Heterogeneous catalyst compositions, methods of preparation thereof, electrodes comprising the same, and methods of use thereof

Abstract

The present disclosure relates, in one aspect, to heterogeneous catalyst compositions comprising at least one transition metal catalyst complex covalently attached to a porous silicon substrate. The present disclosure further relates to methods of preparing the heterogeneous catalyst compositions of the present disclosure, electrodes comprising the same, and methods of use thereof.

Description

[0001]Attorney Docket No.047162-7475WO1(02446) TITLE OF THE INVENTION Heterogeneous Catalyst Compositions, Methods of Preparation Thereof, Electrodes Comprising the Same, and Methods of Use Thereof STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT This invention was made with government support under DE-SC0021173 awarded by the Department of Energy. The government has certain rights in the invention. CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application No.63 / 586,294, filed September 28, 2023, which is incorporated herein by reference in its entirety. BACKGROUND In recent decades, the escalating concerns surrounding the rise in atmospheric carbon dioxide (CO2) levels have spurred urgent efforts to combat the adverse impacts of anthropogenic greenhouse gas emissions. Accordingly, an interest in the development of innovative and efficient carbon capture methods that can curtail the unabated accumulation of CO2 in the atmosphere has arisen. Amid the spectrum of carbon capture technologies, photo- and / or photoelectrochemical reduction of CO2 has emerged as a promising and versatile approach. Such methods permit the direct conversion of CO2into valuable products such as carbon monoxide (CO). This process not only reduces CO2 emissions, but also simultaneously creates valuable chemical feedstocks. Common CO2 reduction methods often rely on heterogeneous or homogeneous catalysts. While heterogeneous catalysts are widely used in industrial settings, the active sites thereof are often ill-defined. Accordingly, an understanding of the structure-function relationship of heterogeneous catalysts sufficient to permit rational improvement is difficult to achieve. Further, homogeneous catalysts are uncommon in industrial processes because of their cost, instability, and the difficulty of product separation. Thus, there is a need in the art for catalysts suitable for CO2 reduction and methods of use thereof. The present disclosure addresses this need. - 1 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) BRIEF SUMMARY In one aspect, the disclosure provides a heterogeneous catalyst composition comprising at least one transition metal catalyst complex covalently attached to a porous silicon substrate. In certain embodiments, the heterogeneous catalyst composition is a composition of formula (I), wherein A, L, m, and Spare defined elsewhere herein: (I). In certain embodiments, the composition is selected from the group consisting of formula (Ia-1), , , (Ib-2), wherein X1, L3, L4, L5, R1a, R1b, L, m, n, o, Spo, and Sphare defined elsewhere herein: , 2). catalyst composition of the disclosure. In certain embodiments, the method comprises (a) contacting an etched, hydrogen-terminated porous silicon (H-porSi) substrate and a transition metal catalyst complex to provide a mixture, wherein the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one non- aromatic, at least partially unsaturated moiety. In certain embodiments, the method comprises (b) heating the mixture to a temperature of about 160 °C. In another aspect, the disclosure provides a method for preparing the heterogeneous catalyst composition of the disclosure. In certain embodiments, the method comprises (a) contacting an at least partially surface oxidized porous silicon substrate and a transition metal catalyst complex to provide a mixture, wherein the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one silatrane - 2 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) moiety. In certain embodiments, the method comprises (b) heating the mixture to a temperature of about 80 °C. In another aspect, the disclosure provides an electrode comprising the heterogeneous catalyst composition of the disclosure. In another aspect, the disclosure provides a method for reducing CO2. In certain embodiments, the method comprises contacting a medium comprising CO2with an electrochemical cell comprising an anode and a cathode, wherein the cathode comprises the heterogeneous catalyst composition of the disclosure. In certain embodiments, the method comprises applying an electric potential between the anode and the cathode. In certain embodiments, the method comprises irradiating the medium. BRIEF DESCRIPTION OF THE FIGURES The drawings illustrate generally, by way of example, but not by way of limitation, various embodiments of the present application. FIGs.1A-1D depict preparation and characterization of freshly etched H-porSi. FIG. 1A provides a schematic depicting certain differences between a flat silicon wafer and a porous silicon wafer. FIGs.1B-1D provide certain exemplary characterization data for H- porSi, including FTIR (FIG.1B) and cross-sectional SEM images (FIGs.1C-1D). FIG.2 provides a XPS survey spectrum of H-porSi. FIG.3A provides a schematic depicting a non-limiting immobilization method for covalent immobilization of a Re catalyst to a porous silicon solid support. FIGs.3B-3D depict exemplary characterization data for 1@porSi, including FTIR (FIG.3B) and the Re 4f region of the XPS spectra thereof. The XPS spectra are of an intact sample, which is representative of the top 10 nm of the surface of 1@porSi, and a powdered sample, which was obtained by scratching the surface of the porous material and mixing, which is representative of the entire pore. FIG.3D provides a cross-sectional SEM image of 1@porSi. FIG.4 depicts an IR spectrum of an exemplary H-porSi immobilized with (Me2bpy)Re(CO)3Cl at 160 °C for 18 hours. The wafer was washed with n-pentane, soaked in 10 mL ethanol for 2 hours, then rinsed with n-pentane. The amount of Re-CO on the surface is significantly lower than the H-porSi functionalized with 1. FIGs.5A-5B provide a full FTIR spectrum of 1@porSi (FIG.5A) and a XPS survey spectrum (FIG.5B) of intact 1@porSi. FIGs.6A-6D provide certain photochemical performance data using 1@porSi as the working electrode. FIG.6A: cyclic voltammograms of a Re catalyst immobilized on porous - 3 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) silicon as the working electrode. Cyclic voltammograms were collected under 1 sun illumination (blue), and dark (black); conditions: 250 mM TBAPF6, 100 mM phenol in MeCN. Ag wire reference electrode, graphite rod counter electrode. FIG.6B: graph depicting product distribution (i.e., H2and CO) as a function of time from a controlled potential electrolysis (CPE) of an exemplary Re-immobilized porous silicon catalyst; conditions: 250 mM TBAPF6in MeCN, CO2, and 100 mM phenol; hybrid electrode WE, graphite rod counter, Ag pseudoreference internally referenced to the ferrocene 0 / 1 couple; potential (- 1.71 V vs. Fc+ / Fc). FIG.6C: graph depicting Faradaic efficiency to H2and CO as a function of time from a controlled potential electrolysis (CPE) of an exemplary Re-immobilized, porous silicon catalyst; conditions: 250 mM TBAPF6in MeCN, CO2, and 100 mM phenol; hybrid electrode WE, graphite rod counter, Ag pseudoreference internally referenced to the ferrocene 0 / 1 couple; potential (-1.71 V vs. Fc+ / Fc). FIG.6D: a bar graph comparing the distribution and amount of product(s) generated after 95 minutes of controlled potential electrolysis (CPE) with an exemplary Re-immobilized, porous silicon catalyst (1@porSi), a freshly etched H-terminated porous silicon wafer (H-por-Si), and a H-terminated porous silicon wafer heated at 160 °C in mesitylene for 18 hours under CO2 (H-porSi heated) and a freshly etched H-terminated porous Si wafer (H-por-Si) under argon. FIG.7 provides a graph showing the total charge passed in CPE of three exemplary Re-immobilized, porous silicon catalysts, demonstrating that the photocathodes are highly reproducible. FIG.8 depicts exemplary FTIR spectra of a heterogeneous catalyst immobilized on porous silicon that displays the reproducibility of catalyst immobilization. FIG.9 provides a graph depicting product distribution (i.e., H2and CO) as a function of time from a controlled potential electrolysis (CPE) for three Re-immobilized, porous silicon catalysts. The photocathodes are highly reproducible. Conditions: 250 mM TBAPF6in MeCN, CO2, and 100 mM phenol; hybrid electrode WE, GC rod counter, Ag pseudoreference internally referenced to the ferrocene 0 / 1 couple; potential (-1.71 V vs. Fc+ / Fc). FIGs.10A-10C provide graphs depicting mass spectrometry ion currents of H2(FIG. 10C), H2 (FIG.10B), and D2 (FIG.10C) for 1@porSi when bulk electrolysis is performed with deuterated phenol. The H2observed is present before the wafer is lowered into solution and before any potential is applied, and thus, without wishing to be bound by any theory, is likely an impurity in the CO2used. HD is only observed after the bulk electrolysis experiment is started, and the amount of HD quickly plateaus. No D2 can be detected. These isotope - 4 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) experiments suggest HD formation from a surface SiH and phenol OD. FIGs.11A-11C provide graphs depicting mass spectrometry ion currents of H2(FIG. 11A), HD (FIG.11B), and D2 (FIG.11C) for H-porSi when bulk electrolysis is performed with deuterated phenol. FIGs.12A-12B provide XPS survey spectra of intact 1@porSi before (FIG.12A) and after (FIG.12B) bulk electrolysis. FIG.12C provides an overlay of IR spectra of 1@porSi before and after bulk electrolysis. FIG.13 depicts an exemplary electrochemical cell comprising a non-limiting, exemplary heterogeneous catalyst composition of the present disclosure. Electrodes and the electrolyte are labeled in the figure. The catholyte and anolyte are separated by an anion exchange membrane (AEM). FIG.14: Comparison betwee hydrogen-terminated porous silicon and thermally oxidized porous silicon. Depicts immobilizing a molecular Re catalyst for photoelectrocatalytic CO2 reduction to CO on hydrogen-terminated porous Si (H-porSi), and further depicts immobilizing molecular Mn catalysts for photoelectrocatalytic CO2 reduction immobilization on thermally oxidized porous Si (SiOx-porSi). FIG.15A: Schematic illustration of the preparation of SiOx-porSi. FIGs.15B-15C: Characterization of SiOx-porSi FTIR (FIG.15B) and Si 2p XPS spectra (FIG.15C). The XPS spectra are of an intact sample, which is representative of the top 10-15 nm of the surface, and a powdered sample, which was obtained by scratching of the porous material and mixing and is representative of the entire pore. FIG.15D: Cross-sectional SEM image and EDS mapping of Si and O in the area highlighted by the yellow rectangle. FIG.16A: Synthesis of Mn complexes (1b-4) with surface attachment groups and control compound 5. FIG.16B: Molecular structure of 4 with ellipsoids drawn at the 35% probability level. FIG.16C: Representative cyclic voltammograms of 3 in MeCN containing 0.25 MnBu4NPF6under Ar, CO2and 100 mM PhOH, and CO2, 2.0 M NEt3and 2.0 MiPrOH. FIG.16D: Performance of 3 in electrocatalytic CO2 reduction in MeCN containing 0.25 MnBu4NPF6in the presence of 100 mM PhOH or 2.0 M NEt3and 2.0 MiPrOH at -1.75 V. FIG. 16E: Comparison of product distribution with FE after forty minutes of CPE using a mixture of 1.0 mM 3, 2.0 M NEt3and 2.0 MiPrOH at various potentials in a CO2saturated solution containing 0.25 MnBu4NPF6 in MeCN. Light blue line depicts current density obtained at 40 minutes in each experiment. FIG.17A: Schematic of the immobilization of 3 and 4 onto SiOx-porSi. FIG.17B: - 5 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Characterization of 3@SiOx-porSi and 4@SiOx-porSi using FTIR spectroscopy. FIGs.17C- 17D: Mn 2p region XPS (FIG.17C), and Si 2p region XPS spectra (FIG.17D) of 3@SiOx- porSi. The XPS spectra are of an intact sample, which is representative of the top 10–15 nm of the surface, and a powdered sample, which was obtained by scratching of the porous material and mixing and is representative of the entire pore. FIG.18A: Cross-sectional SEM image and EDS mapping of Mn, N, Si, and O in the area highlighted by the rectangle. FIG.18B: STEM image of porous Si region. Selected bright spots (likely Mn) are highlighted in circles. FIGs.19A-19C: Electrochemistry of 3@SiOx-porSi as the working electrode in a CO2 saturated solution containing 0.25 MnBu4NPF6in MeCN with 2.0 M NEt3and 2.0 MiPrOH. FIG.19A: Cyclic voltammograms of 3@SiOx-porSi under Ar, CO2, CO2 with 2.0 M NEt3 and 2.0 MiPrOH in the presence of 1 sun illumination. FIG.19B: 1stand 5thscan of 3@SiOx- porSi and 1stof SiOx-porSiΔunder CO2 with 2.0 M NEt3 and 2.0 MiPrOH in the presence of 1 sun illumination. FIG.19C: Linear sweep voltammogram of 3@SiOx-porSi (under 1 sun illumination) and 3@SiOx-porSi-ntype (recorded in the dark) under CO2 with 2.0 M NEt3 and 2.0 MiPrOH. FIGs.20A-20D: Photoelectrocatalytic CO2 reduction using 3@SiOx-porSi and 4@SiOx-porSi as the working electrode in a CO2 saturated solution containing 0.25 MnBu4NPF6in MeCN in the presence of either 100 mM PhOH or 2.0 M NEt3and 2.0 MiPrOH. FIGs.20A-20B: Cumulative FE versus time (FIG.20A) and cumulative amounts of formate, CO and H2(FIG.20B) in a CPE experiment with 2.0 M NEt3and 2.0 MiPrOH at -1.75 V versus Fc+ / Fc using 3@SiOx-porSi under 1 sun illumination. FIG.20C: Comparison of product distribution for CO2reduction with FE after forty minutes of CPE using 3@SiOx- porSi and 4@SiOx-porSi at -1.75 V. Light blue line depicts geometric current density measured at forty minutes. FIG.20D: Comparison of product distribution for CO2reduction with FE after forty minutes of electrolysis using 3@SiOx-porSi as the working electrode with 2.0 M NEt3with 2.0 MiPrOH at various potentials. Blue line depicts geometric current density measured at forty minutes. FIGs.21A-21B: Overlay of FTIR spectra (FIG.21A) and Mn 2p region XPS (FIG. 21B) of 3@SiOx-porSi before and after CPE experiment with 2.0 M NEt3 and 2.0 MiPrOH at -1.75 V versus Fc+ / Fc. FIG.22: Full FTIR spectrum of p-type H-terminated porous Si (H-porSi) with Si-H stretching region in inset. FIG.23: Full FTIR spectrum of n-type H-terminated porous Si (H-porSi-ntype) with - 6 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Si-H stretching region in inset. FIG.24: Cross-sectional SEM image (8,000x magnification) of p-type H-terminated porous Si (H-porSi). FIG.25: XPS survey spectra of intact SiOx-porSi. FIGs.26A-26H: XPS survey spectra of intact and powdered SiOx-porSi; Si 2p region (FIG.26A), O 1s region (FIG.26B), Mn 2p region (FIG.26C), Br 3d region (FIG.26D), N 1s region (FIG.26E), C 1s region (FIG.26F), F 1s region (FIG.26G), and P 2p region (FIG. 26H). FIG.27: BET surface area measurement of SiOx-porSi. For a given BET experiment, at least 5 points were fit between p / porange of 0.05 and 0.2. BET surface area is about 70 m2 / g. FIG.28: Full FTIR spectrum of n-type oxidized porous Si (SiOx-porSi-ntype). FIG.29: Overlay FTIR spectra of three different SiOx-porSi samples etched under identical conditions. The CH and OH stretches present may from residual ethanol used in rinsing the wafers. FIG.30: Surface oxidation of SiOx-porSi under ambient conditions (room temperature, under air) monitored by FTIR spectroscopy between 500-4000 cm-1. FIG.31: Stability of SiOx-porSi under ambient conditions monitored by FTIR spectroscopy over 6 days. Changes in the remaining hydrides and back bonded hydride stretches are minimal (~0.02 absorbance units) suggesting there is no further oxidation. Two peaks between 2300-2400 cm⁻¹ correspond to the absorption of CO₂ in the atmosphere and are not related to any surface species. FIG.32: Homogeneous cyclic voltammograms of 1b-5 under argon in MeCN. Conditions: 1.0 mM [Mn], 100 mMnBu4NPF6, 100 mV / s scan rate, glassy carbon working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. As depicted herein, “1” refers to “1b.” FIG.33: Homogeneous cyclic voltammograms of 1b-5 under CO2in MeCN. Conditions: 1.0 mM [Mn], 100 mMnBu4NPF6, 100 mV / s scan rate, glassy carbon working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. As depicted herein, “1” refers to “1b.” FIG.34: Homogeneous cyclic voltammograms of 1b-5 under CO2with phenol in MeCN. Conditions: 1.0 mM [Mn], 100 mMnBu4NPF6, 100 mM phenol, 100 mV / s scan rate, glassy carbon working electrode, platinum wire counter electrode, silver wire pseudo- reference electrode. As depicted herein, “1” refers to “1b.” - 7 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) FIG.35: Homogeneous cyclic voltammograms of 1b-5 under CO2 with NEt3 andiPrOH. Conditions: 1.0 mM [Mn], 100 mM nBu4NPF6, 2.0 M NEt3, 2.0 MiPrOH, 100 mV / s scan rate, glassy carbon working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. As depicted herein, “1” refers to “1b.” FIG.36: Cyclic voltammogram of Fc and Fc* in the presence or absence of 2.0 M NEt3and 2.0 MiPrOH in MeCN. Conditions: 1.0 mM Fc, 1.0 mM Fc*, 250 mMnBu4NPF6, 100 mV / s scan rate, glassy carbon working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. The reduction potential of Fc* in the absence of NEt3and MiPrOH under CO2 is 0.085 V vs Ag wire [E1 / 2 (noiPrOH & no NEt3) = 0.085V vs Ag wire]. On the other hand, the reduction potential of Fc* in the presence of 2.0 M NEt3and 2.0 MiPrOH under CO2 is 0.151 V vs Ag wire [E1 / 2 (2.0 MiPrOH & 2.0 M NEt3) = 0.151 V vs Ag wire]. Therefore, the redox couple of Fc* (0.654 V vs. Fc+ / Fc) is used as an internal reference. FIGs.37A-37F: Overlay of CPE traces in electrocatalytic CO2reduction with 2.0 M NEt3 and 2.0 MiPrOH at -1.75 V versus Fc+ / Fc using 1b-5. FIGs.37A-37E: Cumulative amounts of formate, CO and H2 from CPE using 1b (FIG.37A), 2 (FIG.37B), 3 (FIG.37C), 4 (FIG.37D), and 5 (FIG.37E). FIG.37F: Faradaic efficiency and current density from CPE using 1b-5. Conditions: 1.0 mM 1b-5, 2.0 M NEt3, 2.0 MiPrOH, 250 mMnBu4NPF6 in MeCN, glassy carbon rod working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. As depicted herein, “1” refers to “1b. FIGs.38A-38D: Cumulative amounts of formate, CO and H2in electrocatalytic CO2reduction with 2.0 M NEt3 and 2.0 MiPrOH at -1.75 V (FIG.38A), -1.95 V (FIG.38B), and - 2.15 V (FIG.38C) versus Fc+ / Fc using 3. FIG.38D: Faradaic efficiency and current density. Conditions: 1.0 mM 3, 2.0 M NEt3, 2.0 MiPrOH, 250 mMnBu4NPF6 in MeCN, glassy carbon rod working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. FIGs.39A-39D: Cumulative amounts of formate, CO and H2in electrocatalytic CO2reduction with 2.0 M NEt3 and 2.0 MiPrOH at -1.75 V (FIG.39A), -1.95 V (FIG.39B), and - 2.15 V (FIG.39C) versus Fc+ / Fc using 4. FIG.39D: Faradaic efficiency and current density. Conditions: 1.0 mM 4, 2.0 M NEt3, 2.0 MiPrOH, 250 mMnBu4NPF6 in MeCN, glassy carbon rod working electrode, platinum wire counter electrode, silver wire pseudoreference electrode FIGs.40A-40D: Cumulative amounts of formate, CO and H2in electrocatalytic CO2reduction with 2.0 M NEt3 and 2.0 MiPrOH at -1.75 V (FIG.40A), -1.95 V (FIG.40B), and - - 8 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) 2.15 V (FIG.40C) versus Fc+ / Fc using 5. FIG.40D: Faradaic efficiency and current density. Conditions: 1.0 mM 5, 2.0 M NEt3, 2.0 MiPrOH, 250 mMnBu4NPF6in MeCN glassy carbon rod working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. FIGs.41A-41C: Overlay of CPE traces in electrocatalytic CO2 reduction with 100 mM phenol at -1.75 V versus Fc+ / Fc using 3 and 4. Cumulative amounts of formate, CO and H2: 3 (FIG.41A) and 4 (FIG.41B). FIG.41C: Faradaic efficiency and current density: 3 and 4. Conditions: 1.0 mM 3 or 4, 100 mM phenol, 250 mMnBu4NPF6in MeCN, glassy carbon rod working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. FIGs.42A-42D: Cumulative amounts of formate, CO and H2 in electrocatalytic CO2 reduction with 100 mM phenol at -1.75 V (FIG.42A), -1.95 V (FIG.42B), and 2.15 V (FIG. 42C) versus Fc+ / Fc using 3. FIG.42D: Faradaic efficiency and current density. Conditions: 1.0 mM 3, 100 mM phenol, 250 mMnBu4NPF6in MeCN, glassy carbon rod working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. FIGs.43A-43D: Cumulative amounts of formate, CO and H2 in electrocatalytic CO2 reduction with 100 mM phenol at -1.75 V (FIG.43A), -1.95 V (FIG.43B), and -2.15 V (FIG. 43C) versus Fc+ / Fc using 4. FIG.43D: Faradaic efficiency and current density. Conditions: 1.0 mM 4, 100 mM phenol, 250 mMnBu4NPF6in MeCN, glassy carbon rod working electrode, platinum wire counter electrode, silver wire pseudoreference electrode. FIG.44: Overlay of FTIR spectra of 1) SiOx-porSi and 2) 3@SiOx-porSi. FIG.45: Overlay of FTIR spectra of 1) SiOx-porSi and 2) 4@SiOx-porSi. FIG.46: XPS survey spectra of intact 3@SiOx-porSi. FIGs.47A-47H: XPS survey spectra of intact and powdered 3@SiOx-porSi; Si 2p region (FIG.47A), O 1s region (FIG.47B), Mn 2p region (FIG.47C), Br 3d region (FIG. 47D), N 1s region (FIG.47E), C 1s region (FIG.47F), F 1s region (FIG.47G), and P 2p region (FIG.47H). FIG.48: XPS survey spectra of intact 4@SiOx-porSi. FIGs.49A-49H: XPS survey spectra of intact and powdered 4@SiOx-porSi; Si 2p region (FIG.49A), O 1s region (FIG.49B), Mn 2p region (FIG.49C), Br 3d region (FIG. 49D), N 1s region (FIG.49E), C 1s region (FIG.49F), F 1s region (FIG.49G), and P 2p region (FIG.49H). FIG.50: Cross-sectional SEM and EDS mapping of Mn, N, Si, and O for 4@SiOx- porSi. - 9 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) FIG.51: Overlay of FTIR spectra of 3@SiOx-porSi before and after digestion. The disappearance of Mn-CO stretches suggests that 3 has been completely removed from the porous Si. FIG.52: Overlay of FTIR spectra of 4@SiOx-porSi before and after digestion. The disappearance of Mn-CO stretches suggests that 4 has been completely removed from the porous Si. FIG.53: Full FTIR spectrum of 3@SiOx-porSi-ntype. FIG.54: Overlay FTIR spectra of: 1) drop-casted 3 on a SiOx-porSi; 2) MeCN- washed drop-casted 3 on H-porSi; It is notable that after washing a drop-casted sample with MeCN, no signals associated with 3 are observed in the IR spectrum, suggesting that any physisorbed 3 is washed away. FIG.55: FTIR spectrum of a SiOx-porSi wafer after attempted immobilization with 5 through heating in MeCN under N2 at 80 °C for 18 hours. The wafer was washed with MeCN, soaked in ethanol for 30 min and then rinsed with pentane. The negligible amount of Mn on the surface is confirmed by the lack of carbonyl stretches. FIG.56: Illustration of how electrical contacts are added to porous Si-based hybrid wafers, so that they can function as electrodes for electrochemistry. FIG.57: Schemetic of the H-cell used for all electrochemical experiments. FIGs.58A-58B: Cyclic voltammograms of 3@SiOx-porSi under Ar (FIG.58A) and CO2 (FIG.58B) with 1 sun illumination. Conditions: 250 mMnBu4NPF6 in MeCN under 1 atm Ar or CO2and 100 mV / s scan rate.