WO2025058928A2 - Silicon nanowire biophotochemical diodes for light-powered carbon dioxide reduction and glycerol valorization - Google Patents

Abstract

An example apparatus is disclosed. The apparatus includes a bio catalyzed cathode to reduce a gas and an anode electrically connected to the bio catalyzed cathode to perform glycerol valorization.

Description

SILICON NANOWIRE BIOPHOTOCHEMICAL DIODES FOR LIGHT- POWERED CARBON DIOXIDE REDUCTION AND GLYCEROL VALORIZATION

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of United States Provisional Patent Application Serial No. 63/581 ,271 , filed September 7, 2023, which is herein incorporated by reference in its entirety.

REFERENCE TO GOVERNMENT FUNDING

[0002] The invention was made with government support under grant Number 2217161 awarded by the National Science Foundation. The government has certain rights in the invention.

FIELD OF THE INVENTION

[0003] The present disclosure describes a multi-chamber bio catalyzed light- powered apparatus to reduce carbon dioxide (CO2).

BACKGROUND

[0004] Artificial photosynthesis is a term that generally refers to any scheme for capturing and then storing energy from sunlight by producing a fuel. Artificial photosynthesis has been shown to be a promising route to store light energy into chemical bonds and close the carbon cycle.

SUMMARY

[0005] In one example, an apparatus includes a bio catalyzed cathode to reduce a gas and an anode electrically connected to the bio catalyzed cathode to perform glycerol valorization.

[0006] In another example, a photochemical diode device includes a cathode and an anode electrically connected to the cathode. The cathode includes a p- doped silicon nanowire array and a first electrode formed on the p-doped silicon nanowire array, wherein the first electrode has been catalyzed using a microbial biocatalyst to facilitate reduction of a gas. The anode includes an n-doped silicon nanowire array, wherein the n-doped silicon nanowire array has been catalyzed with a metallic catalyst and a second electrode formed on the n-doped silicon nanowire array.

[0007] In another example, a method of fabricating a dual-electrode photochemical diode includes fabricating a p-doped silicon nanowire substrate and a separate n-doped silicon nanowire substrate, depositing a metallic catalyst on the n-doped silicon nanowire substrate, forming an electrode on the n-doped silicon nanowire substrate on which the metallic catalyst has been deposited, to form a photoanode of the dual-electrode photochemical diode, forming an electrode on the p-doped silicon nanowire substrate, to form a photocathode of the dual-electrode photochemical diode, and directly interfacing a microbial biocatalyst with the electrode of the p-doped silicon nanowire substrate.

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The teaching of the present disclosure can be readily understood by considering the following detailed description in conjunction with the accompanying drawings, in which:

[0009] FIG. 1A is a schematic diagram illustrating one example of a photochemical diode according to the present disclosure;

[0010] FIG. 1 B illustrates a photograph of a wafer-scale example of the photochemical diode of FIG. 1A;

[0011] FIG. 1 C is a schematic energy diagram of the photochemical diode of FIG. 1A;

[0012] FIG. 2 illustrates a graph that shows the differences in photoelectrochemical performance of the photochemical diode of FIG. 1A, when using platinum versus nickel sputtered onto the photocathode;

[0013] FIG. 3A illustrates a graph that plots photocurrent densities versus bias (vs. RHE) for the n⁺p-Si photocathode of FIG. 1 A under twenty mW/cm² of 740 nm LED continuous irradiation, chopped irradiation, and dark condition;

[0014] FIG. 3B illustrates a graph plotting photocurrent densities versus bias (vs. RHE) of silicon nanowire photocathodes under different light intensities of 740 nm LED illumination;

[0015] FIG. 3C illustrates a graph plotting photocurrent densities versus bias (vs. RHE) of silicon nanowire photocathodes under different pH of electrolytes;

[0016] FIG. 3D illustrates a graph plotting stable current density traces versus time for the silicon nanowire photocathode using an MES buffer for over twelve hours at 0.15 V vs. RHE;

[0017] FIG. 3E illustrates a graph plotting the photocurrent densities versus applied bias (vs. RHE) using a standard 0.5M sulfuric acid and the biocompatible buffer with the silicon nanowire photocathodes under the same light irradiation conditions;

[0018] FIG. 4A illustrates a graph plotting photocurrent densities versus bias (vs. RHE) for p-Si and n⁺p-Si photocathodes;

[0019] FIG. 4B illustrates a graph plotting photocurrent densities versus bias (vs. RHE) for a sputtered Pt-loaded n⁺p-Si photocathode and a Pt wire;

[0020] FIG. 5 illustrates graph plotting the various growth curves of S. ovata with MES buffer-containing media;

[0021] FIG. 6 illustrates nuclear magnetic resonance spectra of abiotic and biotic photoelectrochemical operations;

[0022] FIG. 7A illustrates a graph 700 plotting the time evolution of photocurrent densities of silicon nanowire photocathodes using wild-type S. ovata and methanol-adapted S. ovata at approximately 0.2 V vs. RHE;

[0023] FIG. 7B shows a scanning electron microscope image of wild-type S. ovata;

[0024] FIG. 7C shows a scanning electron microscope image of methanol- adapted S. ovata;

[0025] FIG. 7D illustrates a graph plotting photocurrent densities versus bias (vs. RHE) of bio and abiotic photocathodes;

[0026] FIG. 7E illustrates a graph plotting photocurrent densities versus bias (vs. RHE) of bio-photocathodes under twenty mW/cm² of 740 nm LED irradiation, chopped irradiation, and dark conditions;

[0027] FIG. 7F illustrates a graph plotting photocurrent densities versus bias (vs. RHE) of glycerol oxidation reactions under continuous, chopped, and dark irradiation;

[0028] Figure 7G illustrates a graph plotting the overlap of a PEC linear scan of bioSiNW photocathode and PtAu SiNW photoanode;

[0029] FIG. 8 illustrates a table of photoelectrochemical CO2 reduction using S.ovata/S nanowire biohybrids;

[0030] FIG. 9 illustrates a graph plotting the photoelectrochemical performance of planar n-type silicon and nanowire n-type silicon;

[0031] FIG. 10A illustrates a graph plotting photocurrent densities versus bias (vs. RHE) under twenty mW/cm² of 740 LED irradiation, chopped irradiation, and dark;

[0032] FIG. 10B illustrates a graph showing the long-term operation of the system plotted in FIG. 10A;

