiency at wavelengths longer than 460 nm, indicating that the ALD doping step rendered the entire material p-type and imparted the ability to sustain a strong internal electric field that preferentially drove photogenerated electrons to the GaP/electrolyte interface. Collectively, this work presents a doping strategy with a potentially high degree of controllability for high-aspect ratio III–V materials, where the ZnO ALD film is a practical dopant source for Zn.

Direct electrodeposition of phase-pure crystalline gallium antimonide (GaSb) films has been achieved at low processing temperatures from an aqueous electrolyte. A liquid gallium electrode was used to drive the electroreduction of Sb2O3 dissolved in 0.6 M NaOH. The quality and purity of the resultant films produced depended strongly on the chosen conditions, including temperature, time, and concentration of dissolved Sb2O3. Under select conditions, the direct production of polycrystalline films of GaSb was possible. Raman spectroscopy, powder X-ray diffraction, selected area electron diffraction, scanning electron microscopy, and transmission electron microscopy were separately used to analyze the identity and crystallinity of the electrodeposited films. The cumulative data showed that this electrodeposition process followed the features common both to conventional electrodeposition and melt crystal growth. Accordingly, this electrodeposition process could not be categorized as electrodeposition followed by annealing. Rather, the data implicate this method is akin to a hybrid electrodeposition solution-based crystal growth, where crystalline GaSb films could be grown up to thicknesses of 1 μm in 60 min.

Direct synthesis of crystalline silicon (Si) nanowires at low temperatures has been achieved through an electrochemical liquid–liquid–solid (ec-LLS) process. Liquid metal nanodroplets containing Ga were used as both discrete ultramicroelectrodes and crystal growth seeds for Si nanowires. This new ec-LLS process was performed in propylene carbonate containing SiCl4 at temperatures as low as 60 °C. X-ray diffraction and Raman spectra separately and independently indicated the nanowires were crystalline as prepared. Scanning electron micrographs of Si nanowires grown on both Si(111) and Si(100) substrates further showed that the direction of nanowire growth specifically followed the crystallographic orientation of the underlying substrate, indicating deposition with homoepitaxy. Localized electron diffraction patterns collected from individual Si nanowires in a transmission electron microscope showed the characteristic pattern of the diamond cubic structure of crystalline Si. Additional experiments were performed that indicated the wetting of the electrodeposited Si by the liquid metal influenced the morphology of the resultant nanowire. These cumulative results support the overarching premise that ec-LLS is a unique synthetic method for crystalline Si nanowires at low temperatures.

Chemically modified Si(111) surfaces have been prepared through a series of wet chemical surface treatments that simultaneously show resistance towards surface oxidation, selective reactivity towards chemical reagents, and areal defect densities comparable to unannealed thermal oxides. Specifically, grazing angle attenuated total reflectance infrared and X-ray photoelectron (XP) spectroscopies were used to characterize allyl-, 3,4-methylenedioxybenzene-, or 4-[bis(trimethylsilyl)amino]phenyl-terminated surfaces and the subsequently hydroxylated surfaces. Hydroxylated surfaces were confirmed through reaction with 4-(trifluoromethyl)benzyl bromide and quantified by XP spectroscopy. Contact angle measurements indicated all surfaces remained hydrophilic, even after secondary backfilling with CH3 groups. Surface recombination velocity measurements by way of microwave photoconductivity transients showed the relative defect-character of as-prepared and aged surfaces. The relative merits for each investigated surface type are discussed.

The electrochemical liquid–liquid–solid (ec-LLS) deposition of crystalline germanium (Ge) in a eutectic mixture of liquid gallium (Ga) and indium (In) was analyzed as a function of liquid metal thickness, process temperature, and flux. Through control of reaction parameters, conditions were identified that allow selective nucleation and growth of crystalline Ge at the interface between e-GaIn and a crystalline Si substrate. The crystal growth rates of Ge by ec-LLS as a function of process temperatures were obtained from time-dependent powder X-ray diffraction measurements of crystalline Ge. The driving force, Δμ, for crystal formation in ec-LLS was estimated through analyses of the experimental data in conjunction with predictions from a finite-difference model. The required Δμ for Ge nucleation was tantamount to a supersaturation approximately 102 larger than the equilibrium concentration of Ge in e-GaIn at the investigated temperatures. These points are discussed both in the context of advancing new, low-temperature synthetic methodologies for crystalline semiconductor films and on understanding semiconductor crystal growth more deeply.