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. The solution was not stirred. FIGs.59A-59B: Cyclic voltammograms of 3@SiOx-porSi under CO2, in the presence of 2.0 M NEt3 and 2.0 MiPrOH in MeCN without (FIG.59A) or with (FIG.59B) 1 sun illumination. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2 and 100 mV / s scan rate.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIG.60: Cyclic voltammograms of 4@SiOx-porSi under CO2, in the presence of 2.0 M NEt3and 2.0 MiPrOH in MeCN without and with 1 sun illumination. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 and 100 mV / s scan rate. 4@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIGs.61A-61B: Cyclic voltammograms of 3@SiOx-porSi (FIG.61A) and 4@SiOx- porSi (FIG.61B) under CO2, in the presence of 100 mM phenol in MeCN without or with 1 - 10 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) sun illumination. Conditions: 250 mMnBu4NPF6 and 100 mM phenol in MeCN under 1 atm CO2and 100 mV / s scan rate.3@SiOx-porSi or 4@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. The solution was not stirred. FIGs.62A-62B: Cyclic voltammograms of SiOx-porSi^, without attached catalyst, under CO2in the presence of 100 mM (FIG.62A) phenol or 2.0 M NEt3(FIG.62B) and 2.0 MiPrOH in MeCN without or with 1 sun illumination. The SiOx-porSi^wafer was prepared by submerging a freshly prepared SiOx-porSi wafer in a 5 mL MeCN solution without catalyst in a round bottom flask. The solution was heated under N2 at 70 °C for 18 hours. After allowing the solution to cool, the wafer was soaked in 10 mL ethanol and then rinsed with pentane. Conditions: 250 mMnBu4NPF6, 100 mM phenol or 2.0 M NEt3 with 2.0 MiPrOH, in MeCN under 1 atm CO2 and at 100 mV / s scan rate. SiOx-porSi^working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. The solution was not stirred. FIG.63: Linear sweep voltammograms of 3@SiOx-porSi under CO2, in the presence of 2.0 M NEt3and 2.0 MiPrOH in MeCN under different light intensities (from dark to 1 sun) with stiring at 1200 rpm. The different intensities were run in a randomized order to minimize systematic error. The general shapes of the curves from the linear sweep voltammogram experiments are similar to those obtained from cyclic voltammetry with no distinct redox features associated with 3 observed. Increased light intensity increases the linear sweep voltammogram current, though not linearly, as the geometric current density at 0.05 sun is approximately half of geometric current density at 1 sun. Thus, the intensity of light affects the rates of the photoelectrochemical processes. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2and 100 mV / s scan rate.3@SiOx- porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIG.64: Representative GC-FID (top) and GC-TCD (bottom) trace of the headspace from the working compartment of a CPE experiment using 3@SiOx-porSi under CO2 after 2 hours. CH4* is the internal standard. The presence of O2and N2is likely due to a leak in the H-Cell and / or in the syringe during sampling. The amount of CO (1.16 ^mol) was determined using the area ratio of CO vs. CH4from FID trace and H2(3.74 ^mol) was calculated using the area ratio of H2 (from TCD) vs. CH4 (from FID). Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -1.95 V vs Fc+ / Fc.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. - 11 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) FIGs.65A-65C: CPE experiment with 3@SiOx-porSi at -1.55 V versus Fc+ / Fc; Charge passed (FIG.65A), Cumulative amounts of H2, CO and formate versus time (FIG. 65B) and FE at each individual time point versus time (FIG.65C). Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2with 1 sun illumination at -1.55 V vs Fc+ / Fc.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIGs.66A-66C: CPE experiment with 3@SiOx-porSi at -1.75 V versus Fc+ / Fc. Charge passed (FIG.66A), Cumulative amounts of H2, CO and formate versus time (FIG. 66B) and FE at each individual time point versus time (FIG.66C). Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2with 1 sun illumination at -1.75 V vs Fc+ / Fc.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIGs.67A-67C: CPE experiment with 3@SiOx-porSi at -1.95 V versus Fc+ / Fc. Charge passed (FIG.67A), Cumulative amounts of H2, CO and formate versus time (FIG. 67B) and FE at each individual time point versus time (FIG.67C). Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -1.95 V vs Fc+ / Fc.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIGs.68A-68C: CPE experiment with 3@SiOx-porSi at -2.15 V versus Fc+ / Fc. Charge passed (FIG.68A), Cumulative amounts of H2, CO and formate versus time (FIG. 68B) and FE at each individual time point versus time (FIG.68C). Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -2.15 V vs Fc+ / Fc.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIGs.69A-69C: CPE experiment with 4@SiOx-porSi at -1.75 V versus Fc+ / Fc. Charge passed (FIG.69A), Cumulative amounts of H2, CO and formate versus time (FIG. 69B) and FE at each individual time point versus time (FIG.69C). Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -1.75 V vs Fc+ / Fc.4@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference reference electrode. FIGs.70A-70C: CPE experiment with 4@SiOx-porSi at -1.95 V versus Fc+ / Fc. Charge passed (FIG.70A), Cumulative amounts of H2, CO and formate versus time and (c) FE at each individual time point versus time. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -1.95 V vs Fc+ / Fc. - 12 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) 4@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference reference electrode. FIGs.71A-71C: CPE experiment with 4@SiOx-porSi at -2.15 V versus Fc+ / Fc; Charge passed (FIG.71A), Cumulative amounts of H2, CO and formate versus time (FIG. 71B), and FE at each individual time point versus time (FIG.71C). Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2with 1 sun illumination at -2.15V vs Fc+ / Fc.4@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIG.72: Product distribution after 40 min in a CPE experiment with 4@SiOx-porSi with 2.0 M NEt3and 2.0 MiPrOH at various potentials in CO2-saturated 250 mMnBu4NPF6in MeCN. FIGs.73A-73C: CPE experiment with 3@SiOx-porSi at -1.75 V versus Fc+ / Fc; Charge passed (FIG.73A), Cumulative amounts of H2, CO and formate versus time (FIG. 73B), and FE at each individual time point versus time (FIG.73C). Conditions: 250 mMnBu4NPF6 and 100 mM phenol in MeCN under 1 atm CO2 with 1 sun illumination at -1.75 V vs Fc+ / Fc.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference electrode. FIGs.74A-74C: CPE experiment with 4@SiOx-porSi at -1.75 V versus Fc+ / Fc; Charge passed (FIG.74A), Cumulative amounts of H2, CO and formate versus time (FIG. 74B), and FE at each individual time point versus time (FIG.74C). Conditions: 250 mMnBu4NPF6and 100 mM phenol in MeCN under 1 atm CO2with 1 sun illumination at -1.75 V vs Fc+ / Fc.4@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire reference electrode. FIGs.75A-75C: Reproducible catalytic results over three trials; Charge passed (FIG. 75A), Cumulative amounts of H2, CO and formate versus time (FIG.75B), and FE at each individual time point versus time (FIG.75C) in a CPE experiment with 3@SiOx-porSi at −1.75 V versus Fc+ / Fc. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -1.75 V vs Fc+ / Fc.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference reference electrode. FIGs.76A-76D: Characterization and photocatalytic CO2 reduction performance using control wafer SiOx-porSi^. FIG.76A: FTIR spectra of SiOx-porSi^without attached catalyst. FIGs.76B-76D: Charge passed (FIG.76B), Cumulative amounts of H2, CO and formate versus time (FIG.76C) and FE at each individual time point versus time (FIG.76D) - 13 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) in a CPE experiment with SiOx-porSi^at −1.75 V versus Fc+ / Fc. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -1.75 V vs Fc+ / Fc. SiOx-porSi^working electrode, graphite rod counter electrode, Ag wire pseudoreference reference electrode. FIG.77A: FTIR spectra of SiOx-porSi^without attached catalyst. A SiOx-porSi^wafer was submerged in a 5 mL MeCN solution containing nothing in a round bottom flask. The solution was heated under N2 at 70 °C for 18 hours. After allowing the solution to cool, the wafer was soaked in 10 mL ethanol and then rinsed with pentane. FIGs.77B-77D: Charge passed (FIG.77B), Cumulative amounts of H2, CO and formate versus time (FIG.77C) and FE at each individual time point versus time (FIG.77D) in a CPE experiment with SiOx- porSi^at -1.75 V versus Fc+ / Fc. Conditions: 250 mMnBu4NPF6and 100 mM phenol in MeCN under 1 atm CO2 with 1 sun illumination at -1.75 V vs Fc+ / Fc. SiOx-porSi^working electrode, graphite rod counter electrode, Ag wire pseudoreference reference electrode. FIG.78A: Cyclic voltammograms of 3@SiOx-porSi under CO2, in the presence of 2.0 M NEt3 and 2.0 MiPrOH without or with the red-light excitation. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3and 2.0 MiPrOH in MeCN under 1 atm CO2and 100 mV / s scan rate. 3@SiOx-porSi, graphite rod counter electrode, Ag wire reference electrode. The solution was not stirred. FIGs.78B-78D: Charge passed (FIG.78B), Cumulative amounts of H2, CO and formate versus time (FIG.78C) and FE at each individual time point versus time (FIG.78D) in a CPE experiment with 3@SiOx-porSi at −1.75 V versus Fc+ / Fc. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 with red-light excitation at -1.75 V vs Fc+ / Fc.3@SiOx-porSi, graphite rod counter electrode, Ag wire pseudoreference reference electrode. The FE for formate was ~ 96%, with almost no detectable H2 or CO generated, which is the same as under the standard conditions with 1 sun illumination, with the only change being slightly lower geometric current density. FIG.79: Geometric current density vs. time plot in the photoelectrocatalytic reduction of CO2 using 3@SiOx-porSi under constant potential electrolysis (CPE) with the light intensity changed every 100 s. One hour CPE under 1 sun illumination was performed prior to this experiment, in order to reach steady state current (depleting the redox-active surface species and obtaining a flat current to time plot). These data indicate that the intensity of light affects the rate of CO2 reduction. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 with 1 sun illumination at -1.75 V vs Fc+ / Fc with 1200 rpm stirring.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire reference - 14 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) electrode. FIG.80: Amount of formate versus time and in a CPE experiment with 3@SiOx-porSi and 50 ^mol of TBA formate at -2.15 V versus Fc+ / Fc. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm Ar.3@SiOx-porSi working electrode, graphite rod counter electrode, Ag wire pseudoreference reference electrode. The amount of formate stays at 50 ^mol with no detectable other carbon product (by GC and NMR spectroscopy) generated, indicating that the formate is stable under this condition. FIG.81A: FTIR spectra of H-porSi and SiOx-porSi after heating at 190 °C for 7h and 18h. The amount of Si-oxide increased, as evidenced by the νSi-Oxstretch. FIG.81B: FTIR spectra of 3@SiOx-porSi through immobilization of 3 on SiOx-porSi wafer heated at 190 °C for 7h and 18h. FIG.82A: Cyclic voltammograms of 3@thick-SiOx-porSi under CO2, in the presence of 2.0 M NEt3and 2.0 MiPrOH without or with 1 sun illumination. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2 and 100 mV / s scan rate. 3@thick-SiOx-porSi, graphite rod counter electrode, Ag wire pseudoreference reference electrode. The solution was not stirred. FIGs.82B-82D: Charge passed (FIG.82B), Cumulative amounts of H2, CO and formate versus time (FIG.82C) and FE at each individual time point versus time (FIG.82D) in a CPE experiment with 3@thick-SiOx-porSi at −1.95 V versus Fc+ / Fc. Conditions: 250 mMnBu4NPF6, 2.0 M NEt3 and 2.0 MiPrOH in MeCN under 1 atm CO2with 1 sun illumination at -1.95 V vs Fc+ / Fc.3@thick-SiOx-porSi, graphite rod counter electrode, Ag wire reference electrode. FIG.83: Full FTIR spectrum of 3@SiOx-porSi after CPE. The intensity of the CO stretches was used to estimate that 86% of the Mn initially attached is still present in the form of (Rbpy)Mn(CO)3X (X = solvent or atom form support) after electrolysis. FIG.84: Overlay of FTIR spectra of 4@SiOx-porSi before and after CPE. FIG.85: XPS survey spectra of intact 3@SiOx-porSi after CPE. FIGs.86A-86H: XPS survey spectra of intact and powdered 3@SiOx-porSi after CPE; Si 2p region (FIG.86A), O 1s region (FIG.86B), Mn 2p region (FIG.86C), Br 3d region (FIG.86D), N 1s region (FIG.86E), C 1s region (FIG.86F), F 1s region (FIG.86G), and P 2p region (FIG.86H). FIG.87: XPS survey spectra of intact 4@SiOx-porSi after CPE. FIGs.88A-88H: XPS survey spectra of intact and powdered 4@SiOx-porSi after CPE; Si 2p region (FIG.88A), O 1s region (FIG.88B), Mn 2p region (FIG.88C), Br 3d region - 15 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) (FIG.88D), N 1s region (FIG.88E), C 1s region (FIG.88F), F 1s region (FIG.88G), and P 2p region (FIG.88H). FIG.89: XPS spectra of Mn on the intact sample (representing top 10 nm of 3@SiOx- porSi after CPE) and on the powdered sample (representing all of 3@SiOx-porSi) after CPE. Both show a characteristic Mn(I) doublet indicating that 3 is attached throughout the porous structure and remains to some extent after CPE. Additionally, some of the Mn present in the intact sample is present in a higher oxidation state, but this is not present in the powdered sample. This suggests that some surface oxidation of Mn is occurring. FIG.90: Cross-sectional SEM and EDS mapping of Mn, N, Si, and O on 3@SiOx- porSi after CPE. FIG.91: Cross-sectional SEM and EDS mapping of Mn, N, Si, and O on 4@SiOx- porSi after CPE. DETAILED DESCRIPTION OF THE INVENTION Reference will now be made in detail to certain embodiments of the disclosed subject matter, examples of which are illustrated in part in the accompanying drawings. While the disclosed subject matter will be described in conjunction with the enumerated claims, it will be understood that the exemplified subject matter is not intended to limit the claims to the disclosed subject matter. Throughout this document, values expressed in a range format should be interpreted in a flexible manner to include not only the numerical values explicitly recited as the limits of the range, but also to include all the individual numerical values or sub-ranges encompassed within that range as if each numerical value and sub-range is explicitly recited. For example, a range of “about 0.1% to about 5%” or “about 0.1% to 5%” should be interpreted to include not just about 0.1% to about 5%, but also the individual values (e.g., 1%, 2%, 3%, and 4%) and the sub-ranges (e.g., 0.1% to 0.5%, 1.1% to 2.2%, 3.3% to 4.4%) within the indicated range. The statement “about X to Y” has the same meaning as “about X to about Y,” unless indicated otherwise. Likewise, the statement “about X, Y, or about Z” has the same meaning as “about X, about Y, or about Z,” unless indicated otherwise. In this document, the terms “a,” “an,” or “the” are used to include one or more than one unless the context clearly dictates otherwise. The term “or” is used to refer to a nonexclusive “or” unless otherwise indicated. The statement “at least one of A and B” or “at least one of A or B” has the same meaning as “A, B, or A and B.” In addition, it is to be understood that the phraseology or terminology employed herein, and not otherwise defined, - 16 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) is for the purpose of description only and not of limitation. Any use of section headings is intended to aid reading of the document and is not to be interpreted as limiting; information that is relevant to a section heading may occur within or outside of that particular section. All publications, patents, and patent documents referred to in this document are incorporated by reference herein in their entirety, as though individually incorporated by reference. In the methods described herein, the acts can be carried out in any order, except when a temporal or operational sequence is explicitly recited. Furthermore, specified acts can be carried out concurrently unless explicit claim language recites that they be carried out separately. For example, a claimed act of doing X and a claimed act of doing Y can be conducted simultaneously within a single operation, and the resulting process will fall within the literal scope of the claimed process. Description In one aspect, the present disclosure relates to the development of a photo- or photoelectrocatalyst suitable for reduction of CO2 to one or more valuable products (e.g., formate, formic acid, carbon monoxide (CO), and / or hydrogen). However, as indicated elsewhere herein, common catalysts (i.e., heterogeneous and homogeneous catalysts) each have limitations that hinder industrial application to CO2 reduction processes. Thus, in one aspect, the solution to the technical problem provided by the present application relates to the development of immobilized, well-defined catalyst compositions comprising a light absorbing semiconductor (e.g., silicon) to create hybrid photoelectrodes. In certain aspects, such materials provide the benefits of both homogeneous and heterogeneous catalysts, without the associated limitations. In certain embodiments, such a hybrid material comprises a molecular catalyst (e.g., transition metal complex) covalently tethered to the surface of an immobilized light absorbing semiconductor (e.g., silicon wafer). However, flat silicon substrates (i.e., silicon wafers) have a limited surface area, which significantly hampers the surface coverage of the catalyst and may result in poor selectivity and a lack of catalytic activity and / or efficiency. Thus, in another aspect, the present disclosure relates to a solution to the technical problem of hybrid photoelectrodes with poor catalytic activity and / or selectivity for the reduction of CO2. To address the limitations associated with low catalyst coverage on flat silicon wafers, a porous silicon substrate was prepared by etching, thereby increasing the surface area 106-fold. In certain embodiments, porous silicon substrates have hydride terminated surfaces (H-porSi) or are modified to comprise surface oxidation (SiOx-porSi). In - 17 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) certain embodiments, such surfaces are functionalized (e.g., by condensation with silatrane- functionalized catalyst complexes and / or by hydrosilylation with alkene-functionalized catalyst complexes) to provide the immobilized catalysts of the disclosure with favorable properties for photoelectrocatalytic applications. Thus, in one aspect, the disclosure describes a platform for preparing silicon- immobilized photoelectrocatalyst