[0033] FIG. 10C illustrates a graph showing a product analysis of the system plotted in FIG. 10A;

[0034] FIG. 11A illustrates a graph showing the long-term operation of a system under control (i.e., dark) conditions;

[0035] FIG. 11 B illustrates a graph comparing the average photocurrent densities for the S.ovafa/Si photocathode and PtAu/Si photoanode configuration under irradiated and dark conditions;

[0036] FIG. 11 C illustrates a graph comparing the acetate productions for the S.ovata/S photocathode and PtAu/Si photoanode configuration under irradiated and dark conditions;

[0037] FIG. 11 D illustrates a graph comparing the oxygenate productions for the S.o afa/Si photocathode and PtAu/Si photoanode configuration under irradiated and dark conditions;

[0038] FIG. 12 illustrates a graph 1200 plotting the photocurrent densities versus time of an S. ovata/Si photocathode 102 and a Pt-Au/Si photoanode 104 configuration under twenty mW/cm² and forty mW/cm² of red light irradiation;

[0039] FIG. 13 illustrates a photograph of a horizontal stacking arrangement for an array of photochemical diode devices fabricated according to photochemical diode device 100 of FIG. 1 ;

[0040] FIG. 14 illustrates a photograph of a vertical stacking arrangement for an array of photochemical diode devices fabricated according to photochemical diode device 100 of FIG. 1 ; and [0041] FIG. 15 is a flow diagram illustrating one example of a method for forming a photochemical diode device, according to the present disclosure.

[0042] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures.

DETAILED DESCRIPTION

[0043] Examples of the present disclosure provide an apparatus and method for light-powered reduction of carbon dioxide and valorization of glycerol using silicon nanowire biophotochemical diodes. As discussed above, artificial photosynthesis is a term that generally refers to any scheme for capturing and then storing energy from sunlight by producing a fuel. Artificial photosynthesis has been shown to be a promising route to store light energy into chemical bonds and close the carbon cycle.

[0044] Photochemical diodes, in which two photoelectrodes are wired to drive valuable half-reactions, provide a platform to realize light-driven fuel generation. In bias-free photochemical diode architectures, the utilization of efficient catalysts plays a critical role in determining the overall cell performance, requiring both low onset potential and a high operating current density. However, currently available electrocatalysts rarely provide a low enough thermodynamic barrier to drive CO2 reduction and/or oxygen evolution via the photovoltages provided with semiconducting materials.

[0045] Examples of the present disclosure utilize a highly selective and efficient microbial CO2 biocatalyst, interfaced with a robust silicon (Si) photocathode coupled with a silicon photoanode for a glycerol oxidation reaction (GOR) to drive a bias-free CO2 reduction reaction. In one example, radial doping of silicon photoelectrodes facilitates an approximately 0.85 V of photovoltage in a dualelectrode configuration. Accordingly, the combination of silicon photochemical diodes interfaced with biocatalysts can achieve a maximum photocurrent of 1.2 mA/cm² with approximately eighty percent faradaic efficiency with valuable products. The photochemical diode apparatus of the present disclosure provides a viable platform for utilizing abundant light sources to capture and store CO2 in valuable chemical bonds with the nanomaterial built from abundant silicon.

[0046] In one particular example, a photochemical diode-based apparatus may comprise a multi-chamber bio catalyzed light-powered apparatus that reduces CO2 by converting the CO2 into desired compounds, such as acetate, alcohols, or various polymers, and generates oxygenates via glycerol valorization.

[0047] The intermittent nature of sunlight poses a challenge in harnessing light energy on a practical scale for powering the modern world. One possible solution is to utilize the photon energy for driving uphill electrochemistry that produces chemical fuels that can be stored and transported for later use. Photoelectrochemical (PEC) CO2 reduction is a promising route to store light energy in the chemical bonds and to promote carbon neutrality.

[0048] In one known photochemical diode, a p-type semiconductor photocathode and an n-type semiconductor photoanode are wired together to drive reduction and oxidation reactions with a p/n-PEC device configuration. The photoexcited charge carriers within a solid-state semiconductor electrode are transferred to reactants (e.g., CO2 molecules dissolved in a cathodic electrolyte) at the solid-liquid junction redox reactions. The working principle in this case is that the total harvested light energy from both photoelectrodes (i.e., the photocathode and the photoanode) should be larger than the thermodynamic energy requirement plus the kinetic overpotentials for oxidation and reduction reactions. While the above-described photochemical diode has been demonstrated for a thermodynamically and kinetically more accessible hydrogen evolution reaction (HER), limited success has been achieved in applying the photochemical diode to a CO2 reduction reaction (CORR) due to the higher catalytic overpotentials of currently available abiotic CO2 catalysts.

[0049] For example, a photochemical diode including a gold (Au) nanoparticle loaded amorphous Si photocathode and a bismuth vanadate photoanode has been demonstrated to enable a bias-free PEC system for CO2 reduction, but the reduced products mainly comprised carbon monoxide (CO) and a parasitic hydrogen (H2) with relatively low J_op of 0.24 mA cm^-2. Embedded photovoltaic components, such as perovskite solar cells, may help to increase the total photovoltage and have been demonstrated to produce diatomic and triatomic carbon (C2 and C3) with copper-based electrocatalysts, though the embedded components complicate the device structure and increase the costs to manufacture the device. A bias-free PEC device to reduce CO2 to higher products beyond CO has yet to be achieved for the abiotic version of the photochemical diode.

[0050] Examples of the present disclosure provide bias-free CO2-to-C2 reduction using highly efficient and selective microorganisms (e.g., Sporomusa ovata) that are directly interfaced with a high photovoltage p-type silicon photocathode. The whole-cell catalysts have been demonstrated to be capable of operating at, or operating more positively than, the thermodynamic potential of CO2-to-acetate reduction with the aid of photovoltage.

[0051] In one example platinum (Pt) and/or gold (Au) is deposited on an n-type silicon photoanode for glycerol oxidation reaction (GOR). The photoanode is wired with a p-type photocathode to drive an effective and valuable photoanodic reaction. The nanofabricated photoelectrode in wafer-scale may serve as an efficient and inexpensive light absorber. Approximately 0.85 V of photovoltage may be harvested from the Si photoanode. In a further example, illuminating the photocathode with a low intensity (e g., approximately 740 nm) light emitting diode (LED) may be sufficient to drive biological CO2 reduction reaction (CO2RR) and GOR in tandem.

