Photochemical diodes for unassisted biomass valorization coupled with hydrogen production or carbon dioxide fixation

Abstract

This disclosure provides systems, methods, and apparatus related to photochemical diodes. In one aspect, a device include a photoanode, a photocathode, and a bipolar membrane between the photoanode and the photocathode. The photoanode comprises a first semiconductor, the first semiconductor being N-type doped, a first catalyst disposed over the first semiconductor, and the photoanode being disposed in an anolyte. The photocathode comprises a second semiconductor, the second semiconductor being P-type doped, a second catalyst disposed over the second semiconductor, and the photocathode being disposed in a catholyte. The photoanode and the photocathode are in electrical contact. A hydrogen reduction reaction or a carbon dioxide reduction reaction occurs at the photocathode and a chemical oxidation reaction occurs at the photoanode when the photocathode and the photoanode are illuminated with light.

Description

BACKGROUND

Finding sustainable and renewable energy is currently one of the most urgent challenges facing society today. With the sun providing 173 petawatts (PW) to the earth's surface every year, or enough energy in 1 hour to match the world's yearly energy consumption, artificial photosynthesis offers an attractive route to using solar energy to produce fuels such as hydrogen, a fundamental component for building a carbon-free economy.

One such approach to realizing artificial photosynthesis is presented through the photochemical diode. Photoanodes and photocathodes can be integrated through an ohmic contact, coupling both oxidative and reductive half-reactions in a single device. Conventional approaches have targeted the optimization of the overall water splitting (OWS) reaction in which hydrogen and oxygen are produced. However, the sluggish kinetics of the OER and the high thermodynamic potential requirement of 1.23 V for the OWS limit the current performance of bias-free photoelectrochemical (PEC) systems to ˜3.5 mA/cm². Given that 90% of the overall energy requirements come from OER, alternative oxidative reactions could enable more efficient PEC systems.

SUMMARY

One innovative aspect of the subject matter described in this disclosure can be implemented in a device including a photoanode, a photocathode, and a bipolar membrane between the photoanode and the photocathode. The photoanode comprises a first semiconductor, the first semiconductor being N-type doped, a first catalyst disposed over the first semiconductor, and the photoanode being disposed in an anolyte. The photocathode comprises a second semiconductor, the second semiconductor being P-type doped, a second catalyst disposed over the second semiconductor, and the photocathode being disposed in a catholyte. The first semiconductor is the same semiconductor and the second semiconductor. The photoanode and the photocathode are in electrical contact. A hydrogen reduction reaction (i.e., a hydrogen evolution reaction) or a carbon dioxide reduction reaction occurs at the photocathode and a chemical oxidation reaction occurs at the photoanode when the photocathode and the photoanode are illuminated with light.

Another innovative aspect of the subject matter described in this disclosure can be implemented in a method including providing a device. The device includes a photoanode, a photocathode, and a bipolar membrane between the photoanode and the photocathode. The photoanode includes a first semiconductor, the first semiconductor being N-type doped, with a first catalyst disposed over the first semiconductor. The photoanode is disposed in an anolyte. The photocathode includes a second semiconductor, the second semiconductor being P-type doped, with a second catalyst disposed over the second semiconductor. The photocathode is disposed in a catholyte. The first semiconductor is the same semiconductor and the second semiconductor. The photoanode and the photocathode are in electrical contact. The photoanode and the photocathode are exposed to light. A hydrogen reduction reaction or a carbon dioxide reduction reaction occurs at the photocathode, and a chemical oxidation reaction occurs at the photoanode.

Details of one or more embodiments of the subject matter described in this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages will become apparent from the description, the drawings, and the claims. Note that the relative dimensions of the following figures may not be drawn to scale.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1A and 1B show schematic illustrations of using GOR as an alternative to OER within a photochemical diode system. FIG. 1A shows an energy band diagram of photoelectrodes for HER, OER, and GOR. The photogenerated electrons in the photocathode perform HER while the holes in the photoanode perform GOR and/or OER. Concomitantly, majority carriers recombine at the ohmic contact. FIG. 1B shows current-voltage curves of electrochemical (EC) and PEC HER, GOR, and OER. Predicted operating photocurrent density of bias-free systems are shown for PEC HER-GOR (J_{op,PEC HER-GOR}) and HER-OER (J_{op,PEC HER-OER}).

FIGS. 2A-2C show the electrocatalytic performance of Pt, Au, and PtAu. FIG. 2A shows linear sweep voltammetry (LSV) scans of Pt, Au, and PtAu. FIG. 2C shows the zoomed-in LSV scans. FIG. 2B shows steady-state current densities of Pt, Au, and PtAu at different potentials. Experiments were done in 1 M KOH+1 M glycerol.