This Account describes a new electrochemical synthetic strategy for direct growth of crystalline covalent group IV and III–V semiconductor materials at or near ambient temperature conditions. This strategy, which we call “electrochemical liquid–liquid–solid” (ec-LLS) crystal growth, marries the semiconductor solvation properties of liquid metal melts with the utility and simplicity of conventional electrodeposition. A low-temperature liquid metal (i.e., Hg, Ga, or alloy thereof) acts simultaneously as the source of electrons for the heterogeneous reduction of oxidized semiconductor precursors dissolved in an electrolyte as well as the solvent for dissolution of the zero-valent semiconductor. Supersaturation of the semiconductor in the liquid metal triggers eventual crystal nucleation and growth. In this way, the liquid electrolyte–liquid metal–solid crystal phase boundary strongly influences crystal growth.As a synthetic strategy, ec-LLS has several intrinsic features that are attractive for preparing covalent semiconductor crystals. First, ec-LLS does not require high temperatures, toxic precursors, or high-energy-density semiconductor reagents. This largely simplifies equipment complexity and expense. In practice, ec-LLS can be performed with only a beaker filled with electrolyte and an electrical circuit capable of supplying a defined current (e.g., a battery in series with a resistor). By this same token, ec-LLS is compatible with thermally and chemically sensitive substrates (e.g., plastics) that cannot be used as deposition substrates in conventional syntheses of covalent semiconductors. Second, ec-LLS affords control over a host of crystal shapes and sizes through simple changes in common experimental parameters. As described in detail herein, large and small semiconductor crystals can be grown both homogeneously within a liquid metal electrode and heterogeneously at the interface of a liquid metal electrode and a seed substrate, depending on the particular details chosen for ec-LLS. Third, the rate of introduction of zero-valent materials into the liquid metal is precisely gated with a high degree of resolution by the applied potential/current.The intent of this Account is to summarize the key elements of ec-LLS identified to date, first contextualizing this method with respect to other semiconductor crystal growth methods and then highlighting some unique capabilities of ec-LLS. Specifically, we detail ec-LLS as a platform to prepare Ge and Si crystals from bulk- (∼1 cm3), micro- (∼10–10 cm3), and nano-sized (∼10–16 cm3) liquid metal electrodes in common solvents at low temperature. In addition, we describe our successes in the preparation of more compositionally complex binary covalent III–V semiconductors.

Hybrid organic–inorganic solar cell devices were fabricated utilizing macroporous n-type GaP and poly(3,4-ethylenedioxythiophene):poly(4-styrene sulfonate) (PEDOT:PSS). The high-aspect ratio structures of the macroporous GaP resulted in higher photocurrent and external quantum yield as a function of wavelength. Photocurrent–voltage measurements as a function of light intensity revealed information on the dependence of short-circuit current (Jsc) and open-circuit voltage (Voc) on light intensity. Under 1.0 Sun illumination, hybrid macroporous GaP/PEDOT:PSS devices showed Jsc of 2.34 mA cm−2, Voc of 0.95 V, fill factor of 0.54, and overall efficiency of 1.21%. The extent of the influence of dopant density of GaP on hybrid device performance was probed with current density–voltage measurements. The addition of a gold nanoparticle coating on macroporous GaP prior to PEDOT:PSS coating showed increased device performance, with overall efficiency of 1.81%. Gold-modified planar GaP/PEDOT:PSS showed decreased Jsc and Voc values and lower external quantum yield over all wavelengths.Hybrid organic–inorganic solar cell devices were fabricated utilizing macroporous n-type GaP and poly(3,4-ethylenedioxythiophene):poly(4-styrene sulfonate) (PEDOT:PSS). The high-aspect ratio structures of the macroporous GaP resulted in higher photocurrent and external quantum yield as a function of wavelength.

Functionalization of crystalline gallium phosphide (GaP) (111)B interfaces has been performed through the formation of P–O–R surface bonds. The approach described herein parallels classical Williamson ether synthesis, where hydroxyl groups on etched GaP(111)B surfaces were reacted with halogenated reactants. Grazing angle total internal reflectance infrared spectra showed increased intensities for −CH2– and −CH3 asymmetric and symmetric stretches after reaction with long alkyl halides. Changes in the X-ray photoelectron spectra collected before and after reaction separately corroborated surface attachment to GaP(111)B. Static sessile drop water contact angle measurements for GaP(111)B separately showed increased hydrophobicity following surface modification with long alkyl chains. The surface functionalization reaction rate was increased by the addition of non-nucleophilic bases, consistent with surface deprotonation as the rate-limiting step. Separately, photoelectrochemical measurements conducted before and after reaction with alkyl halides at long wavelengths (λ > 545 nm) showed surface attachment decreased sub-band-gap photocurrents, implying lowered activity of surface traps. Conversely, photoelectrochemical measurements performed after functionalization of p-GaP(111)B with Coomassie Blue sulfonyl chloride showed evidence of persistent sensitized hole injection from the dye into p-GaP.