[0052] FIG. 1A is a schematic diagram illustrating one example of a photochemical diode 100 according to the present disclosure. FIG. 1 B illustrates a photograph of a wafer-scale example of the photochemical diode 100 of FIG. 1A. For instance, the wafer-scale example may be a six-inch, nanofabricated photochemical diode.

[0053] As illustrated, the photochemical diode 100 generally comprises a photocathode 102 and a photoanode 104. In one example, the photocathode 102 comprises a p-type silicon nanowire array (illustrated at microscopic level in FIG. 1A). Carbon dioxide reduction occurs at the photocathode 102 with a microbial CO2 biocatalyst (e.g., S. ovata). In one example, the photoanode comprises an n-type silicon nanowire array (illustrated at microscopic level in FIG. 1A). Glycerol oxidation occurs at the photoanode 104 under red light irradiation. [0054] FIG. 1 C is a schematic energy diagram of the photochemical diode 100 of FIG. 1A. In particular, the example of FIG. 1 C illustrates the photochemical diode 100 coupling a CO2RR (i.e., the electrochemical reduction of carbon dioxide) catalyzed by S. ovata and GOR.

[0055] Experimental results, discussed in greater detail below, have shown that the silicon p-n junction of photochemical diode 100 can achieve a highest operating current density of approximately 1.2 mA/cm² with faradic efficiencies exceeding eighty percent for both cathodic and anodic reactions.

Fundamentals of photoelectrochemistry of the silicon nanowire (SiNW) array in a neutral biological buffer:

[0056] The photoelectrochemical performance of the photochemical diode 100 was abiotically tested with biological buffers in neutral pH under red light irradiation. The wavelength of the red light, 740 nanometer (nm), was well below that of the bandgap of silicon (i.e., 1100 nm), to efficiently excite the carriers and eliminate the energy loss through the thermalization and antimicrobial activity of high-energy photons. A highly doped n⁺ shell was formed on a p-type silicon nanowire array of the photocathode 102 to increase the photovoltage of the photocathode 102. A thirty nm crystalline titanium dioxide (TiCh) was deposited by atomic layer deposition (ALD) to protect the doped n* shell from being oxidized during operation of the photochemical diode 100, and the TiO2 does not interfere with the red light due to the large bandgap of the TiO2.

[0057] Next, three nm of platinum was sputtered onto the top surface of the TiC>2 as a co-catalyst to facilitate the charge transfer to the electrolyte. In other examples, nickel may be sputtered onto the top surface of the TO2 as the cocatalyst. However, although nickel is biocompatible with and connects easily to membrane-bound protein, the experimental results showed that platinum conveyed noticeably higher photoelectrochemical performances (e.g., 200 millivolts (mV) more positive onset potential and four times higher photocurrent generation, approximately 1.4 millamps per square centimeter (mA/cm²), and approximately 0.36 mA/cm² at 0.1 Volts versus reversible hydrogen electrode (RHE)).

[0058] FIG. 2 illustrates a graph 200 that shows the differences in photoelectrochemical performance of the photochemical diode 100 of FIG. 1A, when using platinum versus nickel sputtered onto the photocathode 102. In particular, the graph 200 plots the photocurrent densities versus applied bias (vs. RHE) for TiO2 protected n+p-Si with a three nm film of sputtered platinum (Pt) and a ten nm film of sputtered nickel. Both of the samples were tested in a biological buffer under twenty mW/cm2 of 740 nm LED with a scan rate of ten mV/sec.

[0059] The p-type silicon photocathode 102 shows an onset potential of approximately 0.4 V vs. RHE under irradiation. FIG. 3A illustrates a graph 300 that plots photocurrent densities versus bias (vs. RHE) for the n⁺p-Si photocathode 102 of FIG. 1A under twenty mW/cm² of 740 nm LED continuous irradiation, chopped irradiation, and dark condition. The dark condition and chopped irradiation scans corroborate that the photocurrent was generated from light.

[0060] The experimental results show that the n⁺ radial doping process provides an addition of 200 mV of photovoltage, and the photochemical diode 100 of FIG. 1A shows a total of approximately 400 mV of photovoltage under this red-light irradiation relative to dark Pt electrodes. FIG. 4A illustrates a graph 400 plotting photocurrent densities versus bias (vs. RHE) for p-Si and n⁺p-Si photocathodes 102, while FIG. 4B illustrates a graph 402 plotting photocurrent densities versus bias (vs. RHE) fora sputtered Pt-loaded n⁺p-Si nanowire (SiNW) photocathode 102 a Pt wire. All of the samples in FIGs. 4A and 4B were tested in 0.5 M H2SO4 (unless indicated otherwise) with a scan rate of ten mV/sec at ambient temperatures.

[0061] The influence of light intensity ranging from twenty mW/cm² to seven mW/cm² on photocurrent is shown in FIG. 3B. FIG. 3B illustrates a graph 302 plotting photocurrent densities versus bias (vs. RHE) of silicon nanowire photocathodes 102 under different light intensities of 740 nm LED illumination. The lower light intensities yielded a lower photocurrent and photovoltage. The light intensity of twenty mW/cm² was set as an experimental condition because the intensity is close to the daily average intensity of the near-infrared (IR) region of sunlight, so the intensity can closely simulate natural irradiation conditions. The intensity also exhibits an onset voltage of 0.4 V vs. RHE and 1 mA/cm² at 0.2 V vs. RHE, which sufficiently overlaps with the onset of GOR silicon photoanode 104, approximately zero Volts vs. RHE, for the later combination in a dual-electrode configuration.

[0062] The effect of electrolyte pH was also investigated, since the microenvironment around nanowires creates the local pH gradient. FIG. 3C is a graph 304 plotting photocurrent densities versus bias (vs. RHE) of silicon nanowire photocathodes 102 under different pH of electrolytes. The photocurrents exhibit a clear pH dependency, showing that the catalytic rate of hydrogen evolution on Pt is fast enough relative to the local concentration of protons in neutral pH so that the overall reaction is influenced by the local pH.