FIGS. 3A-3D show the photoelectrochemical performance of the PtAu/Si photoanode. FIG. 3A shows a schematic of the PtAu/Si photoanode showing the dopant layer and electron-hole separation. FIG. 3B shows LSV scans of the photoanode under chopped light, continuous, and no illumination in 1 M KOH+1 M glycerol. FIG. 3C shows the selectivity of PEC GOR using the PtAu/Si photoanode at 0.34 V vs RHE under illumination. Error bars are from three independent measurements. FIG. 3E shows the zoomed-in selectivity toward formic acid (FA) and acetic acid (ACA). FIG. 3D shows LSV scans showing the absolute current density of the PtAu/Si photoanode in 1 M KOH+1 M glycerol and the Pt/SiNW photocathode in 0.5 M H₂SO₄ under 1 sun illumination.

FIGS. 4A and 4B show photoelectrochemical performance of the integrated system of the PtAu/Si photoanode and Pt/SiNW photocathode in a two-electrode configuration. FIG. 4A shows LSV scans under chopped light, illumination, and no illumination. FIG. 4B shows bias-free diurnal stability test cycling between 12 h of illumination and 12 h of dark. The fluctuation in current density can be attributed to the generation of hydrogen gas bubbles on the photocathode surface.

FIG. 5 shows and example of schematic illustration of a photochemical diode.

FIG. 6 shows an example of a flow diagram illustrating a method of operating a photochemical diode.

FIGS. 7A-7C show the PEC CO₂RR performance of Au-NOLI/SiNW photocathodes. FIG. 7A shows the CO and H₂ faradic efficiency (FE) of Au-NOLI/SiNW photocathodes with different mass loadings at −0.17 V vs. RHE. FIG. 7B shows LSV scans of 160 μg/cm² mass loading of Au-NOLI/SiNW photocathode under chopped light, illumination, and no illumination. FIG. 7C shows CO and H₂ FE at different applied potentials of 160 μg/cm² mass loading of Au-NOLI/SiNW photocathode. Experiments were performed in CO₂-saturated 0.1 M KHCO₃ under 1 sun irradiation.

FIGS. 8A-8C show the structure and performance of PtAu/SiNW photoanodes. FIG. 8A shows a SEM image of a PtAu/SiNW photoanode. Scale bar: 200 nm. FIG. 8B shows LSV scans of PtAu/SiNW and PtAu/planar Si photoanodes under chopped light, illumination, and no illumination in 0.1 M glycerol+1 M KOH. FIG. 8C shows LSV scans showing the absolute current density of the Au-NOLI/SiNW photocathode and Au-NOLI dark cathode in CO₂-saturated 0.1 M KHCO₃, along with the PtAu/SiNW photoanode and PtAu dark anode in 0.1 M glycerol+1 M KOH.

FIGS. 9A-9F show the results of experiments on solar-driven bias-free CO₂RR by Au- and Pd-NOLI/SiNW photocathode coupled with PtAu/SiNW photoanode in a two-electrode configuration. FIG. 9A shows LSV scans of the integrated system of Au-NOLI/SiNW photocathode coupled with PtAu/SiNW photoanode under chopped light, illumination, and no illumination. FIG. 9B shows photocurrent density FIG. 9C shows FE of CO and H₂, and FIG. 9D shows FE of anodic products of Au-NOLI/SiNW|PtAu/SiNW integrated system without applied bias over 4 hours. FIG. 9E shows photocurrent density, and FIG. 9F shows FE of cathodic and anodic products of Pd-NOLI/SiNW|PtAu/SiNW integrated system without applied bias over 1 hour. Experiments were performed in CO₂-saturated 0.1 M KHCO₃ for photocathodic chamber and 0.1 M glycerol+1 M KOH for photoanodic chamber under 1 sun irradiation.

DETAILED DESCRIPTION

Reference will now be made in detail to some specific examples of the invention including the best modes contemplated by the inventors for carrying out the invention. Examples of these specific embodiments are illustrated in the accompanying drawings. While the invention is described in conjunction with these specific embodiments, it will be understood that it is not intended to limit the invention to the described embodiments. On the contrary, it is intended to cover alternatives, modifications, and equivalents as may be included within the spirit and scope of the invention as defined by the appended claims.

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. Particular example embodiments of the present invention may be implemented without some or all of these specific details. In other instances, well known process operations have not been described in detail in order not to unnecessarily obscure the present invention.

Various techniques and mechanisms of the present invention will sometimes be described in singular form for clarity. However, it should be noted that some embodiments include multiple iterations of a technique or multiple instantiations of a mechanism unless noted otherwise.