Crystalline InAs films have been prepared directly at room temperature through a new electrochemically induced alloying method by controllably reducing As2O3 dissolved in an alkaline aqueous solution at an indium (In) foil electrode. Steady-state Raman spectra, transmission electron microscopy, and selected area electron diffraction indicated that the as-prepared films crystallize in the zincblende phase with no further thermal treatments. Cyclic voltammetry measurements, optical images, and steady-state Raman spectra confirmed that a clean oxide-free interface is critical for the successful formation of the binary InAs phase. The salient feature of this work is the use of simple aqueous electrochemistry to simultaneously remove passive metal oxides from the In(s) metal surface while controllably reducing dissolved arsenic oxide at the interface to drive the In–As alloying reaction. Raman spectral mapping data illustrate that the resulting film coverage and homogeneity are a strong function of the formal As2O3 concentration and the duration of the electrodeposition experiment. Potential-dependent in situ Raman spectroscopy was used to implicate the solid-state reaction as the rate-limiting step in InAs film formation over the first 160 min, after which solid-state diffusion dominated the kinetics. The collective results establish a precedent for an alternative synthetic strategy for crystalline InAs thin films that does not require vacuum or sophisticated furnaces, toxic gaseous precursors like arsine, or exotic solvents.

The photoelectrochemical properties of p-type gallium phosphide (GaP) (111)A electrodes before and after a two-step chlorination/Grignard reaction sequence have been assessed. Electrochemical impedance spectroscopy indicated both a change in the flat-band potential in water and decreased sensitivity of the band edge energetics toward pH for GaP(111)A surfaces following modification. Separate stability tests were performed to gauge the susceptibilities of unmodified and CH3-terminated p-GaP(111)A photoelectrodes toward reductive degradation under illumination. The steady-state photoelectrochemical results showed modification of GaP(111)A with −CH3 groups significantly enhanced p-GaP stability. Separately, sub-bandgap photocurrent measurements were collected to assess relative changes in surface states acting as recombination centers. In the absence of any dye, the sub-bandgap photocurrent from trapping/detrapping of charge carriers in surface states was higher for unmodified p-GaP photoelectrodes than for CH3-terminated p-GaP(111)A. Further, sensitized photocurrents of p-GaP photoelectrodes with Brilliant Green were systematically higher after modification with −CH3 groups, indicating a deactivation of a surface recombination pathway after surface modification. Collectively, this work illustrates a rational chemical strategy to modify and augment the pertinent interfacial properties of p-GaP photocathodes in an aqueous photoelectrochemical system.

The functionalization of single crystalline gallium phosphide (GaP) (111)A surfaces with allyl groups has been performed using a sequential chlorine-activation/Grignard reaction process. Increased hydrophobicity following reaction of a GaP(111)A surface with C3H5MgCl was observed through water contact angle measurements. Infrared spectra of GaP(111)A samples after reaction with C3H5MgCl showed the asymmetric C═C and C═C–H modes diagnostic of surface-attached allyl groups. The stability of allyl-terminated GaP(111)A surfaces under ambient and aqueous conditions was investigated. XP spectra of allyl-terminated GaP(111)A highlighted a significant resistance against interfacial oxidation both in air and in water relative to the native interface. Electrochemical impedance spectroscopy indicated a change in the flat-band potential of allyl-terminated GaP(111)A electrodes immersed in water relative to native GaP(111)A surfaces. Further, the flat-band potentials for allyl-terminated electrodes were insensitive to changes in solution pH. The utility of surface-bound allyl groups for covalent secondary functionalization of GaP(111)A interfaces was assessed through three separate reactions: Heck cross-coupling metathesis, hydrosilylation, and electrophilic addition of bromine reactions. Addition of aryl groups across the olefins on allyl-terminated GaP(111)A via Heck cross coupling was performed and confirmed through high-resolution F 1s and C 1s XP spectra and IR spectra. Control experiments with GaP(111)A surfaces functionalized with short alkanes indicated no evidence for metathesis. Hydrosilylation reactions were separately performed. Si 2s XP spectra, in conjunction with infrared spectra, similarly showed secondary evidence of surface functionalization for allyl-terminated GaP(111)A but not for CH3-terminated GaP(111)A surfaces. Similar analyses showed electrophilic addition of Br2 across the terminal olefin on an allyl-terminated GaP(111)A surface after exposure to dilute Br2 solutions in CH2Cl2. The work presented herein establishes a set of secondary reaction strategies utilizing allyl-terminated surfaces to modify chemically protected GaP surfaces.