[0063] It also has been observed that under these buffered neutral pH conditions, the buffer concentration influences the overall hydrogen evolution, since the buffer concentration determines the rate of reactant (proton) regeneration. Thus, a fifty millimolar (mM) zwitterionic 2-(N- morpholino)ethanesulfonic acid (MES) buffer was experimentally introduced as an electrolyte, replacing an eighteen mM phosphate buffer. The buffer did not influence the metabolism of S. ovata when S. ovata were cultured in the same concentration of MES.

[0064] FIG. 5 illustrates graph 500 plotting the various growth curves of S. ovata with MES buffer-containing media. Methanol-adapted S. ovata was cultured in Balch-type tubes with the headspace of thirty psi 80%:20% H2 and CO2 with 50mM MES in the media (circles) and without MES (squares), and with 30 psi of 80%:20% N2 and CO2 with 50mM MES in the media (triangles). The error bar stands for the standard deviation of three biological replicates.

[0065] This buffering agent of FIG. 5 has a pKa of 6.27, providing enhanced buffering capacity at the pH near the pKa compared to conventional phosphate buffering agents, which have a pKa of 7.21. FIG. 3D is a graph 306 plotting stable current density traces versus time for the silicon nanowire photocathode 102 using an MES buffer for over twelve hours at 0.15 vs. RHE. The scan rate of all linear voltammetry scan measurements was ten mV/sec.

[0066] FIG. 3E illustrates a graph plotting the photocurrent densities versus applied bias (vs. RHE) using a standard 0.5M sulfuric acid 310 and the biocompatible buffer 312 with the silicon nanowire photocathodes under the same light irradiation conditions. The data plotted in FIG. 3E was obtained with linear sweep voltammetry scans of a standard 0.5 sulfuric acid and the biological buffer electrolyte, indicating the influence of electrolytes on the respective photochemical performances. As a result, the bacterial medium used in the experiments exhibited a relatively higher ohmic drop than standard sulfuric acid due to the higher solution resistance of the bacterial medium, as shown in the electrochemical impedance spectroscopy (EIS) results in the inset of FIG. 3E. Although an additional electrolyte engineering for biocatalysts might reduce the ohmic drop and enhance the electrochemical performances, the voltage drop remains modest around the expected operating current density of the bias-free system (e.g., approximately 0.2 V of drop at one mA/cm²).

[0067] Also, there was no production of acetates observed at this abiotic photoelectrochemical condition due to the excellent hydrogen evolution activity of platinum, whereas the CO2 gas continuously flowed in the device. FIG. 6 illustrates nuclear magnetic resonance (NMR) spectra 600 of abiotic and biotic photoelectrochemical operations, as described above. In FIG. 6, Tetramethylsilane (TMS) was used as an internal standard for quantification. A clear acetate peak appeared at around 1.8 ppm of chemical shift. The abiotic photoelectrochemical operation, without S. ovata did not produce any acetate, due to an excellent hydrogen evolution activity of Pt.

Photoelectrochemical half-reactions for bio-CO2 reduction reaction and glycerol oxidation:

[0068] Next, the microbial catalysts were assembled on the abiotic photocathodes 102 to upgrade the chemistry from hydrogen evolution to CO2 reduction. Two strains of Sporomusa ovata were tested under the photocathodic condition, wild-type, and a methanol-adapted strain, showing a faster autotrophic metabolism. From a chronoamperometry operation over a few days, the adapted strain increased the current density approximately 5-fold higher than the initial state and formed a fully packed hybrid. FIG. 7A illustrates a graph 700 plotting the time evolution of photocurrent densities of silicon nanowire photocathodes using wild-type S. ovata and methanol-adapted S. ovata at approximately 0.2 V vs. RHE.

[0069] The current density of biohybrids in this case indicates how metabolically active bacteria are on the cathodes to accept electrons and how fast the metabolically active bacteria catalyze CO2 and connect an electron flux transferred from electrodes to their CO2 reductive pathway. FIGs. 7B and 7C show scanning electron microscope (SEM) images of the interface after the photoelectrochemical operations for the two strains illustrated in FIG. 7A. For example FIG. 7B shows an SEM image of wild-type S. ovata and FIG. 7C shows a SEM image of methanol-adapted S. ovata.

[0070] The enhanced performance of the adapted S. ovata relative to the wildtype S. ovata is attributed at least in part to the adapted strain’s ability to uptake current more quickly. Even though both strains started from the same bacterial loading density, OD545 approximately 0.04 in the electrolyte, the adapted strain has a much faster metabolism, and thus, can reproduce more quickly, as clearly shown in the two SEM images of FIGs. 7B and 7C. The more direct attachment makes more charge transfer channels at the interface and increases the current density at the same potential. Also, the increased biocatalyst loading on the photocathode 102, which fully covered the surface of photoelectrodes, did not negatively affect the photocurrent generation, but instead increased it. This clearly shows that the rate of photoelectrochemical CO2 reaction is determined by the catalysts and stresses the importance of introducing an efficient catalyst.

[0071] FIG. 7D illustrates a graph 702 plotting photocurrent densities versus bias (vs. RHE) of bio and abiotic photocathodes. FIG. 7E illustrates a graph 704 plotting photocurrent densities versus bias (vs. RHE) of bio-photocathodes under twenty mW/cm² of 740 nm LED irradiation, chopped irradiation, and dark conditions. The data was obtained with linear sweep voltammetry of biotic and abiotic conditions. The conditions plotted in FIG. 7D and FIG. 7E both have a similar cathodic onset potential of around 0.4 V vs. RHE, and the biotic condition shows a slightly higher kinetic rate. The biophotocathodes exhibit over eighty percent (n=3) of faradaic efficiency toward acetate.

[0072] FIG. 8 illustrates a table 800 of photoelectrochemical CO2 reduction using S.ovata/Si nanowire biohybrids. A 720 nm LED with twenty 0mW/cm²was used as a light source to power the photocathodic reactions. The reactor temperature was set as 28-30 °C for optimal performance of the S. ovata. The stirring speed was 250 rpm. 1 H NMR was used to quantify the acetate accumulation. The analytical method is described greater detail below in connection with cathodic product analysis. The chronoamperometry experiment was operated for at least for twelve hours.

[0073] Linear voltammetry scan under dark and chopped illumination confirms a photo-induced electron transfer. FIG. 7E shows a graph 704 plotting photocurrent densities versus bias (vs. RHE) of bio-photocathodes under twenty mW/cm² of 740 nm LED irradiation, chopped irradiation, and dark conditions.