The terms “about” or “approximate” and the like are synonymous and are used to indicate that the value modified by the term has an understood range associated with it, where the range can be ±20%, ±15%, ±10%, ±5%, or ±1%. The terms “substantially” and the like are used to indicate that a value is close to a targeted value, where close can mean, for example, the value is within 80% of the targeted value, within 85% of the targeted value, within 90% of the targeted value, within 95% of the targeted value, or within 99% of the targeted value.

The oxidation of biomass-derived organic compounds has been proposed as an alternative to OER. These oxidative reactions have low potential requirements (<0.3 V vs RHE), showing potential reductions of up to ˜1 V in electrochemical cells when replacing OER. Of such compounds, glycerol, the main byproduct of biodiesel production, can also electrochemically produce products such as glyceraldehyde (GLD), dihydroxyacetone (DHA), glyceric acid (GLA), and lactic acid (LA), which are widely used in the cosmetics, pharmaceutical, and food industries. The value of its products and its low potential requirement make the glycerol oxidation reaction (GOR) an attractive pursuit.

The lower potential requirement of GOR also opens up an avenue for lower photovoltage materials in the photochemical diode design. FIG. 1A shows the redox potentials of HER, OER, and GOR as well as the band diagram of typically used photoelectrodes such as Si, TiO₂, and BiVO₄. For OWS, the redox potential of OER as well as the 1.23 V potential requirement often meant that metal oxides were favored as photoanode materials due to their wide band gaps supplying >1 eV of photovoltage and ability to stack on top of small band gap materials. However, maximum photocurrents for wide band gap materials are limited, with TiO₂ and BiVO₄ supplying 2 and 5 mA/cm², respectively (FIG. 1B).

Compared to OWS, GOR coupled with HER reduces the energy requirement by 700-1000 mV, allowing for smaller band gap materials with larger photocurrents to be used (FIG. 1B). Additionally, selectivity can be steered by choosing a semiconductor material such as Si where the valence band maximum (VBM) lies between the potentials for GOR and the potentials for the OER, preventing any holes from being used for the OER. In this design framework, a selective bias-free device can be achieved by coupling a medium band gap material such as silicon with a low onset potential catalyst for GOR.

Prior to photoelectrochemical experiments, it is necessary to find a suitable catalyst with low onset potentials toward GOR. Pt, commonly used for alcohol oxidation, shows a low onset potential of 0.5 V vs RHE (FIG. 2A) with a peak at 0.84 V vs RHE. However, Pt becomes readily poisoned by CO-like intermediates, dropping the current from 10 to 1 mA/cm² at 0.84 V vs RHE within a few minutes (FIG. 2B). On the other hand, Au has a low binding energy to CO which prevents CO poisoning and can maintain high current densities for GOR (FIG. 2B). However, Au shows significantly larger overpotentials with an onset of over 0.9 V vs RHE. The limitations of both materials can be mitigated by combining the advantages of Pt and Au synergistically. Using a cosputtered PtAu thin film, the onset potential toward GOR is 0.4 V vs RHE, which is 1 V lower than state-of-the-art catalysts for OER (FIG. 2A). PtAu shows enhanced current densities at low potentials and a steady-state current of 10 mA/cm² at 0.84 V vs RHE (FIGS. 2A-2C). It also shows both peaks at 0.84 and 1.3 V vs RHE, being shifted to higher potentials possibly by its bimetallic nature offering oxide resistance, which indicates the synergistic effects between both metals.

FIG. 5 shows and example of schematic illustration of a photochemical diode. A photochemical diode 500 includes a photoanode 510 comprising a first semiconductor 511, a photocathode 520 comprising a second semiconductor 521, and a bipolar membrane 530 between the photoanode 510 and the photocathode 520.

The first semiconductor 511 is N-type doped. A first catalyst 512 is disposed over the first semiconductor 511. The photoanode 510 is disposed in an anolyte 514. The second semiconductor 521 is P-type doped. A second catalyst 522 is disposed over the second semiconductor 521. The photocathode 520 is disposed in a catholyte 524. The first semiconductor 511 is the same semiconductor material as the second semiconductor 521. The photoanode 510 and the photocathode 520 are in electrical contact. When the photocathode 520 and the photoanode 510 are illuminated with light (e.g., sunlight), a hydrogen reduction reaction (i.e., a hydrogen evolution reaction) or a carbon dioxide reduction reaction occurs at the photocathode 520. A chemical oxidation reaction occurs at the photoanode 510.

In some embodiments, the bipolar membrane 530 comprises an anion exchange layer and a cation exchange layer. In some embodiments, the anion exchange layer and a cation exchange layer are both based on hydrocarbon resins.