materials and further describes the design, preparation, and application of exemplary heterogeneous catalyst compositions for the reduction of CO2. Definitions The term “about” as used herein can allow for a degree of variability in a value or range, for example, within 10%, within 5%, or within 1% of a stated value or of a stated limit of a range, and includes the exact stated value or range. The term “alkenyl” as used herein refers to straight and branched chain and cyclic alkyl groups as defined herein, except that at least one double bond exists between two carbon atoms. Thus, alkenyl groups have from 2 to 40 carbon atoms, or 2 to about 20 carbon atoms, or 2 to 12 carbon atoms or, in some embodiments, from 2 to 8 carbon atoms. Examples include, but are not limited to vinyl, -CH=C=CCH2, -CH=CH(CH3), - CH=C(CH3)2, -C(CH3)=CH2, -C(CH3)=CH(CH3), -C(CH2CH3)=CH2, cyclohexenyl, cyclopentenyl, cyclohexadienyl, butadienyl, pentadienyl, and hexadienyl among others. The term “alkoxy” as used herein refers to an oxygen atom connected to an alkyl group, including a cycloalkyl group, as are defined herein. Examples of linear alkoxy groups include but are not limited to methoxy, ethoxy, propoxy, butoxy, pentyloxy, hexyloxy, and the like. Examples of branched alkoxy include but are not limited to isopropoxy, sec-butoxy, tert-butoxy, isopentyloxy, isohexyloxy, and the like. Examples of cyclic alkoxy include but are not limited to cyclopropyloxy, cyclobutyloxy, cyclopentyloxy, cyclohexyloxy, and the like. An alkoxy group can include about 1 to about 12, about 1 to about 20, or about 1 to about 40 carbon atoms bonded to the oxygen atom, and can further include double or triple bonds, and can also include heteroatoms. For example, an allyloxy group or a methoxyethoxy group is also an alkoxy group within the meaning herein, as is a methylenedioxy group in a context where two adjacent atoms of a structure are substituted therewith. The term “alkynyl” as used herein refers to straight and branched chain alkyl groups, except that at least one triple bond exists between two carbon atoms. Thus, alkynyl groups have from 2 to 40 carbon atoms, 2 to about 20 carbon atoms, or from 2 to 12 carbons or, in - 18 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) some embodiments, from 2 to 8 carbon atoms. Examples include, but are not limited to – C ^CH, -C ^C(CH3), -C ^C(CH2CH3), -CH2C ^CH, -CH2C ^C(CH3), and -CH2C ^C(CH2CH3) among others. The term “alkyl” as used herein refers to straight chain and branched alkyl groups and cycloalkyl groups having from 1 to 40 carbon atoms, 1 to about 20 carbon atoms, 1 to 12 carbons or, in some embodiments, from 1 to 8 carbon atoms. Examples of straight chain alkyl groups include those with from 1 to 8 carbon atoms such as methyl, ethyl, n-propyl, n- butyl, n-pentyl, n-hexyl, n-heptyl, and n-octyl groups. Examples of branched alkyl groups include, but are not limited to, isopropyl, iso-butyl, sec-butyl, t-butyl, neopentyl, isopentyl, and 2,2-dimethylpropyl groups. As used herein, the term “alkyl” encompasses n-alkyl, isoalkyl, and anteisoalkyl groups as well as other branched chain forms of alkyl. Representative substituted alkyl groups can be substituted one or more times with any of the groups listed herein, for example, amino, hydroxy, cyano, carboxy, nitro, thio, alkoxy, and halogen groups. The term “amine” as used herein refers to primary, secondary, and tertiary amines having, e.g., the formula N(group)3wherein each group can independently be H or non-H, such as alkyl, aryl, and the like. Amines include but are not limited to R-NH2, for example, alkylamines, arylamines, alkylarylamines; R2NH wherein each R is independently selected, such as dialkylamines, diarylamines, arylalkylamines, heterocyclylamines and the like; and R3N wherein each R is independently selected, such as trialkylamines, dialkylarylamines, alkyldiarylamines, triarylamines, and the like. The term “amine” also includes ammonium ions as used herein. The term “aryl” as used herein refers to cyclic aromatic hydrocarbon groups that do not contain heteroatoms in the ring. Thus, aryl groups include, but are not limited to, phenyl, azulenyl, heptalenyl, biphenyl, indacenyl, fluorenyl, phenanthrenyl, triphenylenyl, pyrenyl, naphthacenyl, chrysenyl, biphenylenyl, anthracenyl, and naphthyl groups. In some embodiments, aryl groups contain about 6 to about 14 carbons in the ring portions of the groups. Aryl groups can be unsubstituted or substituted, as defined herein. Representative substituted aryl groups can be mono-substituted or substituted more than once, such as, but not limited to, a phenyl group substituted at any one or more of 2-, 3-, 4-, 5-, or 6-positions of the phenyl ring, or a naphthyl group substituted at any one or more of 2- to 8-positions thereof. Use of the term “aryl” in combination with another term (e.g., iodide, chloride, boronic acid, and magnesium halide, inter alia), indicates that the aryl group is substituted at - 19 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) one or more positions with the substituent defined by the term used in the combination. For example, an aryl chloride indicates that the aryl is substituted with at least one chloride. The term “catalyst complex” as used herein refers to a composition comprising a catalytically active transition metal in a coordination complex with one or more ligands and / or ions that surround and stabilize the metal center via bonding interactions. The bonding interactions between functional groups of the ligands and / or ions can be ionic, covalent, electrostatic, or it can involve other bonding forces such as coordination with nonbonding electrons, Van der Waals forces, and the like. In one non-limiting example,bpyRe(CO)3Cl is a catalyst complex comprising a Re metal center, wherein coordination of 2,2’-bipyridyl, CO, and Cl ligands stabilizes the transition metal complex. As used herein, the term “contacting” refers to bringing a first molecule, for example, a functionalized transition metal complex, and a second molecule, for example, a surface- etched porous silicon substrate, together in a manner that permits direct physical interaction of the molecules and promotes reactivity. The term “counter anion” as used herein refers to a negatively charged ion that accompanies a cationic species (i.e. positively charged ion) in order to maintain electric neutrality. For example, the chloride ion (Cl-) is the counter anion to sodium (Na+) in NaCl. Non-limiting examples of counter anions include F-, Cl-, Br-, I-, H-, N3-, PhCH=CHCH2-, PhCH-CH=CH2, F3CS(=O)2O- (OTf), (F3CS(=O)2)2N- (NTf2), F3CC(=O)O- (TFA), BF4-, and PF6-. As used herein, the term “covalent bond” refers to a form of chemical bonding that is characterized by the sharing of one or more pairs of electrons between atoms. The term “cycloalkyl” as used herein refers to cyclic alkyl groups such as, but not limited to, cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl, cycloheptyl, and cyclooctyl groups. In some embodiments, the cycloalkyl group can have 3 to about 8-12 ring members, whereas in other embodiments the number of ring carbon atoms range from 3 to 4, 5, 6, or 7. Cycloalkyl groups further include polycyclic cycloalkyl groups such as, but not limited to, norbornyl, adamantyl, bornyl, camphenyl, isocamphenyl, and carenyl groups, and fused rings such as, but not limited to, decalinyl, and the like. Cycloalkyl groups also include rings that are substituted with straight or branched chain alkyl groups as defined herein. Representative substituted cycloalkyl groups can be mono-substituted or substituted more than once, such as, but not limited to, 2,2-, 2,3-, 2,4- 2,5- or 2,6-disubstituted cyclohexyl groups or mono-, di- or tri-substituted norbornyl or cycloheptyl groups, which can be substituted with, for example, amino, hydroxy, cyano, carboxy, nitro, thio, alkoxy, and halogen groups. The term - 20 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) “cycloalkenyl” alone or in combination denotes a cyclic alkenyl group. The terms “halo,” “halogen,” or “halide” group, as used herein, by themselves or as part of another substituent, mean, unless otherwise stated, a fluorine, chlorine, bromine, or iodine atom. The term “haloalkyl” group, as used herein, includes mono-halo alkyl groups, poly- halo alkyl groups wherein all halo atoms can be the same or different, and per-halo alkyl groups, wherein all hydrogen atoms are replaced by halogen atoms, such as fluoro. Examples of haloalkyl include trifluoromethyl, 1,1-dichloroethyl, 1,2-dichloroethyl, 1,3-dibromo-3,3- difluoropropyl, perfluorobutyl, and the like. The term “heteroaralkynyl” as used herein refers to alkynyl groups as defined herein in which a hydrogen or carbon bond of an alkynyl group is replaced with a bond to a heteroaryl group as defined herein. Representative arylalkynyl groups include, but are not limited to, 2-ethynylpyridine and 2-ethynylthiophene. The term “heteroaryl” as used herein refers to aromatic ring compounds containing 5 or more ring members, of which, one or more is a heteroatom such as, but not limited to, N, O, and S; for instance, heteroaryl rings can have 5 to about 8-12 ring members. A heteroaryl group is a variety of a heterocyclyl group that possesses an aromatic electronic structure. A heteroaryl group designated as a C2-heteroaryl can be a 5-ring with two carbon atoms and three heteroatoms, a 6-ring with two carbon atoms and four heteroatoms and so forth. Likewise, a C4-heteroaryl can be a 5-ring with one heteroatom, a 6-ring with two heteroatoms, and so forth. The number of carbon atoms plus the number of heteroatoms sums up to equal the total number of ring atoms. Heteroaryl groups include, but are not limited to, groups such as pyrrolyl, pyrazolyl, triazolyl, tetrazolyl, oxazolyl, isoxazolyl, thiazolyl, pyridinyl, thiophenyl, benzothiophenyl, benzofuranyl, indolyl, azaindolyl, indazolyl, benzimidazolyl, azabenzimidazolyl, benzoxazolyl, benzothiazolyl, benzothiadiazolyl, imidazopyridinyl, isoxazolopyridinyl, thianaphthalenyl, purinyl, xanthinyl, adeninyl, guaninyl, quinolinyl, isoquinolinyl, tetrahydroquinolinyl, quinoxalinyl, and quinazolinyl groups. Heteroaryl groups can be unsubstituted or can be substituted with groups as is discussed herein. Representative substituted heteroaryl groups can be substituted one or more times with groups such as those listed herein. Additional examples of aryl and heteroaryl groups include but are not limited to phenyl, biphenyl, indenyl, naphthyl (1-naphthyl, 2-naphthyl), N-hydroxytetrazolyl, N- hydroxytriazolyl, N-hydroxyimidazolyl, anthracenyl (1-anthracenyl, 2-anthracenyl, 3- anthracenyl), thiophenyl (2-thienyl, 3-thienyl), furyl (2-furyl, 3-furyl) , indolyl, oxadiazolyl, - 21 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) isoxazolyl, quinazolinyl, fluorenyl, xanthenyl, isoindanyl, benzhydryl, acridinyl, thiazolyl, pyrrolyl (2-pyrrolyl), pyrazolyl (3-pyrazolyl), imidazolyl (1-imidazolyl, 2-imidazolyl, 4-imidazolyl, 5-imidazolyl), triazolyl (1,2,3-triazol-1-yl, 1,2,3-triazol-2-yl 1,2,3-triazol-4-yl, 1,2,4-triazol-3-yl), oxazolyl (2-oxazolyl, 4-oxazolyl, 5-oxazolyl), thiazolyl (2-thiazolyl, 4- thiazolyl, 5-thiazolyl), pyridyl (2-pyridyl, 3-pyridyl, 4-pyridyl), pyrimidinyl (2-pyrimidinyl, 4-pyrimidinyl, 5-pyrimidinyl, 6-pyrimidinyl), pyrazinyl, pyridazinyl (3- pyridazinyl, 4- pyridazinyl, 5-pyridazinyl), quinolyl (2-quinolyl, 3-quinolyl, 4-quinolyl, 5-quinolyl, 6- quinolyl, 7-quinolyl, 8-quinolyl), isoquinolyl (1-isoquinolyl, 3-isoquinolyl, 4-isoquinolyl, 5- isoquinolyl, 6-isoquinolyl, 7-isoquinolyl, 8-isoquinolyl), benzo[b]furanyl (2-benzo[b]furanyl, 3-benzo[b]furanyl, 4-benzo[b]furanyl, 5-benzo[b]furanyl, 6-benzo[b]furanyl, 7- benzo[b]furanyl), 2,3-dihydro-benzo[b]furanyl (2-(2,3-dihydro-benzo[b]furanyl), 3-(2,3- dihydro-benzo[b]furanyl), 4-(2,3-dihydro-benzo[b]furanyl), 5-(2,3-dihydro-benzo[b]furanyl), 6-(2,3-dihydro-benzo[b]furanyl), 7-(2,3-dihydro-benzo[b]furanyl), benzo[b]thiophenyl (2- benzo[b]thiophenyl, 3-benzo[b]thiophenyl, 4-benzo[b]thiophenyl, 5-benzo[b]thiophenyl, 6- benzo[b]thiophenyl, 7-benzo[b]thiophenyl), 2,3-dihydro-benzo[b]thiophenyl, (2-(2,3- dihydro-benzo[b]thiophenyl), 3-(2,3-dihydro-benzo[b]thiophenyl), 4-(2,3-dihydro- benzo[b]thiophenyl), 5-(2,3-dihydro-benzo[b]thiophenyl), 6-(2,3-dihydro- benzo[b]thiophenyl), 7-(2,3-dihydro-benzo[b]thiophenyl), indolyl (1-indolyl, 2-indolyl, 3-indolyl, 4-indolyl, 5-indolyl, 6-indolyl, 7-indolyl), indazole (1-indazolyl, 3-indazolyl, 4-indazolyl, 5-indazolyl, 6-indazolyl, 7-indazolyl), benzimidazolyl (1-benzimidazolyl, 2-benzimidazolyl, 4-benzimidazolyl, 5-benzimidazolyl, 6-benzimidazolyl, 7-benzimidazolyl, 8-benzimidazolyl), benzoxazolyl (1-benzoxazolyl, 2-benzoxazolyl), benzothiazolyl (1- benzothiazolyl, 2-benzothiazolyl, 4-benzothiazolyl, 5-benzothiazolyl, 6-benzothiazolyl, 7-benzothiazolyl), carbazolyl (1-carbazolyl, 2-carbazolyl, 3-carbazolyl, 4-carbazolyl), 5H-dibenz[b,f]azepine (5H-dibenz[b,f]azepin-1-yl, 5H-dibenz[b,f]azepine-2-yl, 5H-dibenz[b,f]azepine-3-yl, 5H-dibenz[b,f]azepine-4-yl, 5H-dibenz[b,f]azepine-5-yl), 10,11-dihydro-5H-dibenz[b,f]azepine (10,11-dihydro-5H-dibenz[b,f]azepine-1-yl, 10,11-dihydro-5H-dibenz[b,f]azepine-2-yl, 10,11-dihydro-5H-dibenz[b,f]azepine-3-yl, 10,11-dihydro-5H-dibenz[b,f]azepine-4-yl, 10,11-dihydro-5H-dibenz[b,f]azepine-5-yl), and the like. The term “heterocyclylalkyl” as used herein refers to alkyl groups as defined herein in which a hydrogen or carbon bond of an alkyl group as defined herein is replaced with a bond to a heterocyclyl group as defined herein. Representative heterocyclyl alkyl groups include, but are not limited to, furan-2-yl methyl, furan-3-yl methyl, pyridine-3-yl methyl, - 22 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) tetrahydrofuran-2-yl ethyl, and indol-2-yl propyl. The term “heteroarylalkyl” as used herein refers to alkyl groups as defined herein in which a hydrogen or carbon bond of an alkyl group is replaced with a bond to a heteroaryl group as defined herein. The term “heterocyclylalkyl” as used herein refers to alkyl groups as defined herein in which a hydrogen or carbon bond of an alkyl group as defined herein is replaced with a bond to a heterocyclyl group as defined herein. Representative heterocyclyl alkyl groups include, but are not limited to, furan-2-yl methyl, furan-3-yl methyl, pyridine-3-yl methyl, tetrahydrofuran-2-yl ethyl, and indol-2-yl propyl. The term “heterocyclyl” as used herein refers to aromatic and non-aromatic ring compounds containing three or more ring members, of which one or more is a heteroatom such as, but not limited to, N, O, and S. Thus, a heterocyclyl can be a cycloheteroalkyl, or a heteroaryl, or if polycyclic, any combination thereof. In some embodiments, heterocyclyl groups include 3 to about 20 ring members, whereas other such groups have 3 to about 15 ring members. A heterocyclyl group designated as a C2-heterocyclyl can be a 5-ring with two carbon atoms and three heteroatoms, a 6-ring with two carbon atoms and four heteroatoms and so forth. Likewise, a C4-heterocyclyl can be a 5-ring with one heteroatom, a 6-ring with two heteroatoms, and so forth. The number of carbon atoms plus the number of heteroatoms equals the total number of ring atoms. A heterocyclyl ring can also include one or more double bonds. A heteroaryl ring is an embodiment of a heterocyclyl group. The phrase “heterocyclyl group” includes fused ring species including those that include fused aromatic and non-aromatic groups. For example, a dioxolanyl ring and a benzdioxolanyl ring system (methylenedioxyphenyl ring system) are both heterocyclyl groups within the meaning herein. The phrase also includes polycyclic ring systems containing a heteroatom such as, but not limited to, quinuclidyl. Heterocyclyl groups can be unsubstituted or can be substituted as discussed herein. Heterocyclyl groups include, but are not limited to, pyrrolidinyl, piperidinyl, piperazinyl, morpholinyl, pyrrolyl, pyrazolyl, triazolyl, tetrazolyl, oxazolyl, isoxazolyl, thiazolyl, pyridinyl, thiophenyl, benzothiophenyl, benzofuranyl, dihydrobenzofuranyl, indolyl, dihydroindolyl, azaindolyl, indazolyl, benzimidazolyl, azabenzimidazolyl, benzoxazolyl, benzothiazolyl, benzothiadiazolyl, imidazopyridinyl, isoxazolopyridinyl, thianaphthalenyl, purinyl, xanthinyl, adeninyl, guaninyl, quinolinyl, isoquinolinyl, tetrahydroquinolinyl, quinoxalinyl, and quinazolinyl groups. Representative substituted heterocyclyl groups can be mono-substituted or substituted more than once, such as, but not limited to, piperidinyl or quinolinyl groups, which are 2-, 3-, 4-, 5-, or 6- - 23 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) substituted, or disubstituted with groups such as those listed herein. The term “heterogeneous catalyst” as used herein refers to a composition comprising at least one catalytically active site (e.g., transition metal catalyst complex), wherein the composition exists in a different phase (i.e., physical state) from the reactants in the reaction for which the composition acts as a catalyst. In one non-limiting embodiment, the heterogeneous catalyst is in a solid state, whereas the reactants and / or products are in a liquid or gaseous state. The term “independently selected from” as used herein refers to referenced groups being the same, different, or a mixture thereof, unless the context clearly indicates otherwise. Thus, under this definition, the phrase “X1, X2, and X3are independently selected from noble gases” would include the scenario where, for example, X1, X2, and X3are all the same, where X1, X2, and X3are all different, where X1and X2are the same but X3is different, and other analogous permutations. The term “linker” as used herein refers to a chemically inert region between two linked molecules (e.g., a catalyst complex and a substrate). As used herein, the term “electrode” may refer to the active mass in a power source. For example, it may include either an anode or a cathode, or both an anode and a cathode. The term “cathode” refers to an electrode through which an electric current flows from a polarized electrical device (i.e., reducing electrode). The direction of the electric current is usually opposite to the direction of the electron flow. Therefore, an electron stream enters the cathode of a polarized electrical device and flows, for example, from a connected electrical circuit. The term “anode” refers to an electrode through which an electric current flows into a polarized electrical device (i.e., oxidizing electrode). The direction of the electric current is usually opposite to the direction of the electron flow. In other words, the flow of electrons comes from the anode, for example, into an electrical circuit. The term “L-type ligand” refers to a lone pair (two electron) donating, neutral ligand. Non-limiting examples of “L-type ligands” for use in accordance with the present disclosure include olefins, phosphines (e.g., PPh3), phosphites, carbenes, amines, carbon monoxide (CO), nitrogen (N2), and the like, and combinations thereof. The term “neutral ligand” or “ligand” as used herein, refers to a ligand having no net charge prior to association with, or after dissociation from, a metal center. Non-limiting examples of neutral ligands include alkene (e.g., cyclooctadiene), CO, amine (e.g., NMe3), phosphine (e.g., PPh3), and pyridyl ligands, wherein coordination occurs via the nitrogen lone pair of the pyridyl group. - 24 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) The term “porous silicon” as used herein refers to a substance comprising elemental silicon (e.g., in a crystalline or amorphous state) having a surface containing hydrogen, oxygen or carbon-containing species. In certain embodiments, porous silicon comprises a number of micropores (i.e., pore size is usually less than about 2 nm), mesopores (i.e., pore size is usually about 2-50 nm), or macropores (i.e., pore size is greater than about 50 nm). Furthermore, it should be understood that, upon etching, the surface of the porous material, including the surface of the inner pore wall, may contain a number of Si-H moieties. The term “substituted” as used herein in conjunction with a molecule or an organic group as defined herein refers to the state in which one or more hydrogen atoms contained therein are replaced by one or more non-hydrogen atoms. The term “functional group” or “substituent” as used herein refers to a group that can be or is substituted onto a molecule or onto an organic group. Examples of substituents or functional groups include, but are not limited to, a halogen (e.g., F, Cl, Br, and I); an oxygen atom in groups such as hydroxy groups, alkoxy groups, aryloxy groups, arylalkyloxy groups, oxo(carbonyl) groups, carboxyl groups including carboxylic acids, carboxylates, and carboxylate esters; a sulfur atom in groups such as thiol groups, alkyl and aryl sulfide groups, sulfoxide groups, sulfone groups, sulfonyl groups, and sulfonamide groups; a nitrogen atom