[0074] As a counter anodic reaction to CO2RR, GOR was investigated on an n-type SiNW photoanode. GOR has recently gained attention as an efficient photoanodic reaction for PEC devices. It has been reported that Pt-Au is a suitable GOR catalyst, because Pt-Au combines the advantages of the low overpotential of Pt and the high steady state current of Au synergistically. Photoelectrochemically, co-sputtering Pt and Au on n-type planar Si could yield a very low onset potential of approximately 0.05 V vs. RHE. Additionally, GOR enables the production of financially high-value oxidation compounds, such as glyceric acid (GLA).

[0075] A nanowire array structure replacing the planar silicon with the same catalysts (i.e., co-sputtered Pt-Au with an optimized loading amount) was adopted considering its higher surface area. The high surface area photoelectrodes efficiently increased the photocurrent near the onset potential and an onset potential near 0 vs. RHE was shown under the red light. FIG. 7F illustrates a graph 706 plotting photocurrent densities versus bias (vs. RHE) of glycerol oxidation reactions under continuous, chopped, and dark irradiation. [0076] FIG. 9 illustrates a graph 900 plotting the photoelectrochemical performance of planar n-type silicon and nanowire n-type silicon. A 200 mW/cm² red light LED was used as a light source. FIG. 9 also shows that the photoanode exhibits an approximately 0.45 V photovoltage shift relative to a dark Pt-Au on a glassy carbon electrode. Silicon nanowire architecture gives higher photocurrent near the onset voltage around 0 V vs. RHE. The scan rate was ten mV/s. This could be because the higher surface area and higher loading amount of Pt/Au catalysts improved the static current density in the lower potential region where the amount of catalysts limits the overall performance. FIG. 7G illustrates a graph 308 plotting the overlap of a PEC linear scan of bioSiNW photocathode and PtAu SiNW photoanode, where the expected bias-free current is approximately .2 mA/cm².

Bias-free operation with silicon photochemical diodes:

[0077] For the light-driven bias-free dual-electrode PEC configuration, the photoelectrodes were placed into a two-chamber cell separated by a bipolar membrane (BPM), as shown in FIG. 1A. Utilization of a BPM allows the separation of two different electrolytes with a significant pH difference. GOR is optimal in an alkaline environment with 1 M KOH, whereas microbial CO2RR operates at a neutral pH in an MES buffer. FIG. 10A illustrates a graph 1000 plotting photocurrent densities versus bias (vs. RHE) under twenty mW/cm² of 740 LED irradiation, chopped irradiation, and dark. The scan rate is ten mV/s. The linear scan of the dual-electrode configuration of a bioSiNW photocathode 102 and a Pt-Au SiNW photoanode 104 confirms the bias-free current near approximately 1 .2 mA/cm² and the onset potential of -0.8 V, as shown in FIG. 10A.

[0078] FIG. 10B illustrates a graph 1002 showing the long-term operation of the system plotted in FIG. 10A. FIG. 10C illustrates a graph 1004 showing a product analysis of the system plotted in FIG. 10A. As shown in FIGs. 10B and 10C, the bias-free bio CO2RR-GOR system was able to maintain a photocurrent density of 1.2 mA/cm² even after 1 hour (hr). The main cathodic products are acetate with a faradaic efficiency (FE) of 86.8 ± 14.0 %, and the main anodic product is GLA with a faradaic efficiency (FE) of 38.8 ± 8.0%, with total products reaching around FE_totai of 79.3% ± 9.1 %. The remainder of the products (e.g., lactic acid, acetic acid, and formate) likely come from products that are not analyzable through 1 H NMR, such as tartronic acid and carbonate. The production of glyceraldehyde and dihydroxyacetone was not detected, because both glyceraldehyde and dihydroxyacetone rearrange into lactic acid in alkaline conditions, making it difficult to differentiate them from the production of lactic acid. The product distribution is similar to previously observed PEC GOR results using the Pt-Au catalyst under similar conditions. The acetate production rate of the system was 44.8 ± 11 .6 g/m² day and 0.1 g/L day.

[0079] FIG. 11 A illustrates a graph 1100 showing the long-term operation of a system under control (i.e., dark) conditions. As shown in FIG. 11 A, there is almost no photocurrent (<0.05 mA/cm²), less than 0.014 mA/cm², and, consequently, negligible product formulation, indicating that the products discussed above in connection with FIGs. 10A-10C were being produced from photosynthesis. FIG. 11A shows photocurrent densities versus time for an S.ovata/S photocathode 102 and PtAu/Si photoanode 104 configuration under dark.

[0080] FIG. 11 B illustrates a graph 1102 comparing the average photocurrent densities for the S.ovata/S'\ photocathode 102 and PtAu/Si photoanode 104 configuration under irradiated and dark conditions. FIG. 11 C illustrates a graph 1104 comparing the acetate productions for the S.ovata/Si photocathode 102 and PtAu/Si photoanode 104 configuration under irradiated and dark conditions. FIG. 11 D illustrates a graph 1106 comparing the oxygenate productions for the S.ovata/S photocathode 102 and PtAu/Si photoanode 104 configuration under irradiated and dark conditions. Compared to the irradiated condition, showing approximately 1 mA/cm² of photocurrent, there was a negligible current observed. The higher light intensity (e.g., 40 mW/cm² of irradiation) helped to increase the overall operating bias-free photocurrent to 1.6 mA/cm². FIG. 12 illustrates a graph 1200 plotting the photocurrent densities versus time of an S. ovata/Si photocathode 102 and a Pt-Au/Si photoanode 104 configuration under twenty mW/cm² and forty mW/cm² of red light irradiation. [0081] Thus, examples of the present disclosure provide a light-driven CO2 reduction process using a silicon nanowire array decorated with a microbial catalyst (e.g., S. ovata). The nanofabricated silicon photoelectrodes can harvest 0.85 V from low-intensity red light. The introduction of metabolically adapted and active biocatalysts on the photocathode 102 enhance the photocurrent compared to abiotic and wild-type control. This highly selective and efficient biological photocathodic CO2 reduction, with the onset voltage of 0.4 V vs. RHE, can be wired with a complementary photoanodic glycerol oxidation reaction to realize a bias-free PEC device. The integrated PEC device achieves a maximum of 1.2 mA/cm² of photocurrent with an eighty percent faradaic efficiency of C2 product. [0082] To date, no bias-free dual-electrode PEC device is known to exist without any buried photovoltaic component to drive CO2-to-C2 photoelectrochemical reduction from the light intensity of natural sunlight. Altogether, the proposed integrated system can be a practically feasible solution for solar fuel generation by using earth-abundant silicon and self-replicates and self-repairing microbial catalysts.