In some embodiments, the first semiconductor 511 and the second semiconductor 521 both comprise silicon. In some embodiments, the first catalyst 512 comprises PtAu (e.g., PtAu nanoparticles). In some embodiments, the second catalyst 522 comprises Pt (e.g., Pt nanoparticles) (e.g., for a hydrogen evolution reaction). In some embodiments, the second catalyst 522 comprises Au (e.g., Au nanoparticles) (e.g., for a carbon dioxide reduction reaction). In some embodiments, the anolyte 514 comprises potassium hydroxide. In some embodiments, the anolyte 514 includes glycerol dissolved therein. In some embodiments, the catholyte 524 comprises sulfuric acid.

In some embodiments, a protective layer 540 is disposed on both the first semiconductor 511 and the second semiconductor 521. The first catalyst 512 is disposed on the protective layer 540 of the first semiconductor 511. The second catalyst 522 is disposed on the protective layer 540 of the second semiconductor 521. In some embodiments, the protective layer 540 comprises titanium dioxide (TiO₂).

In some embodiments, the first semiconductor 511 includes nanowires (not shown) on a surface of the first semiconductor. For example, the nanowires may be generated by etching the surface of the first semiconductor 511. The nanowires can increase the performance of the photochemical diode 500, at least in part by increasing the surface area of the photoanode 510.

In some embodiments, the second semiconductor 521 includes nanowires (not shown) on a surface of the second semiconductor. For example, the nanowires may be generated by etching the surface of the second semiconductor 521. The nanowires can increase the performance of the photochemical diode 500, at least in part by increasing the surface area of the photocathode 520.

In some embodiments, the chemical oxidation reaction is not an oxygen evolution reaction. In some embodiments, the chemical oxidation reaction is an oxidation reaction of glucose, 5-hydroxymethylfurfural, or glycerol. In some embodiments, the chemical oxidation reaction is a glycerol oxidation reaction. In some embodiments, the chemical oxidation reaction is a glycerol oxidation reaction, and wherein the glycerol oxidation reaction generates glyceraldehyde (GLD), dihydroxyacetone (DHA), glyceric acid (GLA), or lactic acid (LA).

In some embodiments, a reaction of the photoanode 510 is heavily doped with a P-type dopant. In some embodiments, up to about 10 nanometers (nm) in depth of the reaction includes the P-type dopant.

In some embodiments, a reaction of the photocathode 520 is heavily doped with a N-type dopant. In some embodiments, up to about 10 nm in depth of the reaction includes the N-type dopant.

FIG. 6 shows an example of a flow diagram illustrating a method of operating a photochemical diode. Starting at block 605 of the method 600 shown in FIG. 6, a device is provided. The device may be any of the devices or photochemical diodes described herein.

At block 610, the photoanode and the photocathode are exposed to light (e.g., sunlight). A hydrogen reduction reaction or a carbon dioxide reduction reaction occurs at the photocathode. A chemical oxidation reaction occurs at the photoanode.

In some embodiments, the chemical oxidation reaction is not an oxygen evolution reaction. In some embodiments, the chemical oxidation reaction is an oxidation reaction of glucose, 5-hydroxymethylfurfural, or glycerol. In some embodiments, the chemical oxidation reaction is a glycerol oxidation reaction. In some embodiments, the chemical oxidation reaction is a glycerol oxidation reaction, and wherein the glycerol oxidation reaction generates glyceraldehyde (GLD), dihydroxyacetone (DHA), glyceric acid (GLA), or lactic acid (LA).

The following examples are intended to be examples of the embodiments disclosed herein, and are not intended to be limiting.

EXAMPLES—PHOTOCHEMICAL DIODES—HYDROGEN EVOLUTION

Example—Fabrication of p+n Si Substrates

N-type phosphorus-doped 6″ Si wafers (<100> oriented, 1-10 Ohm-cm, prime, single sided-polished) were used as substrates for the fabrication process. A 6″ silicon wafer was used as the dopant carrier wafer. All the wafers were cleaned in a 4.9% HF bath for 3 minutes to remove native oxides and thoroughly washed with DI water and dried. The carrier wafer was spin-coated with a gallium silicate spin-on-dopant solution at 2200 rpm for 30 seconds and baked on a hotplate at 150° C. for 30 minutes. Afterwards, N-type phosphorus-doped Si wafers, cleaned with HF and water right before this process, were placed gently onto the carrier wafer such that polished surface is touching the dopant layer, and placed into a rapid thermal annealing chamber at 900° C. for 100 seconds under N2. After cooling, these p+n Si substrates are stored in ambient air until use.

Example—TiO₂ Protection Layer

The p+n-Si and SiNW substrates were cleaned in a 4.9% HF bath for 3 minutes and thoroughly washed with DI water and acetone and then dried. Afterwards, a 10 nm TiO₂ layer was deposited at 200° C. using atomic layer deposition and tetrakis(dimethylamido)titanium as the precursor in order to maintain stable performance for prolonged periods of time. After cooling, these substrates are stored in ambient air until use.