in groups such as amines, hydroxyamines, nitriles, nitro groups, N-oxides, hydrazides, azides, and enamines; and other heteroatoms in various other groups. Non-limiting examples of substituents that can be bonded to a substituted carbon (or other) atom include F, Cl, Br, I, OR, OC(O)N(R)2, CN, NO, NO2, ONO2, azido, CF3, OCF3, R, O (oxo), S (thiono), C(O), S(O), methylenedioxy, ethylenedioxy, N(R)2, SR, SOR, SO2R, SO2N(R)2, SO3R, C(O)R, C(O)C(O)R, C(O)CH2C(O)R, C(S)R, C(O)OR, OC(O)R, C(O)N(R)2, OC(O)N(R)2, C(S)N(R)2, (CH2)0-2N(R)C(O)R, (CH2)0-2N(R)N(R)2, N(R)N(R)C(O)R, N(R)N(R)C(O)OR, N(R)N(R)CON(R)2, N(R)SO2R, N(R)SO2N(R)2, N(R)C(O)OR, N(R)C(O)R, N(R)C(S)R, N(R)C(O)N(R)2, N(R)C(S)N(R)2, N(COR)COR, N(OR)R, C(=NH)N(R)2, C(O)N(OR)R, and C(=NOR)R, wherein R can be hydrogen or a carbon-based moiety; for example, R can be hydrogen, (C1- C100) hydrocarbyl, alkyl, acyl, cycloalkyl, aryl, aralkyl, heterocyclyl, heteroaryl, or heteroarylalkyl; or wherein two R groups bonded to a nitrogen atom or to adjacent nitrogen atoms can together with the nitrogen atom or atoms form a heterocyclyl. The term “substrate” as used herein refers to a substance or layer that underlies something, or on which some process occurs. The term “transition metal” as used herein refers to any of the metallic elements within Groups 3 to 12 of the periodic table of elements that have an incomplete inner electron - 25 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) shell and that serve as transitional links between the most and the least electropositive in a series of elements. Transition metals are characterized by multiple valences, colored compounds, and the ability to form stable complex ions. The term “nonmetal” refers to a chemical element that does not have the capacity to lose electrons and form a positive ion. A transition metal element can be any of scandium, titanium, vanadium, chromium, manganese, iron, cobalt, nickel, copper, zinc, yttrium, zirconium, niobium, molybdenum, technetium, ruthenium, rhodium, palladium, silver, cadmium, hafnium, tantalum, tungsten, rhenium, osmium, iridium, platinum, gold, mercury, rutherfordium, dubnium, seaborgium, bohrium, hassium, meitnerium, ununnilium, unununium, and ununbium. In one embodiment, a BMG containing a transition metal element can have at least one of Sc, Y, La, Ac, Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W, Mn, Tc, Re, Fe, Ru, Os, Co, Rh, Ir, Ni, Pd, Pt, Cu, Ag, Au, Zn, Cd, and Hg. Depending on the application, any suitable transitional metal elements, or their combinations, can be used. The alloy composition can comprise multiple transitional metal elements, such as at least two, at least three, at least four, or more, transitional metal elements. The term “X-type ligand” as used herein refers to a single electron donating, ionic ligand. Non-limiting examples of “X-type ligands” for use in accordance with the present disclosure include halides, sulfonates, carboxylates, CN, NO, and hydroxide, and the like (e.g., counter anions), and combinations thereof. Compositions Silicon Substrates In one aspect, the present disclosure relates to catalytically active compositions comprising a solid support-conjugated catalyst, wherein the solid support comprises a light absorbing semiconductor. In certain embodiments, the light absorbing semiconductor comprises silicon (e.g., silicon wafer). The surface area of commercially available, “flat” silicon wafers can be greatly increased (e.g., up to about 106-fold) by the introduction of pores. In certain embodiments, surface etching can be used to prepare porous silicon wafers. Silicon wafers can be etched by one or more methods, including but not limited to methods comprising application of hydrofluoric acid to remove the native oxide and form a hydrogen-terminated silicon surface (i.e., surface Si-H). Porous silicon wafers were prepared according to methods known to those of ordinary skill in the art and characterized by Brunauer-Emmett-Teller (BET) analysis, scanning - 26 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) electron microscopy (SEM), and Fourier transform infrared (FTIR) spectroscopy, to assess surface area, pore morphology, and surface chemistry (i.e., identity of various surface moieties), respectively. In certain embodiments, applied current, HF concentration, time, and silicon type can be varied to generate a variety of pore morphologies. The resultant surface Si-H groups are chemically reactive, and may participate in myriad reactions, including but not limited to H•abstraction. This reactivity of surface Si-H moieties permits quantification of Si-H groups present on porous silicon. Wafers were etched for varying lengths of time and Si-H absorbances were measured by FTIR. Heterogeneous catalyst compositions In one aspect, the present disclosure provides a heterogeneous catalyst composition comprising at least one transition metal catalyst complex covalently attached to a porous silicon substrate. In certain embodiments, the transition metal catalyst complex is a photoelectrocatalyst. In certain embodiments, the transition metal catalyst complex is covalently attached to a Si atom of the porous silicon substrate. In certain embodiments, the porous silicon substrate is an oxidized porous silicon substrate. In certain embodiments, the oxidized porous silicon substrate is a thermally oxidized porous silicon substrate. In certain embodiments, the covalent attachment comprises a covalent bond formed by a reaction between a silatrane moiety of the transition metal catalyst complex and a surface Si-OH moiety of the oxidized porous silicon substrate. In certain embodiments, the porous silicon substrate is a hydrogen-terminated porous silicon substrate. In certain embodiments, the covalent attachment comprises a covalent bond formed by a reaction between a non-aromatic, at least partially unsaturated moiety of the transition metal catalyst complex and a surface Si-H moiety of the hydrogen-terminated porous silicon substrate. In certain embodiments, the transition metal catalyst complex comprises a transition metal center selected from the group consisting of Pd, Cu, Ag, Au, Ni, Pt, Mn, Re, Co, Rh, Ir, Fe, Ru, and Os. In certain embodiments, the covalent attachment occurs via a linker. In certain embodiments, the covalent attachment occurs via a direct bond. In certain embodiments, the - 27 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) linker comprises an optionally substituted C1-C12 alkylenyl moiety. In certain embodiments, the heterogeneous catalyst composition is a composition of formula (I): (I), wherein: Spis a porous silicon substrate; L is selected from the group consisting of a bond, optionally substituted C1-C6 alkylenyl, optionally substituted C1-C6heteroalkylenyl, optionally substituted C2-C6alkenylenyl, optionally substituted C2-C6 heteroalkenylenyl, optionally substituted C3-C8 cycloalkylenyl, optionally substituted C2-C8heterocycloalkylenyl, optionally substituted C6-C10arylenyl, and optionally substituted C2-C10 heteroarylenyl; X1L5L1; independently an L-type ligand, wherein two adjacent L1, L2, L3, L4, or L5groups can combine to form a bidentate L-type ligand; X1is an X-type ligand; and m is an integer is greater than 1; wherein one of L1, L2, L3, L4, L5, and X1, or a combination thereof, is covalently linked to Sp via L. In certain embodiments, L1, L2, L3, L4, and L5are each independently selected from the group consisting of carbon monoxide (CO), P(RA)(RB)(RC), N(RA)(RB)(RC), O(RA)(RB), optionally substituted C2-C12 alkene, and optionally substituted C5-C12 cycloalkene, and optionally substituted C2-C8heteroaryl, wherein RA, RB, and RCare each independently selected from the group consisting of H, optionally substituted C1-C6 alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10 heteroaryl. In certain embodiments, L1and L2combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L1and L3combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L1and L4combine to form a bidentate - 28 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) phosphine ligand or bidentate amine. In certain embodiments, L1and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L2and L3combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L2and L4combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L2and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L3and L4combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L3and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L4and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, the bidentate amine is optionally substituted bipyridine. In certain embodiments, L1is CO. In certain embodiments, L2is CO. In certain embodiments, L3is CO. In certain embodiments, L4is CO. In certain embodiments, L5is CO. In certain embodiments, L1and L2combine to form an optionally substituted bipyridine. In certain embodiments, the heterogeneous catalyst composition is a composition selected from the group consisting of formula (Ia) and (Ib): , wherein: each occurrence of R1aand R1bis independently selected from the group consisting of optionally substituted C1-C6alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8 cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C2-C6alkenyl, optionally substituted C6-C10aryl, optionally substituted C2-C10heteroaryl, halogen, CN, NO2, ORD, N(RD)(RE), C(=O)RD, C(=O)N(RD)(RE), C(=O)ORD, N(RE)S(=O)2RD, and S(=O)2N(RD); RDand REare each independently selected from the group consisting of H, optionally substituted C1-C6alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10heteroaryl; n is selected from the group consisting of 0, 1, 2, and 3; and - 29 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) o is selected from the group consisting of 0, 1, 2, 3, and 4. In certain embodiments, the heterogenous catalyst composition is selected from the group consisting of formula (Ia-1), (Ib-1), (Ia-2), and (Ib-2): , 2), an porous wherein bond L is a covalent bond formed by a reaction between a silatrane moiety of the transition metal catalyst complex and a surface Si-OH moiety of the oxidized porous silicon substrate; and Sphis a hydrogen-terminated porous silicon substrate, wherein bond L is a covalent bond formed by a reaction between a non- aromatic, at least partially unsaturated moiety of the transition metal catalyst complex and a surface Si-H moiety of the hydrogen-terminated porous silicon substrate. In certain embodiments, X1is selected from the group consisting of RF, halogen, OS(=O)2RF, OC(=O)RF, N(C(=O)RF)2, tetracoordinate boronate, hexacoordinate phosphorus, optionally substituted C6-C10 aryl, and optionally substituted C2-C10 heteroaryl, wherein RFis selected from the group consisting of H, optionally substituted C1-C6alkyl, optionally substituted C1-C6 heteroalkyl, optionally substituted C3-C8 cycloalkyl, optionally substituted C2-C8heterocycloalkyl, optionally substituted C6-C10aryl, and optionally substituted C2-C10heteroaryl. In certain embodiments, X1is Cl. In certain embodiments, X1is Br. In certain embodiments, n is 0. In certain embodiments, n is 1. In certain embodiments, o is 0. In certain embodiments, o is 1. - 30 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) In certain embodiments, R1ais CH3. In certain embodiments, R1bis CH3. In certain embodiments, L is -(CH2)4-. In certain embodiments, the porous silicon substrate is a wafer. In certain embodiments, the porous silicon substrate is a bead. In certain embodiments, the porous silicon substrate is a nanotube. In certain embodiments, the porous silicon substrate is a nanorod. In certain embodiments, the silicon comprises p-type silicon. In certain embodiments, the silicon comprises n-type silicon. In certain embodiments, the porous silicon substrate has a pore size ranging from about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 105, 110, 115, or about 120 nm. In certain embodiments, the porous silicon substrate has a pore size of about 70 nm. In certain embodiments, the porous silicon substrate has a surface area of about 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, , or about 400 m2 / g. In certain embodiments, the porous silicon substrate has a surface area of about 200 m2 / g. The porous silicon compositions prepared according to the methods described herein, or otherwise prepared according to methods known to those of ordinary skill in the art, comprise a quantifiable amount of Si-H groups. Further, as similarly described elsewhere herein, Si-H groups may engage in a number of reactions, including hydrogen abstraction and hydrosilylation. Hydrosilylation comprises the addition of an Si-H bond across an unsaturated bond (e.g., alkene or alkyne) to afford organosilicon compounds. Hydrosilylation can be initiated, for example, with UV light at room temperature, by sonication, or with heat (e.g., 120 °C to 200 °C or more), under moisture- and oxygen-free conditions. In certain embodiments, a hydrolysis / condensation reaction can be employed to covalently attach one or more catalysts to the surface of an oxidized, porous silicon particle. In certain embodiments, the oxidized, porous silicon particle comprises one or more surface siloxy moieties (i.e., -SiOH). In certain embodiments, the covalent attachment can be a direct consequence of condensation of a surface siloxy (SiOH) moiety with a silatrane moiety substituent of the catalyst. In one non-limiting embodiment, a silatrane-tethered manganese (Mn) catalyst (i.e.,silatranebpy)Mn(CO)3Br was covalently attached to a surface oxidized, porous silicon composition by a hydrolysis / condensation reaction, as depicted in Scheme 1. - 31 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) In certain embodiments, hydrosilylation can be employed to covalently attach one or more catalysts to the surface of an etched, porous silicon wafer. In certain embodiments, the etched, porous silicon wafer comprises a surface which is at least partially functionalized with hydrogen-terminated silicon moieties (i.e., Si-H). In certain embodiments, the covalent attachment can be a direct consequence of the hydrosilylation reaction (e.g., by hydrosilylation of an alkene-tethered catalyst) or the covalent attachment may be indirect (e.g., hydrosilylation of a bifunctional linker comprising an alkene moiety and another orthogonally reactive moiety suitable for covalent attachment to a catalyst in a subsequent transformation). In one non-limiting embodiment, an alkene-tethered rhenium (Re) catalyst (i.e.,alkenebpy)Re(CO)3Cl (1) was covalently attached to a porous silicon composition by hydrosilylation, as depicted in Scheme 2. In another aspect, the present disclosure provides an electrode comprising the heterogeneous catalyst composition of the present disclosure. In certain embodiments, the electrode is a cathode. Methods In another aspect, the present disclosure provides a method for preparing a heterogeneous catalyst composition comprising at least one transition metal catalyst complex - 32 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) covalently attached to a porous silicon substrate. In certain embodiments, the method comprises (a) contacting an etched, hydrogen-terminated porous silicon substrate and a transition metal catalyst complex to provide a mixture. In certain embodiments, the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one non-aromatic, at least partially unsaturated moiety. In certain embodiments, the method comprises (b) heating the mixture to a temperature of about 160 °C. In certain embodiments, the transition metal catalyst has a concentration of about 0.4 mM during the contacting in (a). In certain embodiments, each contacting occurs in a solvent, optionally wherein the solvent is mesitylene. In certain embodiments, the contacting in (a) occurs for a period of about 18 h. In certain embodiments, the contacting in (b) occurs for a period of about 4 hours. In certain embodiments, each contacting occurs in a solvent, optionally wherein the solvent is mesitylene. In certain embodiments, the non-aromatic, at least partially unsaturated moiety is selected from the group consisting of optionally substituted C2-C6 alkenyl, optionally substituted C2-C6 alkynyl, optionally substituted C5-C10 cycloalkylenyl, optionally substituted C8-C12 cycloalkynylenyl, optionally substituted C2-C6 heteroalkenyl, optionally substituted C2-C6 heteroalkynyl, optionally substituted C5-C10 heterocycloalkylenyl, optionally substituted C8-C12heterocycloalkynylenyl. In certain embodiments, the transition metal catalyst complex comprises a bipyridyl moiety substituted with at least one -(CH2)0-12CH=CH2moiety. In certain embodiments, the bipyridyl moiety substituted with at least one -(CH2)0-12CH=CH2 moiety is further substituted with at least one substituent. In another aspect, the disclosure provides a method for preparing a heterogeneous catalyst composition comprising at least one transition metal catalyst complex covalently attached to a porous silicon substrate. In certain embodiments, the method comprises (a) contacting an at least partially surface oxidized porous silicon substrate and a transition metal catalyst complex to provide a mixture, wherein the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one silatrane moiety. In certain embodiments, the method comprises (b) heating the mixture to a temperature of about 80 °C. In certain embodiments, the transition metal catalyst has a concentration of about 0.2 mM during the contacting in (a). In certain embodiments, each contacting occurs in a solvent. In certain embodiments, the solvent is acetonitrile. In certain embodiments, the contacting in - 33 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) (a) occurs for a period of about 18 h. In certain embodiments, the transition metal catalyst complex comprises at least one bipyridyl moiety substituted with at least is optionally further substituted with one or more substituents. the bipyridyl moiety substituted with at least is further substituted with one or more substituents. In another aspect, provides a method for reducing CO2. In certain embodiments, the (a) contacting a medium comprising CO2with an electrochemical cell comprising an anode and a cathode. In certain embodiments, the cathode comprises at least one transition metal catalyst complex covalently attached to a porous silicon substrate. In certain embodiments, the method comprises (b) applying an electric potential between the anode and the cathode. In certain embodiments, the method comprises (c) irradiating the medium. In certain embodiments, the irradiation of the medium comprises irradiation with a solar simulator. In certain embodiments, the solar simulator comprises a light emitting diode (LED). In certain embodiments, the irradiation comprises about 1 sun of illuminance (100 mW / cm2). In certain embodiments, the irradiation comprises a wavelength of about 380, 390, 400, 410, 420, 430, 440, 450, 460, 470, 480, 490, 500, 510, 520, 530, 540, 550, 560, 570, 580, 590, 600, 610, 620, 630, 640, 650, 660, 670, 680, 690, 700, 710, 720, 730, 740, 750, 760, 770, 780, 790, 800, 810, 820, 830, 840, 850, 860, 870, 880, 890, 900, 910, 920, 930, 940, 950, 960, 970, 980, 990, or about 1000 nm. In certain embodiments, a potential of about -1.75 V is applied. In certain embodiments, formic acid (HC(=O)OH) is produced. In certain embodiments, formate (HC(=O)O-) is produced. In certain embodiments, carbon monoxide (CO) is produced. In certain embodiments, hydrogen (H2) is produced. In certain embodiments, the anode comprises graphite. In certain embodiments, the medium comprises a liquid. In certain embodiments, the medium further comprises at least one electrolyte. In certain embodiments, the electrolyte is tetrabutylammonium hexafluorophosphate - 34 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) (TBAPF6). In certain embodiments, the heterogeneous catalyst composition comprising at least one transition metal catalyst complex covalently attached to a porous silicon substrate is a composition of formula (I). In certain embodiments, the cathode is a composition of formula (I). In certain embodiments, the composition of formula (I) comprises: (I), wherein: S is a porous silic p on substrate; L is selected from the group consisting of a bond, optionally substituted C1-C6alkylenyl, optionally substituted C1-C6 heteroalkylenyl, optionally substituted C2-C6 alkenylenyl, optionally substituted C2-C6heteroalkenylenyl, optionally substituted C3-C8cycloalkylenyl, optionally substituted C2-C8 heterocycloalkylenyl, optionally substituted C6-C10 arylenyl, and optionally substituted C2-C10heteroarylenyl; X1L5L1; independently an L-type ligand, wherein two adjacent L1, L2, L3, L4, or L5groups can combine to form a bidentate L-type ligand; X1is an X-type ligand; and m is an integer is greater than 1; wherein one of L1, L2, L3, L4, L5, and X1, or a combination thereof, is covalently linked to Sp via L. In certain embodiments, L1, L2, L3, L4, and L5are each independently selected from