[0083] Lastly, the present disclosure could be scaled for solar-fuel production and potentially be applied to build a large-scale solar-fuel farm. Materials that might be used in such a large-scale solar fuel farm, including light-absorbing silicon wafers and CO2-reducing bacteria, are inexpensive, highly robust (e.g., stable over months after an operation), and easy to fabricate at large scale. Also, since the disclosed photochemical diode device 100 is powered by a low-intensity of red light, this means that devices fabricated according to the present disclosure could be stacked either vertically or horizontally and irradiated with high-efficiency red-light LEDs, which can be renewably powered from solar panels or grid electricity, as in the case of plant vertical farming.

[0084] FIGs. 13 and 14, for instance, illustrate photographs of example stacking arrangements for the photochemical diode device 100 of FIG. 1 A for a large-scale solar fuel farm. Specifically, FIG. 13 illustrates a photograph of a horizontal stacking arrangement for an array 1300 of photochemical diode devices fabricated according to photochemical diode device 100 of FIG. 1A, while FIG. 14 illustrates a photograph of a vertical stacking arrangement for an array 1400 of photochemical diode devices fabricated according to photochemical diode device 100 of FIG. 1A. A vertical stacking arrangement such as that illustrated in FIG. 14 may be suitable for enhancing the production rates in a confined space of a farming site, while a horizontal stacking arrangement such as that illustrated in FIG. 13 may yield a more streamlined collection of products and provision of reactants. The photochemical diode devices could be continuously run through day and night for valuable chemical generations from low-cost and wasteful feedstocks. The artificial bio-photochemical “leaf” would replace the crops and plants and produce chemicals from, for instance, the greenhouse gas, CO2, and industrial byproduct, glycerol, more efficiently than natural crops.

[0085] FIG. 15 is a flow diagram illustrating one example of a method 1500 for forming a photochemical diode device, according to the present disclosure. Details of the method 1500 are supplied above using information from experimental fabrication processes.

[0086] The method 1500 may begin in step 1502. In step 1504, a p-doped silicon nanowire (p-SiNW) substrate and an n-doped silicon nanowire (n-SiNW) substrate are fabricated.

[0087] In one example, p-type boron-doped six-inch wafers (<100> oriented, 1-30 Ohm-cm, prime, double sided-polished) and n-type phosphorus-doped six- inch Si wafers (<100> oriented, 1-10 Ohm-cm, prime, single sided-polished) were used for fabrication of the photochemical diode 100 of FIG. 1A. The wafers were etched in a 4.9% HF bath for three minutes and washed with deionized (DI) water and dried. In a typical photoresist process, hexamethyldisilazane was applied for two minutes, and MiR 701 photoresist was spin-coated on to have a thickness of one pm.

[0088] The wafers were patterned using a 5x i-line photolithography stepper with a mask that was patterned with a square lattice of 3.75 pm circles and pitch of 10 pm, for final resist pattern of 0.75 pm circles and 2 pm pitch. The resist pattern was developed using MF-26A for sixty seconds, descummed with O2 plasma at fifty W for sixty seconds, and hard baked at 140 °C with UV light. In order to etch the wafer, a low-frequency inductive-coupled plasma deep reaction- ion etch (DRIE) process was used, using O2 and SFe as the etch gas and C4F8 as the passivation gas and a typical DRIE smooth-wall recipe until nanowire lengths of twenty-one pm were achieved. Afterwards, any remaining photoresist was removed with O2 plasma at 250 W for 7.5 minutes.

Fabrication of n⁺p-SiNW and p⁺n-SiNW substrates:

[0089] In one example, a six-inch silicon wafer was used as the dopant carrier wafer. All wafers were cleaned in a 4.9% HF bath for three minutes to remove native oxides and thoroughly washed with DI water and dried. The carrier wafer was spin-coated with either an arsenic silicate spin-on-dopant solution (for n⁺p- SiNW fabrication) or a gallium silicate spin-on-dopant solution (for p⁺n-SiNW fabrication) at 2200 rpm for thirty seconds and baked on a hotplate at 150 °C for thirty minutes.

[0090] Afterwards, p-SiNW and/or n-SiNW substrates, cleaned with HF, water, and acetone right before this process, were placed onto the carrier wafer upside down such that the nanowire surface was interfaced with the dopant layer. The nanowire surface and the dopant layerwere placed into a rapid thermal annealing chamber at 900 °C for either 180 seconds (for n⁺p-SiNW fabrication) or 100 seconds (for p⁺n-SiNW fabrication) under N2. A spin-on-dopant thin film containing a high concentration of arsenic (for n+ doping) in SiC>2 networks was proximately contacted with silicon substrates at high temperatures of over 900 °C for the diffusion of dopants.

[0091] The testing was carried out on a planar substrate first to exclude the complexity of coherent doping on the nanostructured surfaces of nanowires. In one example, the silicon substrate surface should be free of native oxides and vapor molecules to prevent dopant rejections in silicon oxides where the arsenic solubility is much lower than that in silicon. A junction depth, Xj, where the concentrations of shell dopants and substrate dopant, arsenic and boron for n7p- Si are equal, can be calculated based on the theoretical Gaussian distribution of dopants.

where D is the diffusivity of the dopant, t is the diffusion time, N_B is a background concentration, and No is a dopant concentration at the surface. No can be extracted from the theoretical solubility of dopants at the temperature set for the RTA process. For arsenic, No is 2x10²⁰ at 900 °C. For three minutes at 900 °C of the RTA process, Xj for n⁺ layer is approximately 7 nm.

[0092] In optional step 1506 (illustrated in phantom), a protective material layer may be deposited on the p-doped silicon nanowire substrate and the n-doped silicon nanowire substrate.