Example—Deposition of PtAu and Pt Catalyst

The deposition of PtAu and Pt catalyst were performed on a multi-target co-sputtering system with 3×3″ sputter guns, supplied by two 2 kW pulsed DC power supplies and one 1.5 kW DC power supply. After the deposition of TiO2 protection layer, around 4 nm of PtAu catalyst was co-sputtered onto p+n Si substrates with 15 seconds of 50 W power applied on the Pt target and 28 W power on the Au target. Around 2.5 nm of Pt catalyst was sputtered onto SiNW substrates with 20 seconds of 50 W power applied on the Pt target after TiO2 deposition. Both PtAu/Si and Pt/SiNW substrates were stored in ambient air until use.

Example—Electrode Preparation

For both PtAu/Si and Pt/SiNW, the substrates were used as photoanode and photocathode by creating electrically conductive connections to titanium foil. To do this, a Ga—In eutectic was scratched onto the back of the substrate, quick drying silver paint was applied on top of that and taped onto titanium foil using double-sided conductive carbon tape. The electrodes were left to dry for an additional 30 minutes in ambient conditions before being mounted onto the cell for measurements.

For a PtAu film, around 4 nm of PtAu catalyst was sputtered onto a clean glassy carbon electrode (GCE) using the same multi-target co-sputtering system as described previously.

Example—PEC Measurements

For PEC measurements, 4 nm of PtAu was sputtered onto a TiO₂-protected p⁺n-Si wafer (FIG. 3A). TiO₂ is a passivation layer that is commonly used to protect the silicon surface from corrosive alkaline environments. The surface p⁺ layer is formed to increase the photovoltage and to enhance the band bending near the semiconductor surface. A thin catalyst layer is also required due to the front wet-side illumination geometry. Four nanometers was found to be the optimum thickness of PtAu as thinner layers show reduced performance, which can be attributed to a decreased amount of active catalyst, while thicker layers start to decrease photocurrents by reflecting a significant portion of light. FIG. 3B shows the PEC GOR performance of the PtAu/TiO₂/p⁺n-Si (referred to as PtAu/Si onward) photoanode under 100 mW/cm² of air mass (AM) 1.5 simulated sunlight in 1 M KOH+1 M glycerol. The onset potential is around −0.05 V vs RHE and reaches 10 mA/cm² at a potential of 0.5 V vs RHE, which is the lowest overpotential and highest current density reported for PEC GOR. A strongly alkaline solution causes the first deprotonation step to be base catalyzed due to glycerol (pK_a of ˜14) acting as a weak acid. The deprotonated species has been found to be significantly more reactive, which is corroborated by our experiments using neutral and acidic electrolytes.

The product distribution of PtAu/Si is shown in FIG. 3C. At 0.34 V vs RHE, the main products that were produced were GLA with a faradaic efficiency (FE) of 48.1±2.3% and LA with FE=27.3±2.8% with total products reaching around FE_total≈80%. The rest likely comes from additional products such as tartronic acid and carbonate (coming from the oxidation of formic acid) that are not analyzable through ¹H NMR. Additionally, both DHA and GLD rearrange themselves to LA in alkaline electrolytes, making it difficult to differentiate the production of these compounds versus LA. The product distribution is similar to that of previously reported electrochemical GOR results using the PtAu catalyst under similar conditions. PEC control experiments indicate almost no photocurrent, indicating that glycerol is being oxidized on the PtAu surface.

As described earlier, metal oxides are often utilized as photoanodes due to their stability, wide band gap, large photovoltages, and deep VBM level. Surprisingly, despite these advantages and the reduced energy requirement of GOR compared to that of OER, past reports on the PEC performance of GOR on these metal oxides are either comparable to OER or have higher onset potentials, all of which have onset potentials 200-400 mV more positive than in this current work. When the same catalyst is used on TiO₂, the onset potentials and fill factors remain worse than when Si is used. One possible reason could be the poor hole transfer from the VBM to glycerol due to inadequate band alignment. The VBM levels of metal oxides such as TiO₂ and BiVO₄ are 2.7 and 2.1 eV vs NHE, respectively, which are 2-2.5 eV below the redox potential of GOR. On the other hand, the VBM of silicon with a donor concentration of 10¹⁵ cm⁻³ is around 0.42 eV vs NHE, which matches the potential for GOR closely and also prevents OER from occurring.