The substrate-overlayer approach has been used to acquire surface enhanced Raman spectra (SERS) during and after electrochemical atomic layer deposition (ECALD) of CdSe, CdTe, and CdS thin films. The collected data suggest that SERS measurements performed with off-resonance (i.e. far from the surface plasmonic wavelength of the underlying SERS substrate) laser excitation do not introduce perturbations to the ECALD processes. Spectra acquired in this way afford rapid insight on the quality of the semiconductor film during the course of an ECALD process. For example, SERS data are used to highlight ECALD conditions that yield crystalline CdSe and CdS films. In contrast, SERS measurements with short wavelength laser excitation show evidence of photoelectrochemical effects that were not germane to the intended ECALD process. Using the semiconductor films prepared by ECALD, the substrate-overlayer SERS approach also affords analysis of semiconductor surface adsorbates. Specifically, Raman spectra of benzenethiol adsorbed onto CdSe, CdTe, and CdS films are detailed. Spectral shifts in the vibronic features of adsorbate bonding suggest subtle differences in substrate-adsorbate interactions, highlighting the sensitivity of this methodology.

An electrochemical liquid–liquid–solid (ec-LLS) process that yields crystalline silicon at low temperature (80 °C) without any physical or chemical templating agent has been demonstrated. Electroreduction of dissolved SiCl4 in propylene carbonate using a liquid gallium [Ga(l)] pool as the working electrode consistently yielded crystalline Si. X-ray diffraction and electron diffraction data separately indicated that the as-deposited materials were crystalline with the expected patterns for a diamond cubic crystal structure. Scanning and transmission electron microscopies further revealed the as-deposited materials (i.e., with no annealing) to be faceted nanocrystals with diameters in excess of 500 nm. Energy-dispersive X-ray spectra further showed no evidence of any other species within the electrodeposited crystalline Si. Raman spectra separately showed that the electrodeposited films on the Ga(l) electrodes were not composed of amorphous carbon from solvent decomposition. The cumulative data support two primary contentions. First, a liquid-metal electrode can serve simultaneously as both a source of electrons for the heterogeneous reduction of dissolved Si precursor in the electrolyte (i.e., a conventional electrode) and a separate phase (i.e., a solvent) that promotes Si crystal growth. Second, ec-LLS is a process that can be exploited for direct production of crystalline Si at much lower temperatures than ever reported previously. The further prospect of ec-LLS as an electrochemical and non-energy-intensive route for preparing crystalline Si is discussed.

The sensitization of p-GaP by adsorbed CdSe quantum dots has been observed. Nondegenerately doped, planar p-GaP(100) photoelectrodes consistently showed sub-band-gap (>550 nm) photoresponsivity in an aqueous electrolyte containing Eu3+/2+ when CdSe quantum dots (diameters ranging from 3.1 to 4.5 nm) were purposely adsorbed on the surface. Both time-resolved photoluminescence decays and steady-state photoelectrochemical responses supported sensitized hole injection from the CdSe quantum dots into p-GaP. The observation of hole injection in this system stands in contrast to sensitized electron injection seen in other metal oxide/quantum dot material combinations and therefore widens the possible designs for photoelectrochemical energy conversion systems that utilize quantum dots as light-harvesting components.