the group consisting of carbon monoxide (CO), P(RA)(RB)(RC), N(RA)(RB)(RC), O(RA)(RB), optionally substituted C2-C12alkene, and optionally substituted C5-C12cycloalkene, and optionally substituted C2-C8 heteroaryl, wherein RA, RB, and RCare each independently selected from the group consisting of H, optionally substituted C1-C6alkyl, optionally substituted C1-C6 heteroalkyl, optionally substituted C3-C8 cycloalkyl, optionally substituted C2-C8heterocycloalkyl, optionally substituted C6-C10aryl, and optionally substituted C2-C10heteroaryl. - 35 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) In certain embodiments, L1and L2combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L1and L3combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L1and L4combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L1and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L2and L3combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L2and L4combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L2and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L3and L4combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L3and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, L4and L5combine to form a bidentate phosphine ligand or bidentate amine. In certain embodiments, the bidentate amine is optionally substituted bipyridine. In certain embodiments, L1is CO. In certain embodiments, L2is CO. In certain embodiments, L3is CO. In certain embodiments, L4is CO. In certain embodiments, L5is CO. In certain embodiments, L1and L2combine to form an optionally substituted bipyridine. In certain embodiments, the heterogeneous catalyst composition is selected from the group consisting of formula (Ia) and (Ib): (R1a) 1n, each occurrence of R1aand R1bis independently selected from the group consisting of optionally substituted C1-C6 alkyl, optionally substituted C1-C6 heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8heterocycloalkyl, optionally substituted C2-C6 alkenyl, optionally substituted C6-C10 aryl, optionally substituted C2-C10 heteroaryl, halogen, CN, NO2, ORD, N(RD)(RE), C(=O)RD, C(=O)N(RD)(RE), C(=O)ORD, N(RE)S(=O)2RD, and S(=O)2N(RD); RDand REare each independently selected from the group consisting of H, optionally substituted C1-C6 alkyl, optionally substituted C1-C6 heteroalkyl, optionally substituted C3-C8 - 36 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10heteroaryl; n is selected from the group consisting of 0, 1, 2, and 3; and o is selected from the group consisting of 0, 1, 2, 3, and 4. In certain embodiments, the heterogenous catalyst composition is selected from the group consisting of formula (Ia-1), (Ib-1), (Ia-2), and (Ib-2): , 2), Spois an oxidized porous silicon substrate, wherein bond L is a covalent bond formed by a reaction between a silatrane moiety of the transition metal catalyst complex and a surface Si-OH moiety of the oxidized porous silicon substrate; and Sphis a hydrogen-terminated porous silicon substrate, wherein bond L is a covalent bond formed by a reaction between a non- aromatic, at least partially unsaturated moiety of the transition metal catalyst complex and a surface Si-H moiety of the hydrogen-terminated porous silicon substrate. In certain embodiments, X1is selected from the group consisting of RF, halogen, OS(=O)2RF, OC(=O)RF, N(C(=O)RF)2, tetracoordinate boronate, hexacoordinate phosphorus, optionally substituted C6-C10aryl, and optionally substituted C2-C10heteroaryl, wherein RFis selected from the group consisting of H, optionally substituted C1-C6 alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10 heteroaryl. - 37 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) In certain embodiments, X1is Cl. In certain embodiments, X1is Br. In certain embodiments, n is 0. In certain embodiments, n is 1. In certain embodiments, o is 0. In certain embodiments, o is 1. In certain embodiments, R1ais CH3. In certain embodiments, R1bis CH3. In certain embodiments, L is -(CH2)4-. In certain embodiments, the porous silicon substrate is a wafer. In certain embodiments, the porous silicon substrate is a nanoparticle. In certain embodiments, the porous silicon substrate is a nanotube. In certain embodiments, the porous silicon substrate is a nanorod. In certain embodiments, the silicon comprises p-type silicon. In certain embodiments, the silicon comprises n-type silicon. In certain embodiments, the porous silicon substrate has a pore size ranging from about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 105, 110, 115, or about 120 nm. In certain embodiments, the porous silicon substrate has a pore size of about 70 nm. In certain embodiments, the porous silicon substrate has a surface area of about 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, or about 400 m2 / g. In certain embodiments, the porous silicon substrate has a surface area of about 200 m2 / g. EXAMPLES Various embodiments of the present application can be better understood by reference to the following Examples which are offered by way of illustration. The scope of the present application is not limited to the Examples given herein. Materials and Methods All commercial chemicals were used as received, except where noted. The MeCN was dispensed from a solvent system plumbed into the glovebox and immediately used or stored over 4Å molecular sieves. Mesitylene was purchased from Sigma-Aldrich, degassed by purging with dry N2for 20 min and then stored in a glovebox over 4 Å molecular sieves. Tetrabutylammonium hexafluorophosphate (TBAPF6, Sigma-Aldrich) was recrystallized twice from EtOH and dried at 70 °C under vacuum for 2-3 days.95% CO2and 5% methane gas mixture were purchased from Airgas Inc. Nafion 212 membrane was purchase from Ion Power and used as is.49% HF was purchased from Sigma-Aldrich. Si wafers were obtained from University Wafer (P-type, B doped, <100>, SSP Prime Grade) with an estimated - 38 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) resistance of 1-10 ^•cm-1. CPE headspace analyses were performed with a Shimadzu Scientific Instruments Nexis 8030 GC system equipped with a Shimadzu Q-BOND column (Length 30 m, 0.53 mm ID, 20 µm film thickness), a Shimadzu Msieve 5A column (Length 30 m, 0.53 mm ID, 50 µm film thickness) Jetanizer™ FID detector and TCD detector. Mass spectrometric measurements were performed with a Shimadzu Scientific Instruments Triple- Quadrupole 8050 GC-MS NX system, equipped with a AOC-6000 autosampler, a GC-2030 column oven, an Agilent PoraPlotQ column (Length 25 m, 0.25 mm ID, 8 µm film thickness), and a GCMS-8050 mass spectrometer.4,4'-dimethyl-2,2'-bipyridine rhenium tricarbonyl chloride (Me2bpy)Re(CO)3Cl, 4-(4-butenyl)-4'-methyl-2,2'-bipyridine rhenium tricarbonyl chloride (1), 4-(4-butylsilatrane)-4'-methyl-2,2'-bipyridine, 4,4'-bis(4- butylsilatrane)-2,2'-bipyridine, 4-(3-butenyl)-4'-methyl-2,2'-bipyridine, 4,4-bis(3-butenyl)- 2,2'-bipyridine, andnBu4N formate (TBA formate) were synthesized according to literature procedures. Preparation of H-terminated porous silicon (H-porSi) Planar Si wafers (either P-type 1-10 Ω•cm-1 or N-type 0.5 Ω•cm-1) were cut into approximately 1 cm by 1.5 cm rectangles, and then cleaned copiously with acetone and ethanol according to methods known to those of ordinary skill in the art. In a typical etching procedure, a planar wafer was assembled in a custom Teflon etching cell equipped with a Kalrez® O-ring and PTFE screws. Similar etching cells are commercially available, and detailed design drawings have previously been described in the art. The cell was filled with ethanol in order to check for any leaks before etching. The ethanol was removed, and the cell set up within a fume hood. A spiral-shaped Pt counter electrode was lowered into the top opening of the etching cell and adjusted until parallel with the Si wafer below 4 mL of a 1:1 solution of 49% HF and 200 proof EtOH was added to the cell, and a Keithley 2401 SourceMeter was used to apply 40 mA of current for 5 minutes. After etching, the HF solution was removed and the open area of the cell was rinsed 3 times with ethanol. The whole front of the cell was then rinsed copiously with ethanol and dried under N2. H-porSi samples were either used immediately after etching or stored in a glove box. Preparation of 1 immobilized on porous silicon (1@porSi) In a glovebox filled with N2, a freshly etched H-porSi wafer was submerged in a 5 mL mesitylene solution containing 0.4 mM of 1 in a Schlenk flask. The flask was brought out of - 39 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) the glovebox, connected to a Schlenk line, covered with aluminum foil, and the solution was heated under N2at 160 °C for 18 hours. After allowing the solution to cool, the wafer was rinsed with n-pentane and acetonitrile, soaked in ethanol under N2 flow for an hour, and then rinsed with pentane. Any trace amounts of solvent absorbed in 1@porSi were removed under vacuum for 30 minutes prior to characterization. Preparation of heated H-terminated porous silicon (H-porSi-heated) In a glovebox filled with N2, a freshly etched H-porSi wafer was submerged in a 5 mL mesitylene in a Schlenk flask. The flask was brought out of the glovebox, connected to a Schlenk line, covered with aluminum foil, and the wafer was heated in mesitylene under N2at 160 °C for 18 hours. After allowing the flask to cool, the wafer was rinsed with n-pentane and acetonitrile, soaked in ethanol under N2flow for an hour, and then rinsed with pentane. Any trace amounts of solvent absorbed in H-porSi-heated were removed under vacuum for 30 minutes prior to characterization. IR spectroscopy of exemplary compositions Transmittance FTIR spectra were recorded using a Bruker Vertex 80 spectrometer equipped with a liquid nitrogen-cooled MCT detector. A Pike standard solid sample holder was used to support the porous wafer. The position of the wafer was carefully adjusted each time so that the porous area was centered with the beam, and a wide aperture of 8 mm was used. Air was used as the sample background, and 25 scans were taken for both the background and sample. X-ray photoelectron spectroscopy (XPS) X-ray photoelectron spectroscopy (XPS) data was collected on a Kratos Axis Supra+ system with a monochromatic Al Ka X-ray source operated at 150W. A charge neutralizer was used to prevent charging when necessary and all spectra were corrected to the C 1s peak at 284.6 eV. Survey and high-resolution scans were acquired at pass energies of 80 eV and 20 eV, respectively, and the analyzed spot size was 300 x 700 um. Sample preparation The XPS data of the porous Si samples, H-porSi and 1@porSi were prepared using two different methods: (1) XPS on intact sample: 1@porSi was analyzed by XPS after functionalization without any treatment; and (2) XPS on powdered sample: The powdered - 40 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) sample was prepared by scratching the porous Si surface by spatula until the gray Si surface below the porous Si was exposed. BET surface area measurement of H-porSi and 1@porSi A standard multipoint BET program on a micromeritics 3Flex instrument was used for surface area measurements. Wafers were cut in half and massed carefully within the BET sample tube. Using a Micromererics VacPrep, each sample tube was heated to 150 °C under vacuum overnight before being transferred to the 3Flex instrument for analysis. Sample tubes were equipped with isothermal jackets to ensure a constant temperature in the N2 bath. Ultrahigh purity N2was used as the adsorbent, samples were run in triplicate, and at least 5 points between p / p00.05-0.3 were used in the BET analysis. Cross-sectional SEM and EDS of 1@porSi The SEM and EDS were performed on Hitachi SU8230 CFE SEM equipped with a Bruker XFlash 5060FQ annual EDS detector. The cross-section samples were attached to a vertical sample holder using carbon tape and the side of the cross-section area was grounded using copper tapes. A low electron beam voltage of 5 kV was chosen to excite the Re Mα EDS emission line at 1.842 keV only, and to improve the mapping spatial resolution. Due to the EDS spectral resolution limit that led to the overlapping of Re Mα and Si Kα lines, a high beam voltage of 15 kV was used to excite the Re Lα line at 8.645 keV. Inductively Coupled Plasma Mass Spectrometry (ICP-MS) Porous Si-Re samples were digested in 1 mL of 70% aqueous nitric acid (trace metal free grade) for 24 hours to solubilize the surface Re ions. The digested samples were diluted with MilliQ water to a 2% nitric acid concentration. The diluted samples were then mixed with 0.5% HCl / 2% HNO3 aqueous matrix in a 1:2 ratio. These samples were analyzed by an Agilent 7900 inductively coupled plasma mass spectrometer to measure the187Re ion concentration in parts per billion. A calibration curve for187Re concentration ranging from 0.01 to 100 ppb was prepared from commercially available ICP-MS standards (Spex Certiprep). Sample187Re concentrations in ppb were converted first to mol / L, then to number of moles (ncat) by multiplying by the volume of the stock solution (0.01 L) and the dilution factor (35 x 3). Preparation of Si wafers for electrochemistry - 41 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) An electrical contact was made to porous Si wafers samples using the following procedure. The front surface at edge of the porous Si wafer was scratched thoroughly with a diamond tip etching pen to remove SiO2. A small amount of gallium-indium eutectic (MilliporeSigma) was applied to the scratched surface. An insulated wire was stripped to expose about 1 cm of wire at each end. The wire at one end was bent into a spiral shape. One end of the spiral was pressed to the scratched surface on the back of the wafer. The connection was secured with two small piece of Cu tape from front and back. Preparation of 1 immobilized on n-type porous silicon (1@porSi-ntype) In a glovebox filled with N2, a freshly etched n-type H-porSi wafer was submerged in a 5 mL mesitylene solution containing 2 mM of 1 in a Schlenk flask. The flask was brought out of the glovebox, connected to a Schlenk line, covered with aluminum foil, and the solution was heated under N2 at 160 °C for 18 hours. After allowing the solution to cool, the wafer was rinsed with n-pentane and acetonitrile, soaked in ethanol under N2flow for an hour, and then rinsed with pentane. Any trace amounts of solvent absorbed in 1@porSi-ntype were removed under vacuum for 30 minutes prior to characterization. Photoelectrochemistry of 1@porSi CVs with hybrid electrodes 1@porSi were recorded in a two-compartment H-cell. The working and counter compartments were separated by a Nafion 212 membrane. Both sides of the cell were fitted with a Teflon cap that had holes with o-ring and / or septum, which could be sealed around the electrodes and sparging tubing. Both the working and counter compartment consisted of 250 mM TBAPF6, and 100 mM phenol. The working electrode was 1@porSi and the counter electrode was a graphite rod. The reference electrode was an Ag wire. Both compartments were sparged by either Ar or CO2(with 5% of methane as GC internal standard) for 20 minutes, ferrocene was added at the end of the experiment as an internal standard and a CV was recorded with a glassy carbon electrode to minimize potential drift. The CVs were recorded under 1 sun illumination using a VeraSol-2 LED solar simulator. Electrochemistry of 1@porSi-ntype CVs with hybrid electrodes of 1@porSi-ntype were conducted in the same two- compartment H-cell as experiments with 1@porSi. The working electrode was 1@porSi- ntype and the counter electrode was a graphite rod. The reference electrode was an Ag wire. - 42 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Both compartments were sparged by CO2 (with 5% of methane as GC internal standard) for 20 minutes before electrolysis, Ferrocene was added at the end of the experiment as an internal standard and a CV was recorded with a glassy carbon electrode to minimize potential drift. CPE with hybrid electrodes 1@porSi, H-porSi and H-porSi-heated under CO2CPE experiments were conducted in the same two-compartment H-cell as the CV experiments. The working compartment consisted of 250 mMnBu4NPF6and 100 mM phenol. The counter compartment consisted of 100 mM TBAPF6 and 100 mM phenol. The working electrode was the hybrid Si electrode 1@porSi, H-porSi, or H-porSi-heated. The counter electrode was a graphite rod. The reference electrode was an Ag wire. Ferrocene was added at the end of the experiment as an internal reference. Both compartments were sparged with CO2 (with 5% of methane as GC internal standard) for 20 minutes before electrolysis, while stirring the working compartment. The cell was then sealed before electrolysis began. A VeraSol-2 LED solar simulator was used to illuminate the samples and the headspace was regularly sampled using a Shimadzu Nexis GC-2030 equipped with a thermal conductivity detector (TCD) for quantifying O2, N2, and H2 gases, connected in series with a mechanizer and flame ionization detector (FID) for quantifying CO. CPE with H-porSi-heated under Ar Heterogeneous electrolysis experiments were conducted in the same two- compartment H-cell as the CV experiments. The working compartment consisted of 250 mM TBAPF6and 100 mM phenol. The counter compartment consisted of 100 mM TBAPF6and 100 mM phenol. The working electrode was the hybrid Si electrode H-porSi. The counter electrode was a graphite rod. The reference electrode was an Ag wire. Ferrocene was added at the end of the experiment as an internal reference. Both compartments were sparged with Ar 20 minutes, 100 mL of CO was added to the working compartment as a GC internal standard before electrolysis, while stirring the working compartment. The cell was then sealed before electrolysis began. A VeraSol-2 LED solar simulator was used to illuminate the samples and the headspace was regularly sampled using a Shimadzu Nexis GC-2030 equipped with a thermal conductivity detector (TCD) for quantifying O2, N2, and H2gases, connected in series with a mechanizer and flame ionization detector (FID) for quantifying CO. - 43 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Drop casting experiment with complex 1 on H-porSi Complex 1 (1.4 mg, 2.0 mM) was dissolved in 1 mL MeCN in a 4-dram vial.100 µL of a 2.0 mM solution in MeCN were then drop-casted onto the porous area of H-porSi. The resulting porous Si wafer was dried under air at room temperature for 1 hours and then characterized by FTIR spectroscopy. The wafer was then rinsed with 20 mL MeCN three times. There was no visual change in the wafer appearance. The resulting washed porous Si wafer was characterized by FTIR spectroscopy. Immobilization of a control complex (Me2bpy)Re(CO)3Cl on H-porSi In a glovebox filled with N2, a freshly etched H-porSi wafer was submerged in a 5 mL mesitylene solution containing 0.4 mM of (Me2bpy)Re(CO)3Cl in a Schlenk flask. The flask was brought out of the glovebox, connected to a Schlenk line, covered with aluminum foil, and the solution was heated under N2 at 160 °C for 18 hours. After allowing the solution to cool, the wafer was rinsed with n-pentane and acetonitrile, soaked in ethanol under N2flow for an hour, and then rinsed with pentane. CPE with 1@porSi using phenol-d6 as the proton source CPE experiments were conducted in the same two-compartment H-cell as the CV experiments. The working compartment consisted of 250 mM TBAPF6and 100 mM phenol- d6. The counter compartment consisted of 100 mM TBAPF6 and 100 mM phenol. The working electrode was the hybrid Si electrode 1@porSi. The counter electrode was a graphite rod. The reference electrode was an Ag wire. Ferrocene was added at the end of the experiment as an internal reference. Both compartments were sparged with CO2(with 5% of methane as GC internal standard) for 20 minutes before electrolysis, while stirring the working compartment. The cell was then sealed before electrolysis began, and the headspace was regularly sampled using a quadrupole mass spectrometer connected with a Micromeretics 3Flex used for gas flow control. CPE with 1@porSi under13CO2CPE experiments were conducted in the same two-compartment H-cell as the CV experiments. The working compartment consisted of 250 mM TBAPF6and 100 mM phenol. The counter compartment consisted of 100 mM TBAPF6 and 100 mM phenol. The working electrode was the 1@porSi. The counter electrode was a graphite rod. The reference electrode was an Ag wire. Ferrocene was added at the end of the experiment as an internal reference. - 44 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Both compartments were sparged with13CO2 for 15 minutes before electrolysis, while stirring the working compartment. The cell was then sealed before electrolysis began. Headspace sampling was performed with the autosampler tool of the AOC-6000. Samples were equilibrated for 1 min at 40 °C, the syringe was held at 60 °C.100 mL of sample volume was injected with a split injection (split ratio 50) into