The low onset potential and large photocurrent density make PtAu/Si a suitable photoanode candidate for forming a bias-free integrated PEC system. In this demonstration, we used a Pt/TiO2/n⁺p-Si nanowire array (referred to as Pt/SiNW onward) as the photocathode for HER for its low overpotential and high catalytic performance. The photocathode was tested in 0.5 M H₂SO₄ using the same illumination conditions. FIG. 3D shows the overlap of the J-V PEC linear scans of the PtAu/Si photoanode and Pt/SiNW photocathode, where the expected current at no applied bias is seen at the intersection of the two curves. The large current density of 6 mA/cm² at the intersection suggests the feasibility of a bias-free PEC system.

For the solar-driven bias-free integrated PEC system, the photoelectrodes were placed into a two-chamber cell separated by a bipolar membrane (BPM) in a two-electrode configuration. A BPM allows two different pH environments to be used since an alkaline environment is optimal for GOR while an acidic environment is optimal for HER. BPMs also reduce crossover between the electrolytes, preventing the reduction of the oxidized products from GOR. The integrated system shows an onset potential at around −1.2 V and a photocurrent density of up to 6 mA/cm² at zero applied bias (FIG. 4A) with nearly unity H₂ selectivity, corresponding to 112 μmol H₂/h cm². This agrees well with the expected photocurrent density based on FIG. 3D showing that the voltage drop across the membrane is 0.83 V.

The stability of this bias-free GOR-HER system was also tested using diurnal cycling, which mimics the natural sunlight cycle. Potential switching due to light cycling often induces corrosion mechanisms not seen under constant illumination conditions such as rapid etching of the semiconductor or dissolution of the catalyst. Our system was able to maintain a high photocurrent density of over 4 mA/cm² for over 4 days before gradually dropping to zero in 6 days (FIG. 4B). Under continuous illumination, the stability of the integrated system was maintained over a similar total time range. Refreshing the electrolyte reactivated the performance, and X-ray photoelectron spectroscopy showed minor changes in the material other than the slight oxidation of Pt, suggesting that the deactivation mechanism comes from the consumption of electrolyte. A flow-cell configuration is then likely to assist the stability.

Example—Summary of Experimental Work

In this study, we demonstrated a framework for pursuing high current densities for a solar-driven device using photoelectrochemical HER coupled with GOR using silicon. By employing a low overpotential catalyst such as PtAu for GOR, which shows a low onset potential of 0.4 V vs RHE electrochemically, the voltage requirements to couple with HER become substantially lowered compared to OWS. As a result, the PtAu/Si photoanode exhibits a low onset potential of −0.05 V vs RHE and can be coupled with a Pt/SiNW photocathode to achieve a photocurrent density of 6 mA/cm² and stability for over 4 days under no applied bias.

EXAMPLES—PHOTOCHEMICAL DIODES—CARBON DIOXIDE REDUCTION

Example—CO₂RR on Au-NOLI/SiNW Photocathode

Silicon nanowires (SiNWs) have emerged as good candidates for photoelectrode material due to their ability to provide 0.3-0.45 V of photovoltage, high photocurrent, and surface area for more efficient catalyst loading. Yet, the performance of carbon dioxide reduction reaction (CO₂RR) SiNW photocathodes is typically limited near their onset potential owing to the higher overpotential of CO₂RR compared to hydrogen evolution reaction (HER).

Nanoparticle/ordered-ligand interlayer (NOLI) catalysts serve as a candidate for our photoelectrochemical (PEC) system due to their CO₂ to CO/formate selectivity at low overpotential. The origin of this catalytic performance has been attributed to the formation of an ordered ligand layer consecutively dissociated from a dense assembly of nanoparticles. Cations inserted into this interlayer strip their solvation shells and stabilize key intermediates like *CO₂⁻, which leads to an enhancement of catalytic turnover by two orders of magnitude compared to a pristine metal surface.

Specifically, Au-NOLI can convert CO₂ into CO at 100% selectivity at potentials as positive as −0.4 V vs RHE electrochemically. However, despite its unity selectivity toward CO, this potential is still excessively negative for a bias-free integrated system, where the operating potential is −0.1 vs. RHE or higher, depending on the employed photoanode. Therefore, p-type SiNWs are utilized as the semiconductor material, doped with an n⁺ shell to increase the photovoltage by 150-250 mV for the photocathode. This n⁺p-SiNW configuration delivers a photovoltage of approximately 0.4 V, capable of shifting the Au-NOLI cathodic curve towards a more positive region.