[0093] In one example, the protective material layer comprises TiO2. The n⁺p- SiNW and p⁺n-SiNW substrates were cleaned in a 4.9% HF bath for three minutes, washed with DI water and acetone, and then dried. Afterwards, a thirty nm Ti©2 layer was deposited at 250 °C using atomic layer deposition and titanium isopropoxide as the precursor for n⁺p-SiNW, and a ten nm TiO2 thin film was deposited at the deposition temperature of 200 °C using tetrakis(dimethylamido)titanium as the precursor for p⁺n-SiNW in order to maintain a stable device performance for extended operation over time. After cooling, these substrates were stored in ambient air until use.

[0094] In step 1508, a metallic catalyst may be deposited on the n-doped silicon nanowire substrate and on the p-doped silicon nanowire substrate (e.g., over the optional protective material layers).

[0095] In one example, the metallic catalyst comprises at least one of Pt and Au. In one example, Pt-Au and Pt catalysts were deposited on the fabricated TiO2/SiNW photoanode or TiO2/SINW photocathode on a multi-target cosputtering system with sputter guns supplied by two 2 kW pulsed DC power supplies and one 1 .5 kW DC power supply. After the deposition of TiC>2 protection layer, approximately three nm of Pt catalyst was sputtered onto n⁺p-SiNW photocathode substrates with thirty seconds of fifty W power applied on the Pt target. Pt-Au catalyst with a thickness of approximately eleven nm was cosputtered onto p⁺n-SiNW photoanode substrates with forty-five seconds of fifty W power on a Pt target and twenty-eight W power on the Au target. Both Pt/n⁺p- SiNW and PtAu/p⁺n-SiNW substrates were stored in ambient air until use. [0096] In step 1510, an electrode is formed on the n-doped silicon nanowire substrate to form a photoanode of a dual-electrode photochemical diode. The electrode may be formed after the metallic catalyst is deposited on the n-doped silicon nanowire substrate.

[0097] In step 1512, an electrode is formed on the p-doped silicon nanowire substrate to form a photocathode of the dual-electrode photochemical diode.

[0098] For both electrodes, the substrates were used as photoanode and photocathode by creating electrically conductive connections to titanium foil. To do this, a Ga-ln eutectic was applied and then scratched on the back of the Si substrate. Later, a quick drying silver paste was applied on the top of the scratched Si substrate and fixed onto the titanium foil using double-sided conductive carbon tape. The electrodes were left to dry for an additional thirty minutes in ambient conditions before being mounted onto the cell for measurements.

Photoelectrochemical (PEC) characterization:

[0099] All experiments were performed within a set of custom-built PEC cells, whose basic structure is shown in FIG. 1 A. The setup was a two-chamber PEC cell, a cathodic chamber with a working electrode and a reference electrode (Ag/AgCI, 1 M KCI), and the Pt wire counter electrode in an anodic chamber. The working electrode was sealed with an X-ring with a contact area of 0.321 cm². A gas inlet and an outlet were embedded for the purpose of gas purging.

[00100] The cathodic and anodic chambers were separated by a bipolar membrane. Each of the cathodic and anodic chambers had a gas inlet/outlet. The purging gas for each chamber can be independently defined. Each chamber contained a quartz window for PEC experiments. During experiments, the setup was left at the optimal growth temperature of S. ovata, which fluctuated between 28 and 30 °C. FE_acetate and J_acetate were both characterized vs. RHE defined as following:

V vs RHE (V) = V vs. Ag/AgCI (V) + 0.237 (V) + 0.059*pH (EQN. 2) [00101] A 740 nm uniform illumination LED with a power of twenty mW/cm² calibrated with a certified silicon photodiode and a Xenon lamp with an intensity of 100 mW/cm² was applied to operate the photoelectrochemical measurements of biohybrid devices.

[00102] For the bias-free operation, the silicon photocathode and photoanode were assembled in the same two-chamber PEC cell with X-rings, the photocathode was used as a working electrode, the photoanode was used as a counter electrode, and the voltage of working electrode was set to 0 V versus a counter electrodes.

[00103] In step 1514, a microbial biocatalyst may be directly interfaced with the electrode of the p-doped silicon nanowire substrate.

[00104] In one example, the microbial biocatalyst is S. ovata, as discussed above. The nanowire-bacteria hybrids were prepared on the SiNW photocathodes with the inorganic medium with fifty mM of MES as an electrolyte and 740 nm irradiation as a photon source. The pH value of the electrolyte was adjusted if needed by adding a corresponding amount of hydrochloric acid or 1 M sodium hydroxide into a buffer, and a digital pH meter was used to measure the pH by taking out five ml of pH-adjusted buffer after approximately thirty minutes of equilibration time. The pH-adjusted electrolyte was added to the acid sterilized cathodic (fifteen ml) and anodic (thirty ml) chambers, respectively. Abiotic chronoamperometry experiments were conducted, typically at approximately 0.3 V vs. RHE forone day, with purging eighty percent, N2/ten percent, H2/ten percent CO2 gas to make an anaerobic environment. In parallel, the revived cells in the betaine medium were autotrophically cultured twice in yeast medium with eighty percent H₂/twenty percent CO2 to adapt the revived cells to autotrophic metabolisms. The methanol-adapted S. ovata was cultured in the yeast media containing two percent methanol as a sole electron donor before the two hydrogen cycles to upregulate the methanol oxidizing paths.

[00105] The method 1500 may end in step 1516. Cathodic Product analysis:

[00106] Liquid products for CO2 reduction from the cathode were quantified after electrolysis by proton NMR (1 H-NMR) spectroscopy with 3- (Trimethylsilyl)propionic-2,2,3,3-d4 acid sodium salt (TMSP-d4) as an internal standard for quantification. Acetate was the sole detectable product of the CO2 reducing metabolism of S. ovata owing to the highly selective metabolic pathway of S. ovata. Twenty volume percent of a D2O-TMSP standard was added to all the catholyte solutions. The standard solution was prepared by adding twenty mg TMSP to twenty-five g of D2O, FE_acetate and the incremental mole of acetic acid is calculated according to:

(EQN. 4) where a represents the conversion factor between acetate concentration and the ratio of the NMR peak area of acetate (1 .8 ppm) to the NMR peak area of TMSP (0 ppm). The conversion factor is determined by the slope of a six-point calibration curve between zero and ten mM. V_eiectroiyte represents the volume of electrolyte (typically approximately 0.7 ml). Before starting the reaction, 0.7 ml of electrolyte was sampled, and the fresh electrolyte was subsequently added back to the electrochemical cell to maintain the total volume of electrolyte.