the injection port of the GC which was held at 250 °C. Helium was used as a carrier gas with a column flow rate of 0.5 mL / min. The sample was separated on an Agilent PoraPlotQ column (Length 25 m, 0.25 mm ID, 8 µm film thickness) with the following temperature program: 40 °C (hold 2 min) to 100 °C with a heat rate of 20 °C / min (hold 2 min). The MS interface and ion source were held at 250 °C, sample ionization was performed with electron ionization (70 eV) and positive ions were analyzed with a Q3 scan between 10 and 400 m / z (event time 100 ms) for the full time of the GC run (7 min). MRM scans were performed for the following molecules and transitions (collision energy 35V, 100ms per event):13CO (29m / z -> 13 and 16m / z),13CO2 (45m / z -> 13 and 29m / z), and12CO (28m / z -> 12 and 16m / z). Measurements and data post-processing were performed with GCMS Real Time Analysis and Postrun Analysis 4.50 SP1. Characterization of molecular Mn(I) complexes NMR spectra were recorded on Agilent DD2-400 or -500 MHz spectrometers at ambient temperature. Chemical shifts for1H and13C NMR spectra are reported in ppm and referenced to residual internal protio solvent. IR spectra of molecular compounds were collected on a Bruker Alpha II spectrometer as compressed solids using an ATR attachment. Mass spectrometric measurements of molecular compounds were recorded with Shimadzu Scientific Instruments QToF 9030 LC-MS system. UV-Vis absorption spectroscopy was performed on a Shimadzu UV-3600Plus under nitrogen in 1 cm quartz cells. Cyclic voltammograms of molecular complexes Homogeneous cyclic voltammetry measurements with 1b-5 in solution were collected on a CH Instruments model 600D / 650D potentiostat with a 3mm glassy carbon working electrode, a platinum wire counter electrode, and a silver wire pseudoreference electrode. All homogeneous cyclic voltammograms were recorded in 100 mMnBu4NPF6 in MeCN. The MeCN was dispensed from a solvent system and immediately used.1.0 mM [Mn] was used in all cases. Glassy carbon electrodes were polished between scans on a felt pad using a 0.05 μm alumina slurry. Ferrocene (Fc) and / or [bis(pentamethylcyclopentadienyl)iron(II)] (Fc*) were added at the end of the experiment as an internal standard in all cyclic voltammetry - 45 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) measurements to minimize potential drift. Fc was added in MeCN containing 0.25 MnBu4NPF6with 100 mM phenol. On the other hand, Fc* was added in MeCN containing 0.25 MnBu4NPF6 with 2.0 M NEt3 and 2.0 MiPrOH because the Fc+ / Fc couple is no longer observable in the presence of NEt3(see section S.IV). All cyclic voltammograms were recorded at a scan rate of 100 mV / s. CO2 reduction electrolysis using molecular complexes CPE experiments with molecular complexes were conducted using a three-electrode system in a two-compartment H-cell, with the compartments separated by an ultra-fine frit (average porosity: 0.9-1.4 micrometers). The working electrode was glassy carbon rod electrode, the counter electrode was a platinum wire electrode and the reference electrode was a silver wire pseudoreference electrode. The electrodes and gas-sparging lines were fitted through gas-tight ports on both sides of the cell. The working compartment contained 1.0 mM [Mn] and 250 mMnBu4NPF6, with either 100 mM phenol or 2.0 M NEt3and 2.0 MiPrOH. The counter compartment solution was of identical composition but did not contain [Mn] catalyst. Both compartments were sparged for 20 minutes before electrolysis with CO2 containing 5% CH4 which was used as internal standard for quantification of gaseous products via gas chromatography. The cell was sealed before the CPE began, and the headspace was regularly sampled using a Shimadzu Nexis GC-2030 gas chromatograph equipped with a thermal conductivity detector (TCD) for quantifying H2 gas, connected in series with a mechanizer and flame ionization detector (FID) for quantifying CO. Formate was quantified using ion chromatography (IC) by taking 500 µL aliquots of the working solution. A 32-fold dilution of these aliquots using distilled water was performed followed by filtration through a 0.22 µm syringe filter prior to IC analysis. The IC instrument was a 930 Compact IC Flex from Metrohm equipped with a Metrosep C4 – 150 / 4.0 column for anion separation and a conductivity detector. Preparation of 1b and 2 immobilized on hydride terminated porous Si (1b@porSi, 2@porSi) In a glovebox filled with N2, a freshly etched H-porSi wafer was submerged in a 5 mL mesitylene solution containing 0.4 mM of 1b or 2 in a thick-wall pressure flask. The flask was sealed and brought out of the glovebox, covered with aluminum foil, and the solution was heated under N2 at 160 °C for 18 hours. After allowing the solution to cool, the wafer was rinsed with pentane and MeCN, soaked in 10 mL ethanol under N2flow for an hour, and then rinsed with pentane. Any trace amounts of solvent absorbed in 1b@porSi and 2@porSi - 46 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) were removed under vacuum (less than 30 mTorr) for 30 minutes prior to characterization. Preparation of SiOx-porSi and SiOx-porSi-ntype A freshly etched hydrogen-terminated porous silicon (H-porSi or H-porSi-ntype) wafer was positioned with the porous side face-up in a crystallization dish and placed in an oven set at 190 °C. The H-porSi and H-porSi-ntype samples were heated for 7 hours and 12 hours, respectively, to facilitate the formation of a silicon oxide (SiOx) layer. The growth of this oxide layer in SiOx-porSi and SiOx-porSi-ntype was monitored periodically using transmittance FTIR spectroscopy as shown in Figure 2b of the text. Preparation of 3@SiOx-porSi, and 4@SiOx-porSi A SiOx-porSi wafer was submerged in a 5 mL MeCN solution containing 2.0 mM of 3 or 4 in a round bottom flask. The flask was sealed, covered with aluminum foil, and the solution was heated under N2at 80 °C for 18 hours. After allowing the solution to cool, the wafer was rinsed with MeCN, soaked in 10 mL ethanol and then rinsed with pentane. Any trace amounts of solvent absorbed in 3@SiOx-porSi, 4@SiOx-porSi were removed under vacuum (less than 30 mTorr) for 30 minutes prior to characterization. Preparation of SiOx-porSi^A SiOx-porSi wafer was submerged in 5 mL MeCN in a round bottom flask. The solution was heated under N2 at 70 °C for 18 hours. After allowing the solution to cool, the wafer was soaked in 10 mL ethanol and then rinsed with pentane. Any trace amounts of solvent absorbed in SiOx-porSi^were removed under vacuum (less than 30 mTorr) for 30 minutes prior to characterization. IR spectroscopy of porous Si based materials Transmittance FTIR spectra were recorded using a Bruker Vertex 80 spectrometer equipped with a liquid nitrogen-cooled MCT detector. A Pike standard solid sample holder was used to support the porous wafer. The position of the wafer was carefully adjusted each time so that the porous area was centered with the beam, and a wide aperture of 8 mm was used. Air was used as the sample background, and 25 scans were taken for both the background and sample. The sample chamber was constantly purged with dry N2. - 47 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) X-ray photoelectron spectroscopy (XPS) XPS data was collected on a Kratos Axis Supra+ system with a monochromatic Al K^X-ray source operated at 150 W. A charge neutralizer was used to prevent charging when necessary and all spectra were corrected to the C 1s peak at 284.6 eV. Survey and high- resolution scans were acquired at pass energies of 80 eV and 20 eV, respectively, and the analyzed spot size was 300 x 700 ^m. Sample preparation: The oxidized porous Si samples for XPS analysis, SiOx-porSi, 3@SiOx-porSi, 4@SiOx-porSi, 3@SiOx-porSi after CPE, and 4@SiOx-porSi after CPE, were prepared using two different methods: (1) XPS on intact sample: The oxidized porous Si sample was analyzed by XPS without any treatment; and (2) XPS on powdered sample: The powdered sample was prepared by scratching the oxidized porous Si surface with a spatula until the gray Si surface below the oxidized porous Si was exposed. The powdered sample was then pressed into a small piece of indium metal, which served to immobilize the powder. BET surface area measurement of SiOx-porSi A standard multipoint BET program on a micromeritics 3Flex instrument was used for surface area measurements. The instrument was calibrated by running a standard provided by the manufacturer in tandem with the porous samples. Wafers were cut in half and massed carefully within the BET sample tube. Using a Micromeritics VacPrep, each sample tube was heated to 150 °C under vacuum overnight before being transferred to the 3Flex instrument for analysis. Sample tubes were equipped with isothermal jackets to ensure a constant temperature in the N2bath. Ultrahigh purity N2was used as the adsorbent, samples were run in triplicate, and at least 5 points between p / p00.05-0.3 were used in the BET analysis. An example BET plot is provided below in Figure S8. The experimentally determined BET surface areas allow for the comparison of porous samples presented in this paper to porous samples from the literature.5Cross-sectional SEM and EDS of SiOx-porSi, 3@SiOx-porSi, 4@SiOx-porSi, 3@SiOx-porSi after CPE and 4@SiOx-porSi after CPE The SEM and EDS were performed on a Hitachi SU8230 CFE SEM equipped with a Bruker XFlash 5060FQ annual EDS detector. The cross-section samples were attached to a vertical sample holder and the side of the cross-section area was grounded using copper tape. A low electron beam voltage of 5 kV was chosen to image the porous area, and a beam - 48 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) voltage of 10 kV (chosen to excite the Mn Mα EDS emission line at 5.898 keV) was used on the same region for EDS mapping. Aberration-corrected scanning transmission electron microscopy (STEM) of 3@SiOx-porSi Sample preparation: The TEM specimen was prepared using a TESCAN plasma (Xe) FIB lift-out technique. To protect the surface layer, a thin coat of nitrocellulose (dissolved in MeCN) was first applied to the sample surface, followed by ion beam-induced deposition of a protective Pt layer. The plasma FIB (Focused Ion beam milling) was used to mill trenches and undercut the lamella, which was then lifted out, transferred to a TEM grid, and thinned to electron transparency using low-energy Xe+ ions, resulting in a high-quality specimen for microstructural analysis. For STEM observation a JEOL NEOARM S / TEM was used for the HAADF-STEM imaging at 200 kV. A 6C probe size and a 40 μm condenser aperture was used, yielding approximately 62.5 pA probe current. For EELS A 4 cm camera length was used for imaging and 2 cm for EELS. The convergence angle was 25 mrad. Gatan DigitalMicrograph software was used to capture and analyze spectroscopic data. Inductively coupled plasma mass spectrometry (ICP-MS) sample preparation for hybrid photoelectrodes Sample preparation for hybrid photoelectrodes: Samples of 3@SiOx-porSi or 4@SiOx-porSi, before or after CPE experiment were placed into a glass beaker containing 10 mL of a 1 M KOH aqueous solution. The wafer was digested for 15 minutes at 60 °C. Subsequently, the wafer was carefully removed from the beaker using tweezers, rinsed with 5.0 mL of Milli-Q water, and returned to the digestion beaker. Following this, the water was evaporated using a heating plate. The residual solid was subjected to digestion in 1.0 mL of 70% aqueous nitric acid (trace metal free grade) for 24 hours to facilitate the solubilization of surface-bound Mn ions. Finally, the digested samples were diluted with Milli-Q water to achieve a final nitric acid concentration of 2%. ICP-MS sample preparation of solution after CPE using 3@SiOx-porSi: Samples of post CPE solution were concentrated under reduced pressure and then the residual solid was transferred to a porcelain bowl. The mixture was at 800 °C in air for 3 hours to remove organic material. The orange-brown residue was sonicated in 3 mL of 10% aqueous nitric acid (trace metal free grade) to solubilize the Mn ions. The digested samples were combined with 2 mL of a 0.5% HCl / 2% HNO3aqueous matrix. Samples were analyzed using an Nexion 5000, multi-quadrupole inductively coupled - 49 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) plasma mass spectrometer to measure the55Mn ion concentration in parts per billion. A calibration curve for55Mn concentration ranging from 1 to 50 ppb was prepared from commercially available ICP-MS standards (Millipore-Sigma). (Photo-)electrochemistry of 3@SiOx-porSi, 4@SiOx-porSi, and SiOx-porSi^(Photo-)electrochemical measurements using 3@SiOx-porSi, 4@SiOx-porSi, or SiOx- porSi^hybrid photoelectrodes were recorded in a two-compartment H-cell. The working and counter compartments were separated by a Nafion 212 membrane. Both sides of the cell were fitted with a Teflon cap that had holes with O-ring and / or septum, which could be sealed around the electrodes and sparging tubing. Both the working and counter compartment consisted of 250 mMnBu4NPF6, with or without 100 mM phenol as well as with or without 2.0 M NEt3 and 2.0 MiPrOH. The working electrode was 3@SiOx-porSi, 4@SiOx-porSi, or SiOx-porSi^and the counter electrode was a graphite rod. The pseudoreference electrode was an Ag wire. Both compartments were sparged by either Ar or CO2 (with 5% of methane as GC internal standard) for 20 minutes. A cyclic voltammogram was recorded with a glassy carbon electrode to minimize potential drift. An Oriel VeraSol-2 LED solar simulator composed of an LSS-7120 LED controller and an LSH 7520 LED head lamp was used as the light source. The cyclic voltammograms were recorded under 1 sun illumination without stirring. The linear sweep voltammograms were recorded under the same conditions but with stirring. See Figure S34 for schematic illustration of the H-cell. CPE with hybrid electrodes 3@SiOx-porSi, 4@SiOx-porSi, SiOx-porSi^under CO2 CPE experiments were conducted in the same two-compartment H-cell as the cyclic voltammetry experiments. The working compartment consisted of 250 mMnBu4NPF6 and 2.0 M NEt3with 2.0 MiPrOH (or 100 mM phenol). The counter compartment consisted of 250 mMnBu4NPF6 and 2.0 M NEt3 with 2.0 MiPrOH (or 100 mM phenol). The working electrode was the oxidized Si electrode 3@SiOx-porSi, 4@SiOx-porSi, and SiOx-porSi^. The counter electrode was a graphite rod. The reference electrode was an Ag wire. Both compartments were sparged with CO2 (with 5% of methane as GC internal standard) for 20 minutes before electrolysis, while stirring the working compartment. The cell was then sealed before CPE began. During the experiment, the solution in the working compartment was stirred at a speed of 1200 rpm. A VeraSol-2 LED solar simulator was used to illuminate the samples and the headspace was regularly sampled using a Shimadzu Nexis GC-2030 - 50 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) equipped with a thermal conductivity detector (TCD) for quantifying O2, N2, and H2 gases, connected in series with a mechanizer and flame ionization detector (FID) for quantifying CO. Formate was quantified using ion chromatography (IC) by taking 500 µL aliquots of the working solution. A 32-fold dilution of these aliquots using distilled water was performed followed by filtration through a 0.22 µm syringe filter prior to IC analysis using the same instrument described previously. CPE with hybrid electrodes 3@SiOx-porSi under CO2under red-light excitation A CPE experiment was conducted in the same two-compartment H-cell as the cyclic voltammetry experiments. The working compartment consisted of 250 mMnBu4NPF6and 2.0 M NEt3 with 2.0 MiPrOH. The counter compartment consisted of 250 mMnBu4NPF6 and 2.0 M NEt3with 2.0 MiPrOH. The working electrode was the oxidized Si electrode 3@SiOx- porSi. The counter electrode was a graphite rod. The reference electrode was an Ag wire. CPE experiment was performed with 3@SiOx-porSi under the standard conditions (2.0M NEt3 and 2.0 MiPrOH, -1.75 V applied potential) using only red-light excitation (between 600-700 nm) to confirm that the SiOx-porSi material and not 3 was responsible for absorbing light in the photoelectrocatalytic system. At this wavelength 3 does not absorb. CPE with hybrid electrodes 3@SiOx-porSi under CO2under different light intensity Light dependent CPE experiments were conducted in the same two-compartment H- cell as the cyclic voltammetry experiments. The working compartment consisted of 250 mMnBu4NPF6 and 2.0 M NEt3 with 2.0 MiPrOH. The counter compartment consisted of 250 mMnBu4NPF6and 2.0 M NEt3with 2.0 MiPrOH. The working electrode was the oxidized Si electrode 3@SiOx-porSi. The counter electrode was a graphite rod. The reference electrode was an Ag wire. Both compartments were sparged with CO2(with 5% of methane as GC internal standard) for 20 minutes before CPE, while stirring the working compartment. The cell was then sealed before electrolysis began. During the experiment, the solution in the working compartment was stirred at a speed of 1200 rpm. A VeraSol-2 LED solar simulator was used to illuminate the samples. Prior to the experiment, a one-hour CPE under 1 sun illumination was conducted to deplete the surface species, achieving a stable current-time plot. Once the current stabilized, the light intensity was varied at 100-second intervals through a sequence of 1, 0.8, 0.6, 0.4, 0.2, 0.1, 0.05, 0.01, to 0 (dark) Sun, before returning to 1 Sun. - 51 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) CPE with hybrid electrodes 3@SiOx-porSi under argon with additional formate The CPE experiment was conducted in the same two-compartment H-cell as the cyclic voltammetry experiments. The working compartment consisted of 250 mMnBu4NPF6, 14.4 mg (50 ^mol) TBA formate and 2.0 M NEt3with 2.0 MiPrOH. The counter compartment consisted of 250 mMnBu4NPF6and 2.0 M NEt3with 2.0 MiPrOH. The working electrode was the hybrid electrode 3@SiOx-porSi. The counter electrode was a graphite rod. The reference electrode was an Ag wire. Both compartments were sparged with Ar for 20 minutes before electrolysis, while stirring the working compartment. The cell was then sealed before CPE began. During the experiment, the solution in the working compartment was stirred at a speed of 1200 rpm. A VeraSol-2 LED solar simulator was used to illuminate the samples. Formate was quantified using IC by taking 500 µL aliquots of the working solution. A 32-fold dilution of these aliquots using distilled water was performed followed by filtration through a 0.22 µm syringe filter prior to IC analysis. IC was performed using the same instrument described previously. Preparation of 3 immobilized on n-type porous Si (3@SiOx-porSi-ntype) The SiOx-porSi-ntype wafer was submerged in a 5 mL MeCN solution containing 2.0 mM of 3 in a round bottom flask. The flask was sealed covered with aluminum foil, and the solution was heated under N2 at 80 °C for 12 hours. After allowing the solution to cool, the wafer was rinsed with MeCN, soaked in 10 mL ethanol and then rinsed with pentane. Any trace amounts of solvent absorbed in 3@SiOx-porSi-ntype were removed under vacuum (less than 30 mTorr) for 30 minutes prior to characterization. Electrochemistry of 3@SiOx-porSi-ntype Cyclic voltammograms using 3@SiOx-porSi-ntype were recorded in the same two- compartment H-cell as experiments with 3@SiOx-porSi. The working electrode was 3@SiOx- porSi-ntype and the counter electrode was a graphite rod. The reference electrode was an Ag wire. Both compartments were sparged by CO2 (with 5% of methane as GC internal standard) for 20 minutes before electrolysis. The cyclic voltammograms were recorded under ambient light without stirring. A cyclic voltammogram was recorded with a glassy carbon electrode to calibrate the Ag wire reference. Photochemical CO2reduction by homogeneous complexes 3 and 4 under - 52 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) photoelectrochemical conditions These experiments were conducted in the same two-compartment H-cell as the cyclic voltammetry experiments. The working compartment consisted of 0.4 mM 3 or 4, 100 mMnBu4NPF6and 2.0 M NEt3and 2.0 MiPrOH in MeCN. The counter compartment consisted of 100 mMnBu4NPF6 in MeCN. Both compartments were sparged with CO2 (with 5% of methane as GC internal standard) for 20 minutes before the experiment, while stirring the working compartment. The cell was then sealed before the experiment began. A VeraSol-2 LED solar simulator was used to illuminate the catalyst solution and the headspace was sampled using a Shimadzu Nexis GC-2030 equipped with a thermal conductivity detector (TCD) for quantifying O2, N2, and H2gases, connected in series with a mechanizer and flame ionization detector (FID). Drop-casting experiment with complex 3 on SiOx-porSi Complex 3 (1.4 mg, 2.0 mM) was dissolved in 1 mL MeCN in a 4-dram vial.100 µL of a 2.0 mM solution in MeCN was then drop-cast onto the porous area of SiOx-porSi. The