FIG. 7A illustrates the CO₂RR gas product selectivity of the Au-NOLI/SiNW photocathodic half-reaction at varying mass loadings of Au catalysts, with a CO faradic efficiency (FE) (96.0±3.0%) at 160 μg/cm². In order to trigger NOLI formation, a dense assembly of nanoparticles was formed, which leads to a strong ligand interdigitation that is needed for collective ligand dissociation at low overpotential. At lower mass loadings (20 to 80 μg/cm²), the Au nanoparticles were not sufficiently proximate to one another, hindering the formation of NOLI and resulting in non-optimal CO selectivity (FIG. 7A). With a mass loading of 160 μg/cm², assemblies of Au nanoparticles form on the nanowires, facilitating the formation of NOLI and thereby enhancing CO selectivity. Further increasing the mass loading only results in similar CO selectivity (FIG. 7A), therefore, the lowest catalyst loading (i.e., 160 μg/cm²) needed to achieve good CO selectivity was chosen as the photocathode condition for subsequent experiments.

FIG. 7B shows the linear sweep voltammetry (LSV) curve of Au-NOLI/SiNW with mass loading of 160 μg/cm². The onset potential is around 0.2 V vs. RHE, with the photocurrent density reaching and plateauing at 4 mA/cm² at a relatively positive potential of −0.3 V vs. RHE. FIG. 7C depicts the CO₂RR gas products of Au-NOLI/SiNW at various applied potentials. The CO selectivity starts to increase as the potential is more negative than the onset potential (around 0.2 V vs. RHE) and plateaus at around 95% FE after reaching approximately −0.07 V vs. RHE. This demonstrates that the combination of low onset potential, high selectivity, and photocurrent density renders Au-NOLI/SiNW a good candidate for a photocathode to integrate with a photoanode to realize bias-free CO₂RR.

Example—PtAu/SiNW Photoanode

The glycerol oxidation reaction (GOR) has shown promise as a substitute for the oxygen evolution reaction (OER) in the photoanodic reaction, resulting in improved overlapping current density between the photocathode and photoanode. It has been demonstrated that sputtered PtAu on planar n-type Si displays a low onset potential and high photocurrent, capable of being combined with an HER photocathode to achieve a high operating photocurrent density. Yet, owing to the higher overpotential of CO₂RR compared to HER, the overlap between the Au-NOLI/SiNW photocathode and PtAu/planar Si photoanode is limited. Instead of utilizing planar n-type Si, we employed n-type SiNWs with p⁺ shell formed on the fabricated nanowires as the photoanode material (FIG. 8A). Further control of catalyst loading was achieved by adjusting the thickness of the sputtered PtAu catalysts onto the SiNWs. FIG. 8B shows the LSV curves of PtAu/SiNW and PtAu/planar Si GOR photoanodes. The SiNW photoelectrodes, with higher surface area, intensified the photocurrent near the onset potential and displayed a better fill factor compared to the planar Si counterpart. This improvement could be credited to the increased loading of PtAu catalysts, which enhanced the photocurrent density in the lower potential region (approximately −0.1 V to 0.4 V vs. RHE), where the catalyst quantity limits the overall performance. In FIG. 8C, the overlap of LSV scans between the Au-NOLI/SiNW photocathode and PtAu/SiNW photoanode is depicted, alongside the dark electrochemical LSV curves of Au-NOLI on carbon paper and PtAu on glassy carbon electrode. The overlap between the photocathodic and photoanodic curves is significant, with the use of SiNW aiding in providing sufficient photovoltages for the cathodic and anodic curves to overlap.

Example—Solar-Driven CO₂ Reduction Toward CO with Glycerol Oxidation

For the bias-free integrated system, the Au-NOLI/SiNW photocathode and PtAu/SiNW photoanode were positioned in a two-chamber cell separated by a bipolar membrane (BPM) in a two-electrode configuration. The adoption of a BPM facilitates the use of two distinct pH environments: an alkaline environment with 1 M KOH for GOR, and a neutral pH of 6.8 in 0.1 M KHCO₃ for CO₂RR. The integrated system exhibited an onset potential of approximately −0.7 V and a bias-free photocurrent density of 1.5 mA/cm² (FIGS. 9A and 9B) with a CO FE of 96.7±1.0% (FIG. 9C). The bias-free photocurrent density remained above 1 mA/cm², and the CO FE could be maintained above 80% for at least 4 hours (FIGS. 9B and 9C). The anodic products are glycerate with a FE of 51.9±6.5% and lactate with FE=22.7±2.6% with total products reaching around FE_total=90% (FIG. 9D). The product distribution was similar to previously reported PEC GOR results using PtAu catalysts under similar conditions.

Example—Solar-Driven CO₂ Reduction Toward Formate with Glycerol Oxidation

Adhering to this framework, Pd-NOLI, known for its ability to selectively produce formate at low overpotential electrochemically, was applied to the CO₂RR photocathode to produce formate. Following adjustments of mass loading, the Pd-NOLI/SiNW photocathode was paired with the PtAu/SiNW photoanode, resulting in an integrated system with a bias-free photocurrent density of approximately 2 mA/cm² (FIG. 9E). The cathodic products demonstrated a ratio exceeding 3:1 of formate to hydrogen (FIG. 9F), whereas the major anodic products included glycerate with a FE of 44.4±5.7% and lactate with FE=42.3±10.6%. The distribution of anodic products exhibits slight variations compared to the Au-NOLI/SiNW photocathode and PtAu/SiNW photoanode system, which is attributed to the minor differences in operating potential between the two systems (FIG. 8C).