The solar energy conversion properties of thin Si and GaP nanowire photoelectrodes in photoelectrochemical cells have been examined through sets of finite-element simulations. A discussion describing the motivation behind nanostructured, high aspect ratio semiconductor photoelectrode designs and a brief survey of current experimental results reported for nanostructured semiconductor photoelectrodes in photoelectrochemical cells are presented first. An analysis is then shown that outlines the primary recombination pathways governing the steady-state current-potential behaviors of thin, cylindrical nanowire photoelectrodes, with explicit expressions detailing the differences between planar and cylindrical photoelectrodes arising from the solution of carrier fluxes in planar and cylindrical geometries. Results from finite-element simulations used to model the key features of thin nanowire photoelectrodes under low-level injection conditions are shown that illustrate which recombination pathway(s) is operative under various experimental conditions. Specifically, the respective effects of non-uniform doping, tapering along the length, variation in charge carrier mobilities and lifetimes, changes in nanowire radius, and changes in the density of surface defects on the observable photocurrent-potential responses are reported. These cumulative results serve as guides for future experimental work aimed at improving the attainable solar energy conversion efficiencies of doped semiconductor nanowire photoelectrodes. Lastly, separate simulations that model lightly doped nanowire photoelectrodes under high-level injection conditions are discussed. These results suggest discrete, ohmic-selective contacts may afford a way to circumvent the stringent doping requirements discussed herein for thin nanowire photoelectrodes.

The direct electrodeposition of crystalline germanium (Ge) nanowire film electrodes from an aqueous solution of dissolved GeO2 using discrete ‘flux’ nanoparticles capable of dissolving Ge(s) has been demonstrated. Electrodeposition of Ge at inert electrode substrates decorated with small (<100 nm), discrete indium (In) nanoparticles resulted in crystalline Ge nanowire films with definable nanowire diameters and densities without the need for a physical or chemical template. The Ge nanowires exhibited strong polycrystalline character as-deposited, with approximate crystallite dimensions of 20 nm and a mixed orientation of the crystallites along the length of the nanowire. Energy dispersive spectroscopic elemental mapping of individual Ge nanowires showed that the In nanoparticles remained at the base of each nanowire, indicating good electrical communication between the Ge nanowire and the underlying conductive support. As-deposited Ge nanowire films prepared on Cu supports were used without further processing as Li+ battery anodes. Cycling studies performed at 1 C (1624 mA g–1) indicated the native Ge nanowire films supported stable discharge capacities at the level of 973 mA h g–1, higher than analogous Ge nanowire film electrodes prepared through an energy-intensive vapor–liquid–solid nanowire growth process. The cumulative data show that ec-LLS is a viable method for directly preparing a functional, high-activity nanomaterials-based device component. The work presented here is a step toward the realization of simple processes that make fully functional energy conversion/storage technologies based on crystalline inorganic semiconductors entirely through benchtop, aqueous chemistry and electrochemistry without time- or energy-intensive process steps.

The steady-state photoelectrochemical responses of p-GaP photoelectrodes immersed in aqueous electrolytes and sensitized separately by six triphenylmethane dyes (rose bengal, rhodamine B, crystal violet, ethyl violet, fast green fcf, and brilliant green) have been analyzed. Impedance measurements indicated that these p-GaP(100) photoelectrodes operated under depletion conditions with an electric field of ∼8.5 × 105 V cm–1 at the p-GaP/solution interface. The set of collected wavelength-dependent quantum yield responses were consistent with sensitization occurring specifically from adsorbed triphenylmethane dyes. At high concentrations of dissolved dye, the measured steady-state photocurrent–potential responses collected at sub-bandgap wavelengths suggested unexpectedly high (>0.1) net internal quantum yields for sensitized hole injection. Separate measurements performed with rose bengal adsorbed on p-GaP surfaces pretreated with (NH4)2S verified efficient sensitized hole injection. A modified version of wxAMPS, a finite-difference software package, was utilized to assess key operational features of the sensitized p-GaP photocathodes. The net analysis showed that the high internal quantum yield values inferred from the experimental data were most likely afforded by the internal electric field present within p-GaP, effectively sweeping injected holes away from the interface and minimizing their participation in deleterious pathways that could limit the net collection yield. These simulations defined effective threshold values for the charge carrier mobilities (≥10–6 cm2 V–1 s–1 and ≥10–1 cm2 V–1 s–1 at dopant densities of 1018 and 1013 cm–3, respectively), hole injection rate constants (≥1012 s–1), and surface trap densities (1012 cm–2) needed to attain efficient hole collection with the quality of p-GaP materials used here. The cumulative experimental and modeling data thus provide insight on design strategies for assembling new types of dye-sensitized photocathodes that operate under depletion conditions.