resulting porous Si wafer was dried under air at room temperature for 1 hour and characterized by FTIR spectroscopy. The wafer was then rinsed with 20 mL MeCN three times. There was no visual change in the wafer appearance. The resulting washed porous Si wafer was characterized by FTIR spectroscopy. Immobilization of control complex 5 on SiOx-porSi A SiOx-porSi wafer was submerged in a 5 mL MeCN solution containing 2.0 mM of 5 in a round bottom flask. The flask was sealed covered with aluminum foil, and the solution was heated under N2at 70 °C for 18 hours. After allowing the solution to cool, the wafer was rinsed with MeCN, soaked in 10 mL ethanol and then rinsed with pentane. The wafer was characterized by FTIR spectroscopy. Preparation and CPE with hybrid electrode 3@thick-SiOx-porSI A freshly etched hydrogen-terminated porous silicon (H-porSi) wafer was positioned face-up in a crystallization dish and subsequently transferred to an oven set at 190 °C. The sample was heated for 18 hours to facilitate the formation of a thicker silicon oxide (SiOx) layer. The development of this layer was monitored using transmittance FTIR spectroscopy. Experiments Investigating Ability of 3 and 4 to Photochemically Reduce CO2in Solution - 53 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) One explanation for the ability of 3@SiOx-porSi and 4@SiOx-porSi to reduce CO2 to formate is that 3 and 4 leach into solution and then photochemically reduces CO2. In order to eliminate this possibility, a 0.4 mM solution of 3 and 4 was irradiated in the same cell and conditions used for the CPE experiments with 3@SiOx-porSi and 4@SiOx-porSi. The 0.4 mM solution contained far more catalyst than was shown to be present on the surface of the wafer and represents a theoretical maximum as a 0.4 mM solution of 3 and 4 is used to functionalize the p-type wafer to make 3@SiOx-porSi and 4@SiOx-porSi. It was observed that a trace amount of formate was produced under these conditions. This suggests that catalyst leaching, which was shown to be minimal, is not responsible for the observed catalysis using 3@SiOx-porSi and 4@SiOx-porSi. Example 1: Design, preparation, and characterization of porous silicon-supported catalyst compositions (e.g., porous silicon-supported rhenium (Re) catalysts) A common strategy to immobilize a molecular catalyst onto a planar hydrogen- terminated Si wafer is to append an alkene to an ancillary ligand on the catalyst and then perform a hydrosilylation reaction with surface Si–H bonds. To enable this process with porous Si, a hydrogen-terminated porous p-type Si (H-porSi) was prepared using a Galvanostatic etching process. Transmission IR spectra of exemplary H-porSi wafers exhibit three SiHxbands at 2138, 2116, and 2089 cm-1, which are assigned to the SiH3, SiH2, and SiH stretches, respectively (FIG.1B). Additionally, a sharp signal at 908 cm-1is observed, which corresponds to the SiH2bending mode. The intensity of these characteristic SiHxbands can be used to confirm reproducibility in the generation of Si-H sites across different H-porSi samples. To ensure that the IR intensities are meaningful, the IR spectra were carefully aligned and used a wide aperture that sampled most of the porous area. Similar approaches have been previously employed with porous Si to gain semiquantitative information via IR spectroscopy. The band observed at 1104 cm-1is typical of an Si–O stretch. Since Czochralski (Cz) type Si wafers were used, this band reflects the interstitial O atoms within the Si lattice, and the intensity of the absorbance of this Si–O stretch is similar before and after galvanostatic etching. X-ray photoelectron spectroscopy (XPS) of the H-porSi wafer showed a peak at 99.9 eV that is consistent with hydrogen-terminated Si (FIG.2). There was also a small broad feature at 100.7 eV suggesting that the surface was slightly oxidized, presumably from exposure to trace oxygen. To evaluate the surface area of H-porSi, BET analysis with a - 54 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) nitrogen isotherm was performed. The wafers have a BET surface area of approximately 200 m2 / g, which is within the range for other examples of porous Si. The pore morphologies of the H-porSi wafers were explored by cross-sectional SEM (FIGs.1C-1D). The wafers have a pore depth of roughly 3.2 µm and the internal pore structure is not well-ordered but sponge- like with roughened walls. Overall, H-porSi can be easily and reproducibly made, and it can be characterized in detail, a feature which is critical for achieving reproducible catalysis. The Re complex, (Rbpy)Re(CO)3Cl (1,Rbpy = 4-(but-3-en-1-yl)-4'-methyl-2,2'- bipyridine), with a pendant alkene, was prepared according to methods known to those of ordinary skill in the art. Complex 1 was then immobilized via hydrosilylation of a freshly prepared H-porSi wafer. Specifically, a H-porSi wafer was immersed in a 0.4 mM solution of 1 in mesitylene at 160 °C for 18 hours to form 1@porSi (FIG.3A; Scheme 1). Under these conditions, the alkene group on theRbpy ligand reacts with an Si–H bond on the surface to form a covalent Si–C bond. After the hydrosilylation reaction, 1@porSi was rinsed with n- pentane, soaked in ethanol for 2 hours, and then rinsed with n-pentane again to wash away any 1 physisorbed to the surface (FIG.4). To remove any residual solvent trapped within the pores, the sample was dried in vacuo. Control experiments are consistent with covalent attachment of 1 and suggest that the washing procedure removes physisorbed material. To quantify amount of Re attached to the surface, a wafer of 1@porSi was digested in acid and the extract analyzed by Inductively Coupled Plasma Mass Spectrometry (ICP-MS), wherein 341.5 nmol of Re were observed in 1@porSi, which is approximately 1000 times more Re than a monolayer of 1 on a flat Si wafer. 1@porSi was characterized using FTIR spectroscopy (FIG.5A), XPS (FIG.5B), and cross-sectional SEM. The transmittance IR spectrum of 1@porSi exhibited two prominent signals at 2019 cm-1and 1922 cm-1, respectively, which are assigned as carbonyl stretches from an immobilized bpy-ligated Re tricarbonyl complex. The low energy band is broad and may comprise overlapping signals, as has been observed in related systems. The intensity of the carbonyl stretches can be used to determine the approximate relative surface coverage of 1. After immobilization of 1, the intensity of the surface Si–H stretching peak(s) at approximately 2100 cm-1decreased by approximately 60%. This reduction can be attributed to both the hydrosilylation reaction with 1 and surface oxidation. Even though the immobilization was conducted under dry, oxygen-free conditions, the Si–O stretches at 1104 cm-1increased greatly above the intensity expected for an unoxidized Cz-type wafer, and a new peak was observed at 2250 cm-1, corresponding to the partially oxidized OSi–H stretch. - 55 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) XPS analysis provides additional evidence of the immobilization of 1 onto H-porSi. Considering that XPS typically analyzes the top 10-15 nm of a material, which accounts for less than 1% of the pore depth of Si as judged by SEM, two different sampling methods were utilized to probe the distribution of Re throughout the sample. First, an intact sample was analyzed to measure the elemental distribution of the top 10-15 nm of the pores. Second, powdered sample was obtained by scratching and scraping only the pores off the silicon wafer. This measurement provides an average distribution of elements throughout the porous area. Both XPS measurements exhibited a Re 4f doublet, with a binding energy of approximately 41.5 eV for the 4f7 / 2 peak. This confirms the presence of a Re(I) species throughout the porous structure. The N 1s region in the XPS spectrum also displayed features supporting the presence of catalyst (1) on the surface. Cross-sectional SEM confirmed that the internal pore structure of 1@porSi was similar to H-porSi, indicating that the immobilization conditions do not change the wafer. EDS mapping further confirmed the presence of Re in 1@porSi. Example 2: Exemplary catalytic reaction(s) of porous silicon-supported catalyst compositions (i.e., CO2 reduction) The present disclosure describes a non-limiting use of certain porous-silicon supported catalyst compositions (i.e., CO2reduction), for example, utilizing a porous-silicon supported Re catalyst (“Re@porSi” or “1@porSi”) as a hybrid photoelectrode, wherein the Re@porSi or 1@porSi is prepared by hydrosilylation of (alkenebpy)Re(CO)3Cl by an etched, porous silicon wafer. A non-limiting, exemplary reaction cell is described herein, wherein the cell comprises a graphite rod counter electrode, Ag wire reference electrode, and Re@porSi working electrode immersed in 250 mM TBAPF6, 100 mM phenol in MeCN under CO2 (FIG.13). The photoelectrochemical properties of 1@porSi were explored using cyclic voltammetry in MeCN withnBu4NPF6(TBAPF6) electrolyte and 100 mM phenol. Significant current increase was observed with decrease of the potential in CV scans with potential onset around -1.0 V, both in the present or absence of CO2. However, unlike that which was observed using a flat Si wafer immobilized with 1, no discernable redox couples associated with 1 were observed for 1@porSi. An estimated photovoltage of approximately 500 mV was obtained by comparing the catalytic onset potentials in the cyclic voltammograms of 1@porSi under CO2with 1 sun illumination, as compared to that of a darkened n-type porous Si wafer with 1 attached (1@porSi (n-type); FIG.6A). This is consistent with many reported - 56 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) photovoltages using p-Si photoelectrodes, which range from 0.2 to 0.6 V. To examine the products of photoelectrocatalysis by 1@porSi, controlled potential electrolysis (CPE) experiments were run using a CO2-saturated solution of 250 mM TBAPF6 in MeCN with 100 mM phenol as the proton source. The applied potential was at -1.71 V (versus Fc+ / Fc) and the illumination was 1 sun using a solar simulator. CO and H2 were the only detected products, wherein CO is the only detected carbon product. Both products were quantified by sampling the headspace and analyzing them using gas chromatography. A significant amount of charge (~8 C) was passed in less than 6 hours using 1@porSi, with little decay over time (FIG.7). After approximately 5.5 hours of CPE, around 36 μmol of CO and 7 μmol of H2were produced, with Faradaic efficiencies (FE) for CO (FECO) and H2(FEH2) of 91% and 18%, respectively (FIG.6B). The FEH2 rapidly dropped to less than 10% after an hour and remained low for the rest of the photoelectrocatalysis (FIG.6C). During the same time period, FECO remained relatively consistent in the range of 70-95%. Both FECO and CPE current show slow decrease over time, possible due to catalyst deactivation or degradation of the porous structure. Isotopic labeling reaction was performed with13CO2. Under identical conditions, the electrolysis produces13CO (as confirmed by GC-MS), which confirmed that CO was catalytically produced in this reaction. Control experiments were performed to test the contribution of the H-porSi support to the CO2and H2production by 1@porSi. CPE using freshly etched H-porSi, without any catalyst, under standard catalytic conditions indicated smaller currents than CPE with 1@porSi. The primary product was H2with a small amount of CO (hydrogen-terminated Si materials have been reported to reduce CO2). Under argon instead of CO2, only H2 was formed. Another experiment used a H-porSi wafer that had been heated at 160 °C for 18 hours in just mesitylene, the same conditions used for the immobilization of 1 but without any 1. The results with this wafer under standard CO2reduction conditions were similar to those for freshly etched H-porSi, with significantly less current than for 1@porSi and more H2than CO. These experiments demonstrate that immobilization of 1 on H-porSi is crucial for generating significant quantities of CO using 1@HporSi. While the overall FE of H2is 18% for 1@porSi, an apparent FEH2of 145% was observed in the first 10 minutes of photoelectrolysis. Similarly, in the formation of H2 in CPE by H-porSi (no catalyst), FEH2values exceeded 300% after 95 minutes for electrolyses under CO2 or Ar. Without wishing to be bound by any theory, these faradaic efficiencies suggest that the hydrogen evolution reaction is occurring at least partially by a non-electrochemical or partially electrochemical pathway for the porous electrodes with and without catalysis. - 57 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Careful CPE experiments using deuterated phenol suggest hydrides on the silicon electrode are participating in this pathway leading to inflated FEH2values (FIGs.10A-10C and FIGs. 11A-11C). The 1@porSi hybrid photoelectrocatalyst performs dramatically better than the same rhenium catalyst hydrosilylated to (i.e., covalently conjugagted to) a planar p-doped Si wafer. Under the same catalytic conditions, 1@porSi reduces CO2to CO with 90% Faradaic efficiency, while 1 on a planar wafer has a FECO = 18%.1@porSi gives relatively stable catalytic performance, with CO generated at an almost constant rate over 6 hours (FIG.6A), while 1 on a planar wafer decomposes after 15 minutes. XPS (FIGs.12A-12B) and FTIR spectra (FIG.12C) of 1@porSi before and after CPE showed that the large majority of 1 (~75%) is still present on the surface electrolysis. Reproducible catalytic results are obtained using 1@porSi, with very similar kinetics, FEs, and current densities achieved over three trials, versus the irregular behavior of 1 on planar wafers (despite much effort). The reproducibility is in part due to the ability to gauge the quality of the initial H- porSi wafer and then control the coverage obtained during immobilization of 1, using simple and semiquantitative characterization by transmission IR spectroscopy. This degree of control is unique in the immobilization of molecular catalysts on solid supports. Example 3: Preparation and Characterization of the Mn(I) Complex All Mn(I) complexes were synthesized under an N2 atmosphere using standard Schlenk techniques or an MBraun glovebox. [4-(3-butenyl)-4'-methyl-2,2'-bipyridine] manganese tricarbonyl bromide (1b) 2,2'- bipyridine (L1, 80 mg, 0.36 mmol), THF (10 mL), and a stir bar were added to a Schlenk flask containing a greaseless Teflon stopcock seal to form a dark orange solution. The flask was covered with foil and heated at 55 °C in an oil bath overnight. The solvent was removed under vacuum and the flask was brought back into the glovebox. The yellow solids were washed with pentane and Et2O (2 x 3 mL) to remove an orange impurity, then dissolved in - 58 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) THF to form a deep red solution, filtered, and the solvent was removed under vacuum to afford 1b as bright yellow / orange microcrystals (127 mg, 85%).1H NMR (CDCl3): δ 9.09 (d, 1H, bpy), 9.06 (d, 1H, bpy), 7.90 (s, 2H, bpy), 7.32 (s, 2H, bpy), 5.85 (m, 1H, =CH), 5.13- 5.07 (m, 2H, =CH2), 2.89 (m, 2H, CH2), 2.55 (s, 3H, Me), 2.48 (m, 2H, CH2) ppm.13C{1H} NMR (CDCl3): δ 221.9, 155.5, 155.4, 153.9, 153.3 (2C), 153.2, 136.2 (=CH), 127.3, 126.6, 123.1, 122.4, 116.8 (=CH2), 34.8, 33.9, 21.5 ppm. FTIR: 2024, 1905br cm−1. UV-Vis (DCM, λ, ε): 423 (1800) nm. HRMS (ESI) m / z: [M − Br]+Calcd for C18H16N2O3Mn 363.0536; Found 363.0515. [4,4-bis(3-butenyl)-2,2'-bipyridine] manganese tricarbonyl bromide (2) , -2,2'- bipyridine (L2, 96 mg, 0.36 mmol), THF (10 mL), and a stir bar were added to a Schlenk flask containing a greaseless Teflon stopcock seal to form a dark orange solution. The flask was covered with foil and heated at 55 °C in an oil bath overnight. The solvent was removed under vacuum and the flask was brought back into the glovebox. The yellow solids were washed with pentane and Et2O (2 x 3 mL) then dissolved in DCM, filtered, and solvent was removed under vacuum to afford 2 as a bright yellow solid (155 mg, 89%).1H NMR (CDCl3): δ 9.11 (d, 2H, bpy), 7.90 (s, 2H, bpy), 7.34 (d, 2H, bpy), 5.68 (m, 1H, =CH), 5.13- 5.08 (m, 2H, =CH2), 2.90 (m, 4H, CH2), 2.48 (m, 4H, CH2) ppm.13C{1H} NMR (CDCl3): δ 221.9, 155.4, 153.3 (bpy CH), 136.2 (=CH), 126.6 (bpy CH), 122.4, 116.8 (=CH2), 34.8 (CH2), 33.9 (CH2) ppm. FTIR: 2024, 1911br cm−1. UV-Vis (DCM, λ, ε): 425 (2600) nm. HRMS (ESI) m / z: [M − Br]+Calcd for C21H20N2O3MnBr 403.0849; Found 403.0837. [4-(4-butylsilatrane)-4'-methyl-2,2'-bipyridine] manganese tricarbonyl bromide (3) mg, 0.621 - 59 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) mmol), Mn(CO)5Br (170 mg, 0.618 mmol), THF (10 mL), and a stir bar were added to a Schlenk flask containing a greaseless Teflon stopcock seal to form a dark orange / red solution. The solution was covered in foil and stirred overnight. The solvent was removed under vacuum to afford dark red / orange microcrystals. The solids were washed with pentane and Et2O (2 x 3 mL), redissolved in THF to form a dark red / orange solution, filtered, and dried under vacuum to afford 3 as a bright orange powder (353 mg, 92%).1H NMR (CDCl3): δ 9.04 (s, 2H, bpy), 7.90 (d, 2H, bpy), 7.30 (s, 2H, bpy), 3.76 (s, 6H, OCH2), 2.81 (m, 8H, NCH2+CH2), 2.55 (s, 3H, Me), 1.86 (s, 2H, CH2), 1.71 (s, 2H, CH2), 1.50 (s, 2H, CH2), 0.49 (s, 2H, SiCH2).13C{1H} NMR (CDCl3): δ 222.0, 155.8, 155.6, 155.2, 153.1 (bpy CH), 150.3, 127.2 (bpy CH), 126.9 (bpy CH), 123.1 (bpy CH), 122.5 (bpy CH), 57.9 (OCH2), 51.2 (NCH2), 35.4 (CH2), 34.2 (CH2), 24.8 (CH2), 21.5 (Me), 16.3 (SiCH2) ppm. FTIR: 2030, 1932sh, 1919 cm−1. UV-Vis (DCM) λmax, nm (ε): 406 (1900). UV-Vis (THF) λmax, nm (ε): 413 (1900), 296 (12,000) 257sh (10,000). HRMS (ESI) m / z: [M − Br]+Calcd for C24H29N3O6SiMn 538.1201; Found 538.1178. [4,4'-bis(4-butylsilatrane)-2,2'-bipyridine] manganese tricarbonyl bromide (4) mg, 0.325 mmol), Mn(CO)5Br (89 mg, 0.32 mmol), THF (10 mL), and a stir bar were added to a Schlenk flask containing a greaseless Teflon stopcock seal to form a dark orange solution. The solution was covered in foil and stirred overnight. The solvent was removed under vacuum to afford a bright yellow powder. The solids were washed with pentane and Et2O (2 x 3 mL), redissolved in DCM to form an orange solution, filtered, and dried under vacuum to afford 4 as a bright yellow / orange powder (250 mg, 94%). Yellow crystals suitable for X-ray diffraction of 4∙2(CH2Cl2) were grown from a concentrated DCM solution of 4 layered under Et2O at −35 °C.1H NMR (CDCl3): δ 9.02 (m, 2H, bpy), 7.88 (s, 2H, bpy), 7.29 (m, 2H, bpy), 3.75 (s, 12H, OCH2).2.80 (s, 16H, NCH2+CH2), 1.72 (s, 4H, CH2), 1.50 (s, 4H, CH2), 0.49 (s, 4H, SiCH2) ppm.13C{1H} NMR (CDCl3): δ 155.7, 155.4, 152.9 (bpy CH), 126.6 (bpy CH), 122.5 (bpy CH), 57.9 (OCH2), 51.2 (NCH2), 35.4 (CH2), 34.2 (CH2), 24.8 (CH2), 16.2 (SiCH2) ppm. Resonances associated with the CO peaks were not observed. FTIR: 2024, - 60 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) 1923, 1899 cm−1. UV-Vis (DCM) λmax, nm (ε): 411 (1500) 310sh (4900), 294 (9700), 246sh (11,000). HRMS (ESI) m / z: [M − Br]+Calcd for C33H46N4O9Si2MnBr: 753.2179; Found 753.2148. [4,4'-dimethyl-2,2'-bipyridine] manganese tricarbonyl bromide (5) Complex 5 was synthesized according to a method reported in the literature. Example 4: Preparation of a Thermal Silicon Oxide Layer on Porous Si (SiOx-porSi) One of the challenges associated with using H-porSi, and other types of hydrogen- terminated Si, as a support is that it is not possible to obtain a perfect monolayer of a catalyst via hydrosilylation for steric reasons. As a consequence, Si-H bonds remain on the surface after catalyst immobilization. Although the formation of a native oxide layer for freshly etched porous Si can be extremely slow under ambient conditions, the surface can oxidize when subjected to heat, which is required for hydrosilylation, or under the conditions used for CO2 photoelectroreduction. The subsequent restructuring of the surface complicates analysis. In addition, the apparent production of H2 from reactive Si-H sites and the ability of some Si- H sites to reduce CO2 stoichiometrically introduces further complexity. It has been postulated herein that a solution would be to generate a protective silicon oxide layer on H-porSi that does not change throughout functionalization and electrochemical experiments. H-porSi was prepared using a previously reported procedure in which low- doped p-type wafers (1-10 ^ cm2) were assembled in a teflon cell, exposed to a 1:1 solution of HF:EtOH and 40 mA current was applied for 5 minutes (FIG.15A). The H-porSi was characterized using FTIR spectroscopy, where three characteristic hydride stretches (νSi-Hx) are observed around ~2100 cm-1(FIG.22). BET analysis was used to determine the surface area and scanning electron microscopy was utilized to examine the pore morphology (FIG. 29). It has previously been demonstrated that thermal oxides can be formed by heating porous Si to high