Anodic Product analysis:

[00107] Liquid products for glycerol oxidation from the anode were quantified after electrolysis by 1 H qNMR spectroscopy with water suppression using dimethyl sulfoxide as an internal reference. Faradaic efficiencies were reported as is without normalization. A relaxation delay of forty-two seconds was used. A standard D₂O-DMSO solution was prepared by adding 400pl of DMSO to 100 g of D2O. A ten volume percent of D2O-DMSO standard was added to the analyte solutions. The same equation for FE_aCetate and n_acetic add was used for anodic product analysis, and the number of electrons used for an electrochemical reaction (e.g., eight electrons for acetate) was modified depending upon the oxidation products.

[00108] Although the above examples are described with respect to a particular microbial catalyst (e.g., S. ovata), it should be noted that other microbial catalysts may be used. For example, the microbial catalyst may be a particular type of microbial microorganism for cathodic reactions. A non-exhaustive list of other possible microbial catalysts is shown in Table 1 , below, along with the reactant and products that can be formed using the photochemical diode device described herein:

Host microorganisms Reactant(s)/Product(s)

methyl-1 -butanol

Ralstonia palustris CO2/PhB, Carotenoids

Thiobaciilus denitrificans NO3F/N2b

Escherichia coli H2O/H2;

Glucose, CO2/Succinate

Saccharomyces hexose/shikimic acid cerevisiae

A. vinelandii N2 & h2O/Nh3

& h2 CO2/formic acid

Cupriavidus necator CCL/methyl ketones, butanediol, ethylene,

PhB, propanol

TABLE 1

[00109] In addition, although the above examples are described with respect to a particular anodic oxidation reaction to oxidize glycerol to form value-added products, it should be noted that other organic oxidation reactions may occur to form other desired products/gases. For example, the other desired anodic reactions may include oxidation of biomasses, glucose, 5-hydroxymethylfurfural (HMF), alcohols (e.g., methanol, ethanol, benzyl alcohol, isopropanol, and the like), diols (e.g., 1 ,3-propanediol, 1 ,2-propanediol), ethylene glycol, urea, polymers (plastics), and the like.

[00110] It will be appreciated that variants of the above-disclosed and other features and functions, or alternatives thereof, may be combined into many other different systems or applications. Various presently unforeseen or unanticipated alternatives, modifications, variations, or improvements therein may be subsequently made by those skilled in the art which are also intended to be encompassed by the following claims.

Claims

What is claimed is:

1. An apparatus comprising: a bio catalyzed cathode to reduce a gas; and an anode electrically connected to the bio catalyzed cathode to perform glycerol valorization.

2. The apparatus of claim 1 , wherein the bio catalyzed cathode and the anode each comprise a respective silicon nanowire array.

3. The apparatus of claim 2, wherein the respective silicon nanowire array of the bio catalyzed cathode is p-doped, and the respective silicon nanowire array of the anode is n-doped.

4. The apparatus of claim 1 , wherein the bio catalyzed cathode comprises a microbial biocatalyst.

5. The apparatus of claim 4, wherein the microbial biocatalyst comprises sporomusa ovata.

6. The apparatus of claim 4, wherein the microbial biocatalyst comprises at least one of: moorella thermoacetica, acidithiobacillus ferrooxidans, Clostridium aceticum, Clostridium ljungdahlii, ralstonia eutropha, ralstonia palustris, thiobacillus denitrificans, escherichia coli, saccharomyces cerevisiae, a. vinelandii, or cupriavidus necator.

7 The apparatus of claim 1 , wherein the bio catalyzed cathode and the anode are activated via a light source.

8. The apparatus of claim 7, wherein the light source irradiates the bio catalyzed cathode and the anode with red light.

9. The apparatus of claim 1 , wherein the cathode is catalyzed with at least one of: platinum or gold.

10. The apparatus of claim 1 , wherein the gas is carbon dioxide.

11. A photochemical diode device comprising: a cathode, comprising: a p-doped silicon nanowire array; and a first electrode formed on the p-doped silicon nanowire array, wherein the first electrode has been catalyzed using a microbial biocatalyst to facilitate reduction of a gas; and an anode electrically connected to cathode, comprising: an n-doped silicon nanowire array, wherein the n-doped silicon nanowire array has been catalyzed with a metallic catalyst; and a second electrode formed on the n-doped silicon nanowire array.

12. The photochemical diode device of claim 11 , wherein the microbial biocatalyst comprises sporomusa ovata.

13. The photochemical diode device of claim 11 , wherein the microbial biocatalyst comprises at least one of: moorella thermoacetica, acidithiobacillus ferrooxidans, Clostridium aceticum, Clostridium ljungdahlii, ralstonia eutropha, ralstonia palustris, thiobacillus denitrificans, escherichia coli, saccharomyces cerevisiae, a. vinelandii, or cupriavidus necator.

14. The photochemical diode device of claim 11 , wherein the metallic catalyst comprises at least one of: platinum or gold.

15. A method of fabricating a dual-electrode photochemical diode, the method comprising: fabricating a p-doped silicon nanowire substrate and a separate n-doped silicon nanowire substrate; depositing a metallic catalyst on the n-doped silicon nanowire substrate; forming an electrode on the n-doped silicon nanowire substrate on which the metallic catalyst has been deposited, to form a photoanode of the dualelectrode photochemical diode; forming an electrode on the p-doped silicon nanowire substrate, to form a photocathode of the dual-electrode photochemical diode; and directly interfacing a microbial biocatalyst with the electrode of the p-doped silicon nanowire substrate.

16. The method of claim 15, further comprising: after the fabricating, but prior to depositing the metallic catalyst, depositing a protective material layer on the p-doped silicon nanowire substrate and the n- doped silicon nanowire substrate.

17. The method of claim 16, wherein the protective material layer comprises titanium oxide.

18. The method of claim 15, wherein the microbial biocatalyst comprises sporomusa ovata.

19. The method of claim 15, wherein the microbial biocatalyst comprises at least one of: moorella thermoacetica, acidithiobacillus ferrooxidans, Clostridium aceticum, Clostridium ljungdahlii, ralstonia eutropha, ralstonia palustris, thiobacillus denitrificans, escherichia coli, saccharomyces cerevisiae, a. vinelandii, or cupriavidus necator.

20. The method of claim 15, wherein the metallic catalyst comprises at least one of: platinum or gold.