Crystalline GaAs (c-GaAs) has been prepared directly through electroreduction of As2O3 dissolved in an alkaline aqueous solution at a liquid gallium (Ga(l)) electrode at modest temperatures (T ≥ 80 °C). Ga(l) pool electrodes yielded consistent electrochemical behavior, affording repetitive measurements that illustrated the interdependences of applied potential, concentration of dissolved As2O3, and electrodeposition temperature on the quality of the resultant c-GaAs(s). Raman spectra indicated the composition of the resultant film was strongly dependent on both the electrodeposition temperature and dissolved concentration of As2O3 but not to the applied bias. For electrodepositions performed either at room temperature or with high (≥0.01 M) concentrations of dissolved As2O3, Raman spectra of the electrodeposited films were consistent with amorphous As(s). X-ray diffractograms of As(s) films collected after thermal annealing indicated metallurgical alloying occurred only at temperatures in excess of 200 °C. Optical images and Raman spectra separately showed the composition of the as-electrodeposited film in dilute (≤0.001 M) solutions of dissolved As2O3(aq) was pure c-GaAs(s) at much lower temperatures than 200 °C. Diffractograms and transmission electron microscopy performed on as-prepared films confirmed the identity of c-GaAs(s). The collective results thus provide the first clear demonstration of an electrochemical liquid–liquid–solid (ec-LLS) process involving a liquid metal that serves simultaneously as an electrode, a solvent/medium for crystal growth, and a coreactant for the synthesis of a polycrystalline semiconductor. The presented data serve as impetus for the further development of the ec-LLS process as a controllable, simple, and direct route for technologically important optoelectronic materials such as c-GaAs(s).

Photoactive ZnGeP2 nanowires have been prepared by solid-source sublimation chemical vapor deposition using Sn catalysts. Nanowire films with areas >0.5 cm2 on Si(100) and Si(111) substrates were deposited with variable nanowire length and diameter. Transmission electron microscopy (TEM), scanning TEM (STEM), and polarized Raman microscopy indicated nanowires exhibited single-crystal character and compositional homogeneity. Photoelectrochemical measurements performed in an aqueous electrolyte indicated the as-prepared ZnGeP2 nanowires were p-type and capable of passing sustained cathodic photocurrents under white light illumination. The presented results identify a straight-forward approach to the preparation of II–IV–V2 nanowire films with features suitable for optical and photoelectrochemical energy conversion/storage applications.

Crystalline gallium arsenide (GaAs) (111)A and gallium nitride (GaN) (0001) surfaces have been functionalized with alkyl groups via a sequential wet chemical chlorine activation, Grignard reaction process. For GaAs(111)A, etching in HCl in diethyl ether effected both oxide removal and surface-bound Cl. X-ray photoelectron (XP) spectra demonstrated selective surface chlorination after exposure to 2 M HCl in diethyl ether for freshly etched GaAs(111)A but not GaAs(111)B surfaces. GaN(0001) surfaces exposed to PCl5 in chlorobenzene showed reproducible XP spectroscopic evidence for Cl-termination. The Cl-activated GaAs(111)A and GaN(0001) surfaces were both reactive toward alkyl Grignard reagents, with pronounced decreases in detectable Cl signal as measured by XP spectroscopy. Sessile contact angle measurements between water and GaAs(111)A interfaces after various levels of treatment showed that GaAs(111)A surfaces became significantly more hydrophobic following reaction with CnH2n–1MgCl (n = 1, 2, 4, 8, 14, 18). High-resolution As 3d XP spectra taken at various times during prolonged direct exposure to ambient lab air indicated that the resistance of GaAs(111)A to surface oxidation was greatly enhanced after reaction with Grignard reagents. GaAs(111)A surfaces terminated with C18H37 groups were also used in Schottky heterojunctions with Hg. These heterojunctions exhibited better stability over repeated cycling than heterojunctions based on GaAs(111)A modified with C18H37S groups. Raman spectra were separately collected that suggested electronic passivation by surficial Ga–C bonds at GaAs(111)A. Specifically, GaAs(111)A surfaces reacted with alkyl Grignard reagents exhibited Raman signatures comparable to those of samples treated with 10% Na2S in tert-butanol. For GaN(0001), high-resolution C 1s spectra exhibited the characteristic low binding energy shoulder demonstrative of surface Ga–C bonds following reaction with CH3MgCl. In addition, 4-fluorophenyl groups were attached and detected after reaction with C6H4FMgBr, further confirming the susceptibility of Cl-terminated GaN(0001) to surface alkylation. However, the measured hydrophobicities of alkyl-terminated GaAs(111)A and GaN(0001) were markedly distinct, indicating differences in the resultant surface layers. The results presented here, in conjunction with previous studies on GaP, show that atop Ga atoms at these crystallographically related surfaces can be deliberately functionalized and protected through Ga–C surface bonds that do not involve thiol/sulfide chemistry or gas-phase pretreatments.