CONCLUSION

Further details regarding the embodiments disclosed herein can be found in Jia-An Lin et al., Photochemical Diodes for Simultaneous Bias-Free Glycerol Valorization and Hydrogen Evolution, J. Am. Chem. Soc. 2023, 145, 24, 12987-12991, which is herein incorporated by reference.

In the foregoing specification, the invention has been described with reference to specific embodiments. However, one of ordinary skill in the art appreciates that various modifications and changes can be made without departing from the scope of the invention as set forth in the claims below. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of invention.

Claims

1. A device comprising: a photoanode comprising a first semiconductor, the first semiconductor being N-type doped, a first catalyst disposed over the first semiconductor, the photoanode being disposed in an anolyte;a photocathode comprising a second semiconductor, the second semiconductor being P-type doped, a second catalyst disposed over the second semiconductor, the photocathode being disposed in a catholyte, the first semiconductor being the same semiconductor material as the second semiconductor, and the photoanode and the photocathode being in electrical contact; anda bipolar membrane between the photoanode and the photocathode; anda hydrogen reduction reaction or a carbon dioxide reduction reaction occurs at the photocathode and a chemical oxidation reaction occurs at the photoanode when the photocathode and the photoanode are illuminated with light.

2. The device of claim 1, wherein the bipolar membrane comprises an anion exchange layer and a cation exchange layer.

3. The device of claim 2, wherein the anion exchange layer and a cation exchange layer are both based on hydrocarbon resins.

4. The device of claim 1, wherein the first semiconductor and the second semiconductor both comprise silicon.

5. The device of claim 1, wherein a protective layer is disposed on both the first semiconductor and the second semiconductor.

6. The device of claim 5, wherein the protective layer comprises titanium dioxide (TiO2).

7. The device of claim 1, wherein the first catalyst comprises PtAu, and wherein the second catalyst is a catalyst from a group Pt, Au, and PdAu.

8. The device of claim 1, wherein the anolyte comprises potassium hydroxide, and wherein the catholyte comprises sulfuric acid.

9. The device of claim 1, wherein the anolyte includes glycerol dissolved therein.

10. The device of claim 1, wherein the second semiconductor includes nanowires on a surface of the second semiconductor.

11. The device of claim 1, wherein the chemical oxidation reaction is not an oxygen evolution reaction.

12. The device of claim 1, wherein the chemical oxidation reaction is an oxidation reaction of glucose, 5-hydroxymethylfurfural, or glycerol.

13. The device of claim 1, wherein the chemical oxidation reaction is a glycerol oxidation reaction.

14. The device of claim 1, wherein the chemical oxidation reaction is a glycerol oxidation reaction, and wherein the glycerol oxidation reaction generates glyceraldehyde (GLD), dihydroxyacetone (DHA), glyceric acid (GLA), or lactic acid (LA).

15. The device of claim 1, wherein a surface of the photoanode is heavily doped with a P-type dopant, and wherein up to about 10 nanometers in depth of the surface includes the P-type dopant.

16. The device of claim 1, wherein a surface of the photocathode is heavily doped with a N-type dopant, and wherein up to about 10 nanometers in depth of the surface include the N-type dopant.

17. A device comprising: a photoanode comprising N-type doped silicon, a PtAu catalyst disposed over the photoanode, the photoanode being disposed in an anolyte;a photocathode P-type doped silicon, a Pt catalyst disposed over the photocathode, the photocathode being disposed in a catholyte, and the photoanode and the photocathode being in electrical contact; anda bipolar membrane between the photoanode and the photocathode.

18. A method comprising: providing a device, the device including: a photoanode comprising a first semiconductor, the first semiconductor being N-type doped, a first catalyst disposed over the first semiconductor, the photoanode being disposed in an anolyte,a photocathode comprising a second semiconductor, the second semiconductor being P-type doped, a second catalyst disposed over the second semiconductor, the photocathode being disposed in a catholyte, the first semiconductor being the same semiconductor and the second semiconductor, and the photoanode and the photocathode being in electrical contact, anda bipolar membrane between the photoanode and the photocathode; andexposing the photoanode and the photocathode to light, a hydrogen reduction reaction or a carbon dioxide reduction reaction occurring at the photocathode, and a chemical oxidation reaction occurring at the photoanode.