The photoelectrochemical responses of two archetypal metal–organic frameworks (MOFs), MOF-5 and MOF-177, have been assessed. Films of MOF-5 and MOF-177 were grown on carboxylic-acid-terminated conductive fluorine-doped tin oxide substrates. Separate analyses by powder X-ray diffraction, Raman spectroscopy, and fluorescence spectroscopy collectively indicated these films prepared via a solvothermal method in diethylformamide were free of residual impurities such as ZnO clusters and residual organics. Exposure of these films to white light illumination while immersed in acetonitrile electrolytes elicited measurable photocurrents. Wavelength-dependent analysis of the photoresponses showed that the measured photocurrents were induced by ultraviolet light and that the spectral response profiles followed closely the light absorption profiles of each respective material. Attenuation of the induced photocurrents was noted after prolonged ultraviolet light illumination and/or exposure of the films to H2O(l), indicating that the observed photoresponse properties are directly related to the structural integrity of these MOFs. The cumulative data illustrate that such MOFs have innately light-sensitive properties that are atypical in high surface area materials.

An electrochemical liquid–liquid–solid (ec-LLS) process that produces large amounts of crystalline semiconductors with tunable nanostructured shapes without any physical or chemical templating agent is presented. Electrodeposition of Ge from GeO2(aq) solutions followed by dissolution into a liquid Hg electrode, saturation of the liquid alloy, and precipitation can yield polycrystalline Ge(s) under ambient conditions. A unique advantage of ec-LLS is that it involves precipitation under electrochemical control, where the applied bias precisely defines the flux of Ge into the liquid electrode. Fidelity of the saturation and precipitation of Ge from liquid electrodes affords a variety of material morphologies, including dense films of oriented nanostructured filaments with large aspect ratios (>103). Electrodeposition involving a liquid electrolyte, a liquid electrode, and a solid deposit under ambient conditions represents a conceptually unexplored direct wet-chemical route for the preparation of bulk quantities of crystalline group-IV semiconductors without the time- and energy-intensive processing steps required in traditional preparations of semiconductor materials.

Nitrogen alloyed gallium phosphide (GaP1–xNx) nanowires have been prepared by annealing gallium phosphide (GaP) nanowires in flowing NH3(g) at 750 °C. X-ray diffraction patterns and electron microscopy showed that changes in the annealing conditions afforded controlled alloying of N without effecting a complete conversion to gallium nitride (GaN). Raman measurements on nanowire films and individual nanowires highlighted intense new signatures, consistent with symmetry reduction from N incorporation in the zincblende lattice. The resultant optical properties and photoresponse of the GaP1–xNx nanowire films were investigated by wavelength-dependent diffuse reflectance and photoelectrochemical measurements, respectively. Diffuse reflectance measurements showed progressively lower reflectivity of visible light for nanowire films annealed in increasingly higher levels of NH3(g), indicating an increased light absorption. Corresponding photoelectrochemical measurements of the GaP1–xNx nanowires revealed an increased quantum efficiency, relative to GaP, for energy conversion of light with wavelengths longer than 545 nm. The presented data set thus identifies a methodology for improving the solar energy conversion properties of GaP nanowire film photoelectrodes for visible light.

Ultrathin films of germanium (Ge) have been electrodeposited onto surface-enhanced raman spectroscopy (SERS)-active, polycrystalline gold (Au) nanoparticle film electrodes from aqueous solutions containing dissolved GeO2. An overlayer SERS strategy was employed to use the SERS-activity of the underlying Au electrode to enhance the Raman signatures separately for the Ge phonon mode and vibrational modes of surface groups. Electrochemical and spectroscopic data are presented that demonstrate monolayer-level detection of the electrodeposited material and the preparation of crystalline Ge films exhibiting quantum-confinement effects. Potential-dependent Raman spectra are shown that identify electrodeposition conditions where Ge films can be deposited with either long- or short-range crystalline order. Raman spectra collected with electrodeposited Ge films immersed in solutions containing CN−(aq) did not indicate a significant presence of pinholes that exposed the underlying Au(s) substrate. Raman spectra were also collected that identified a potential-dependence for Ge hydride formation at the interface of these films. Separate spectra were collected for the oxidative dissolution of Ge in solution and the complete dry oxidation of Ge to GeOx in air. These data sets cumulatively represent the first demonstration of the overlayer SERS strategy to follow surface chemical processes at crystalline, nanostructured, Ge materials in situ and in real time.Keywords: electrodeposition; germanium; Raman; SERS; surface chemistry

Single-crystalline gallium phosphide (GaP) surfaces have been functionalized with alkyl groups via a sequential Cl-activation, Grignard reaction process. X-ray photoelectron (XP) spectra of freshly etched GaP(111)A surfaces demonstrated reproducible signals for surficial Cl after treatment with PCl5 in chlorobenzene. The measured Cl content consistently corresponded to approximately a monolayer of coverage on GaP(111)A. In contrast, GaP(111)B surfaces treated with the same PCl5 solution under the same conditions exhibited macroscale roughening and yielded XP spectra that showed irreproducible Cl surface content often below the limit of detection of the spectrometer. The Cl-activated GaP(111)A surfaces were reactive toward alkyl Grignard reagents. Sessile contact angle measurements between water and GaP(111)A after various levels of treatment showed that GaP(111)A surfaces became significantly more hydrophobic following reaction with either CH3MgCl or C18H37MgCl. GaP(111)A surfaces reacted with C18H37MgCl demonstrated wetting properties consistent with surfaces modified with a dense layer of long alkyl chains. High-resolution C 1s XP spectra indicated that the carbonaceous species at GaP(111)A surfaces treated with Grignard reagents could not be ascribed solely to adventitious carbon. A shoulder in the C 1s XP spectra occurred at slightly lower binding energies for these samples, commensurate with the formation of Ga−C bonds. High-resolution P 2p XP spectra taken at various times during prolonged direct exposure to ambient laboratory air indicated that the resistance of GaP(111)A to surface oxidation was greatly enhanced after surface modification with alkyl groups. GaP(111)A samples that had been functionalized with C18H37− groups exhibited less than 0.1 nm of surface oxide after 7 weeks of continuous exposure to ambient air. GaP(111)A surfaces terminated with C18H37− groups were also used as platforms in Schottky heterojunctions with Hg. In comparison to freshly etched GaP(111)A, the alkyl-terminated GaP(111)A samples yielded current−voltage responses that were in accord with metal−insulator−semiconductor devices and indicated that this reaction strategy could be used to alter rates of heterogeneous charge transfer controllably. The wet chemical surface functionalization strategy described herein does not involve thiol/sulfide chemistry or gas-phase pretreatments and represents a new synthetic methodology for controlling the interfacial properties of GaP and related Ga-based III−V semiconductors.

The steady-state photoelectrochemical responses of semiconductor nanowire arrays in a nonaqueous regenerative photoelectrochemical cell were analyzed. Experimental and numerical simulation data were collected to determine the extent that dopant density levels, ND, have on the efficiency of semiconductor nanowire photoelectrodes with radii (r) comparable to the width of the depletion region (W). Films of Si nanowires (r < 40 nm) were prepared by metal-assisted chemical etching of single-crystalline Si(111) substrates with known bulk optoelectronic properties and utilized as photoelectrodes in a methanolic electrolyte containing dimethylferrocene (dmFc) and dimethylferrocenium (dmFc+). This photoelectrochemical system featured definable values for the rate of heterogeneous charge-transfer, the interfacial equilibrium barrier height (Φb), and the rate of surface recombination. Under white light illumination, the photocurrent−potential responses of Si nanowire arrays were strongly influenced by the ratio between the nanowire radius and the depletion region width (r/W). Lightly doped Si nanowire arrays consistently showed lower light-saturated photocurrents than heavily doped Si nanowire arrays despite having hole diffusion lengths, Lp, that were larger by a factor of 2. Measurement of the wavelength-dependent external quantum yields for the Si nanowire arrays separated out the effects from the underlying Si substrate and confirmed that carrier-collection was either significantly enhanced or suppressed by the Si nanowires depending on the value of r/W established by the Φb and ND. Digital simulations of nanowire heterojunctions using a two-dimensional semiconductor analysis software package (TeSCA) and known system parameters are presented that further explore the quantitative interplay between r/W and collection efficiency for nanowire photoelectrodes. The implications for designing low-cost semiconductor photoelectrodes using nanowire-based heterojunction architectures are examined, and tolerances for control over doping levels in semiconductor nanowire photoelectrodes are discussed.