temperatures (typically 700-1200 °C) in the presence of oxygen but this depends on the exact type of porous Si and in some cases leads to damage of the porous structure. Here, relatively gentle heating of H-porSi to 190 °C was sufficient to create an oxide layer that does not cause physical damage to the porous structure and has surface Si- OH sites for catalyst attachment (FIG.15A). As shown in FIG.15B, the formation of the oxide layer in these oxidized porous Si samples (SiOx-porSi) was monitored by FTIR - 61 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) spectroscopy. As soon as one hour after heating, the characteristic hydride stretches (νSi-Hx) around ~2100 cm-1decrease as porous Si is oxidized. Stretches assigned to backbonded hydrides (νOSi-H) around ~2200 cm-1and Si oxygen bonds (νSi-O) around 1100 cm-1appear as the surface oxidizes. These changes are similar to those reported for oxidation of other porous Si samples. The increase in the OH stretching band at ~3300 cm-1is likely due to the generation of Si-OH bonds, but there could also be some physisorption of water molecules in the pores since the oxide is expected to be more hydrophilic than H-porSi. The predominance of hydrogen bonded OH groups at ~3300 cm-1over isolated OH (ca.3600 cm-1) may also suggest the presence of water. While FTIR spectroscopy cannot be used to gauge the thickness or evenness of an oxide layer, the decrease of the initial Si-Hx stretch can be used to gauge the extent of surface oxidation. Here, after seven hours, the intensity of the Si-Hx stretches is reduced by approximately 80%, indicating the majority of Si atoms on the surface have been at least partially oxidized. This suggests that less than a monolayer of oxide is present if the pores oxidize evenly. For electrocatalytic applications, the presence of a thin oxide layer is optimal as electrons need to be passed through this insulating, protective layer to any attached catalyst. Consequently, used herein are SiOx-porSi samples that had been heated for seven hours and control experiments demonstrated that although further heating leads to a continued decrease in the Si-Hx features, the electrocatalytic behavior is worse (vide infra). A major advantage of the facile characterization of SiOx-porSi using FTIR spectroscopy is that this technique can be readily used to ensure reproducibility across samples in the etching and oxidation steps by ensuring the absorbance of the νSi-Hx, νOSi-H, and νSi-O are roughly comparable in the SiOx-porSi samples (FIG.25). Samples of SiOx-porSi were additionally characterized by XPS (FIG.16C and FIGs. 27-28) and cross-sectional SEM and EDS (FIG.15D). Since XPS typically only probes approximately the top 10 nm of a sample, measurements were recorded on two types of samples, intact and powdered. Intact samples were unmodified SiOx-porSi. The two characteristic Si 2p signals, at 99 and 103 eV, confirm the presence of both elemental Si and SiOx, respectively, at the top of the pores (FIG.15C). The powdered samples were created by scraping and grinding up the porous Si area and pressing the powder into indium metal. The Si 2p signal at 103 eV in the black trace in FIG.15C indicates that the oxide is found throughout the porous structure. Cross-sectional SEM confirms that the overall porous structure does not greatly change after oxidation (FIG.15D) and the sample of SiOx-porSi - 62 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) shows roughly the same pore depth and structure as H-porSi (FIG.24). Further, EDS mapping shows a higher concentration of Si in the bulk Si than in the porous region and oxygen maps perfectly over the porous area. Taken together XPS and EDS confirm the presence of an oxide layer throughout the porous structure. Finally, as determined by FTIR spectroscopy, samples of SiOx-porSi are stable when left in ambient conditions for at least seven days (FIGs.26A-26H). Overall, SiOx-porSi has many attractive qualities as a support because of its stability, reproducibility, facile fabrication, and easy characterization. Example 5: Synthesis and Characterization of Mn(I) Catalysts with Surface Attachment Groups It was recently demonstrated that the Mn(I) complex (bpy)Mn(CO)3Br electrocatalytically reduces CO2to formate with FE of approximately 60% in an acetonitrile (MeCN) solution containing 2.0 M isopropanol (iPrOH) and 2.0 M triethylamine (NEt3). To generate an immobilized system for CO2reduction to formate, variants of (bpy)Mn(CO)3Br were synthesized with functional groups on bpy that would enable covalent attachment SiOx- porSi. To compare the performance of SiOx-porSi to H-porSi as a support, complexes were also prepared with functional groups on bpy that would enable immobilization on H-porSi. The new complexes 1b and 2 contain either one or two pendant alkenes, respectively, on the bpy ligand for immobilization through hydrosilylation to H-porSi. The new complexes 3 and 4 contain either one or two pendant silatrane groups, respectively, on the bpy ligand. The silatrane groups can undergo hydrolysis-type reactions with surface Si-OH groups to form strong Si-O bonds to the surface. The silatrane group was selected because it is air and moisture stable and results in strong binding to metal oxide surfaces upon mild heating. The literature compound (4,4'-Me2bpy)Mn(CO)3Br (5, 4,4'-Me2bpy = 4,4'-dimethyl-2,2'- bipyridine) served as control compound. Complexes 1b, 2, 3, 4, and 5 (i.e., 1b-5) were prepared in yields of 85-94% through the reaction of the substituted bpy ligands with Mn(CO)5Br in THF (FIG.16A). All complexes were fully characterized by1H and13C NMR, FTIR, and UV-Vis spectroscopies. Complex 4 was also characterized by single crystal X-ray diffraction analysis and has similar geometric parameters around Mn to related complexes without silatranes (FIG.16B). Cyclic voltammograms of complexes 1b-5 were recorded in MeCN withnBu4NPF6as the electrolyte under Ar (FIG.16C and FIG.31). Two quasi-reversible redox events were observed at approximately -1.75 V and -2.2 V vs ferrocenium / ferrocene (Fc+ / Fc). These are assigned as reduction of the Mn center (MnI / Mn0) and then one-electron reduction of the bpy ligand - 63 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) ([Mn0bpy] / [Mn0bpy-]), following literature reports of other (bpy)Mn(CO)3X (X = Cl or Br) complexes (FIG.31). Cyclic voltammograms of 1b-5 recorded under 1 atm of CO2in the presence of either 100 mM phenol (PhOH) or with 2.0 MiPrOH and 2.0 M NEt3 display significant current enhancement at the potentials of both the first and second reduction waves (FIG.16C and FIGs.32-34). The current enhancement at the second wave is more significant, but these results are consistent with electrocatalytic CO2reduction at both potentials. This is in agreement with the presence of both low and high overpotential pathways for electrocatalytic CO2reduction by this class of catalysts. To determine the products from electrocatalytic CO2 reduction using 1b-4, controlled potential electrolysis (CPE) was performed (FIGs.35-36, FIGS.37A-37F, FIGs.38A-38D, FIGs.39A-39D, FIGs.40A-40D, and FIGs.41A-41C). The headspace was analyzed using gas chromatography to detect CO and H2, and aliquots from the reaction solution were analyzed by ion chromatography to detect formate. FIG.16D summarizes the results using 3 with different proton sources at potentials of -1.75 V. In the presence of PhOH, CO is the major product (FE = 67% after 2.5 hours, FIGs.39A-39D and FIGs.40A-40D), whereas in the presence of NEt3 andiPrOH, formate is the major project (FE = 82% after 2.5 hours, FIG. 36 and FIGs.37A-37F). In the case of 3, even higher than expected FEs for formate are observed with NEt3 andiPrOH compared to (bpy)Mn(CO)3Br. As the applied potential was increased from -1.75 V to -2.15 V in the presence of NEt3andiPrOH, formate remains the major product although the FE decreases and more H2 is produced (FIG.16E). There is an increase in current density at higher applied potentials. Similar results to 3 were observed for compounds 1b, 2, 4, and 5 in CPE experiments under CO2 (FIG.36, FIGs.38A-38D, FIGs. 39A-39D, and FIGs.41A-41C). Overall, these combination of spectroscopic and electrochemical characterization, catalysis, and crystallographic data suggests that modifications made to the bpy ligand to enable immobilization do not significantly affect the reactivity of the series of Mn complexes, with 1b-4 behaving analogously to 5. Example 6: Catalyst Immobilization on Porous Si Initially, attempts were made to immobilize 1b and 2 through a hydrosilylation reaction with H-porSi using the procedure previously utilized for molecular Re complexes. However, this strategy was not successful due to decomposition of 1b and 2 at the high temperatures required (160 °C). Attempts to perform the functionalization at lower temperatures (120 °C) did not lead to decomposition of 1b or 2 but no complex was immobilized as determined by FTIR spectroscopy, suggesting that high temperatures are - 64 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) required for hydrosilylation. An alternative strategy to facilitate hydrosilylation reactions is via sonication at room temperautre, but sonication of the H-porSi damaged the wafer. These experiments highlight some of the limitations associated with immobilizing reactive catalysts on H-porSi. Complexes 3 and 4 were attached to SiOx-porSi by heating a heterogeneous mixture containing an MeCN solution of the complex and SiOx-porSi at 80 °C for eighteen hours (FIG.17A). The wafer was then rinsed with pentane and MeCN, soaked in 10 mL ethanol for an hour, and then rinsed again with pentane. The resulting hybrid materials, 3@SiOx-porSi and 4@SiOx-porSi, were characterized using a series of different techniques (FIGs.42A-42D, FIGs.43A-43D, FIGs.44-46, FIGs.47A-47H, FIG.48, FIGs.49A-49H, and FIGS.50-51). To quantify the amount of Mn attached to the surface, wafers of 3@SiOx-porSi and 4@SiOx- porSi were digested in 2.0 M KOH aqueous solution, then acidified with 70% nitric acid, and the extracts were analyzed by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) (Table 1, FIGs.50-51). Base was used because KOH dissolves the porous structure in approximately 10 minutes ensuring that all attached catalyst is digested. Samples from 3@SiOx-porSi and 4@SiOx-porSi with a geometric area of 1.2 cm2contained 810(80) nmol and 530(10) nmol, respectively, of Mn. Table 1. ICP-MS data used to calculate the amount of Mn in 3@SiOx-porSi, 4@SiOx-porSiaMn CatalystMeasured conc. ( ^g / L)Dilution factor Total Mn nmol . FTIR spectroscopy was used to examine the attachment of 3 and 4 via their characteristic CO stretches and changes in the extent of oxidation of the SiOx-porSi support (FIG.17B, FIGs.42A-42D, and FIGs.43A-43D). The FTIR spectra of 3@SiOx-porSi and 4@SiOx-porSi exhibited two prominent CO stretching bands between 1930 and 2030 cm-1(FIG.17B). The frequency of the CO stretches is similar to those observed for powder samples of 3 and 4 suggesting that the complexes are still intact on the surface. Comparison of the FTIR spectra of 3@SiOx-porSi and 4@SiOx-porSi with that of SiOx-porSi indicated that there is some further oxidation of the surface during silatrane attachment. This is not unexpected given that there were still some reactive Si-H bonds on the surface of SiOx-porSi, - 65 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) which likely react with trace O2 and H2O (in the MeCN) at 80 °C. The oxidation of surface Si-H bonds is supported by the 50% decrease in the intensity of the surface Si-H stretching bands at approximately 2100 cm-1, along with an increase in the intensity of the v(Si–O) bands. Overall, the changes to the surface, as determined by FTIR spectroscopy, are significantly smaller upon catalyst attachment to SiOx-porSi compared to H-porSi, suggesting that it is a more stable support. The XPS spectra of 3@SiOx-porSi and 4@SiOx-porSi were collected from both intact and powdered samples (FIGs.17C-17D, FIGs.44-46, and FIGs.47A-47H). FIG.17C shows the Mn 2p region of the XPS spectra of 3@SiOx-porSi, while the related data for 4@SiOx- porSi is in FIGs.47A-47H. For 3@SiOx-porSi and 4@SiOx-porSi the Mn 2p doublet was observed for intact and powdered samples providing evidence for the presence of Mn both on the surface and in the pores. The major Mn peaks appear at ~654 eV and ~641 eV, consistent with Mn in the +1 oxidation state and the presence of an immobilized version of 3 or 4. Additionally, the Mn 2p spectra exhibit weaker satellite peaks at ~646 and ~658 eV. These same satellite peaks are observed in powered samples of molecular 3 and 4, again suggesting that the complexes are intact on the surface. A comparison of the Si 2p region of the XPS spectra of powdered 3@SiOx-porSi (which examines the bulk sample) with unfunctionalized SiOx-porSi indicates the ratio of silicon oxide (~103 eV) to elemental Si (~99 eV) remained relatively unchanged, suggesting no significant change to the bulk. However, a slight increase in the ratio of silicon oxide to Si was observed in the intact samples of 3@SiOx-porSi suggesting there was some surface oxidation during immobilization in agreement with results from FTIR spectroscopy and previous observations for immobilization of a Re catalyst on H- porSi through hydrosilylation Cross-sectional SEM showed that the internal pore structures of 3@SiOx-porSi and 4@SiOx-porSi were similar to that of SiOx-porSi, highlighting that the catalyst immobilization procedure did not significantly impact the structure of the wafer (FIG.18A and FIGs.49A-49H). EDS elemental mapping spectra were used to elucidate the elementary composition inside the pores of 3@SiOx-porSi and 4@SiOx-porSi (FIG.18A and FIG.50). These spectra show that Mn and N are both present throughout the porous section of the material, consistent with the intact catalyst being present in the pores. To further test for the presence of 3 in the pores of 3@SiOx-porSi, a cross-section of 3@SiOx-porSi was observed at high magnification using aberration-corrected scanning transmission electron microscopy (STEM). To meet the electron transparency requirement for STEM observation, a thin layer (lamella) of the sample was made using a focused ion beam (FIB) lift-out technique. A high- - 66 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) magnification image from the porous Si area is shown in FIG.18B. The image contains many bright spots, which are hypothesized to be individual Mn atoms. Consistent with this proposal when EDS and electron energy loss spectroscopy (EELS) were performed on a sample region containing the bright spots, the Mn K and L edges are clearly shown in the EDS and a peak at around 644 eV, corresponding to Mn, is detected by EELS. As in the SEM-EDS, signals consistent with N, Si, and O were observed in the porous Si layer in the TEM-EDS. To gain evidence for covalent attachment of the complexes, a solution of 3 in MeCN was drop-cast onto a SiOx-porSi wafer, which presumably resulted in physisorption of 3 (FIG.52). The CO region of the surface FTIR spectrum of this sample showed peaks at 2026 and 1938 cm-1. However, when the SiOx-porSi wafer containing the drop-cast sample was washed with MeCN, no signals associated with the CO stretches remain in the IR spectrum, suggesting that the procedure of soaking in MeCN removes any complexes physisorbed to the surface. Moreover, soaking a SiOx-porSi wafer in an MeCN solution of 5 under N2 at 80 °C for 18 hours and subsequently washing the wafer with pentane and ethanol (the same functionalization procedure used for 3 and 4) resulted in no CO stretches being observed by FTIR spectroscopy (FIG.53). This was expected since 5 lacks a functional group for covalent attachment to the surface. Overall, the characterization data shows that 3 and 4 have been successfully immobilized on SiOx-porSi and are likely intact and present in the pores of the material. Covalent attachment of the complexes is supported by washing experiments and electrochemical results are described elsewhere herein. Example 7: Photoelectrochemical CO2Reduction Using 3@SiOx-porSi and 4@SiOx- porSi Cyclic voltammograms were obtained for 3@SiOx-porSi and 4@SiOx-porSi in contact with the standard electrolyte solution, 0.25 MnBu4NPF6in MeCN, with one sun illumination under Ar or CO2, in the absence of any added proton source. All the cyclic voltammograms (and the linear sweep voltammogram for 3@SiOx-porSi) are quite similar, with a gradual increase in current observed starting at approximately -1.2 V (FIG.19A, FIG. 56, and FIGs.61A-61B). The lack of well-defined signals for the immobilized catalyst have been attributed to the complexities associated with using porous Si as an electrode, which have been described previously. A notable feature of the cyclic voltammograms of 3@SiOx- porSi under Ar and CO2 with one sun illumination is that the initial scan is similar to subsequent scans. This stands in contrast to studies using catalysts immobilized on H-porSi, where the initial cyclic voltammograms were dominated by electrochemical reactions that - 67 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) occur at the surface. The absence of an initial surface response from SiOx-porSi indicates that it is an electrochemically well-behaved system for immobilizing catalysts, especially compared to H-porSi. When cyclic voltammograms of 3@SiOx-porSi and 4@SiOx-porSi were recorded with one sun illumination under a CO2 atmosphere in the presence of 2.0 M NEt3 and 2.0 MiPrOH, significant photoelectrocatalytic current is observed with an onset of approximately - 1.3 V (FIGs.19A-19B, FIGs.58A-58B, and FIGs.59A-59B). The current remains constant in five consecutive scans, suggesting that a stable photoelectrocatalytic process is occurring (FIG.19B). Similarly, when cyclic voltammograms of 3@SiOx-porSi and 4@SiOx-porSi were recorded with one sun illumination under a CO2atmosphere in the presence of 100 mM PhOH, significant current was observed starting at around -1.2 V (FIGs.59A-59B). Control experiments with a photoelectrode that had been subjected to the silatrane immobilization conditions but without the Mn catalyst (denoted as SiOx-porSiΔ) did not show a significant amount of current under standard conditions (one sun illumination, CO2atmosphere, 2.0 M NEt3, and 2.0 MiPrOH; FIG.19B and FIG.60). A small amount of current is generated when SiOx-porSiΔis reduced under a CO2 atmosphere with 100 mM PhOH in the presence of 1 sun illumination, but this is far less than for systems containing Mn (FIG.60). These results show that the presence of Mn is required to generate significant current during photoelectrocatalysis. CPE at -1.75 V was used to examine the product distribution from photoelectrocatalytic reduction of CO2using 3@SiOx-porSi under standard conditions (FIGs. 20A-20B). Seven hours of CPE with 3@SiOx-porSi formed 25.5 mmol of formate with ~7 C of charge passed (FIG.20A and 63). This corresponds to a FE for formate of 71%, with only a small amount of H2 generated and almost no CO. No liquid products, such as methanol, were detected using NMR spectroscopy after seven hours of CPE. Earlier in the electrolysis, after forty minutes, the FE for formate was a remarkable 96%, with almost no detectable H2 or CO generated (FIG.20C and FIG.64). The geometric current density decreased by 33% from its maximum at 0.6 mA / cm2after forty minutes to 0.4 mA / cm2after seven hours. Interestingly, the geometric current density observed in CPE is similar to the current density observed in cyclic voltammetry at -1.75 V, which is indicative of a relatively stable catalytic system. Based on the geometric current density from CPE after forty minutes and the amount of Mn on the surface (from ICP-MS), the maximum TOF at -1.75 V is 11.4 h-1. A photoelectrochemical reduction of13CO2using 3@SiOx-porSi produced13C- labelled formate (by GC-MS and NMR spectroscopy, FIG.80, FIGs.81A-81B, and FIGs. - 68 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) 82A-82D), confirming that CO2 is the source of the formate generated in the reaction. Consistent with cyclic voltammetry experiments, a control CPE evaluating the photoelectrocatalytic activity of a SiOx-porSiΔelectrode without any attached catalyst under the same conditions passed very little charge, ~0.5 C after four hours, and produced predominantly H2 (1.0 mmol) with almost no detectable CO and formate (FIGs.74A-74C). Thus, the immobilization of 3 is crucial for generating significant quantities of formate, and SiOx-porSiΔby itself is not efficient for CO2 reduction. A photovoltage for electrocatalysis of approximately 280 mV was obtained by comparing the current produced by 3@SiOx-porSi versus a photoelectrode containing 3 immobilized on a degenerately doped n-type porous Si wafer at a potential of -1.75 V using cyclic voltammetry (FIG.19C). An applied potential of - 1.75 V was selected because this was the potential at which formate was produced with the highest FE. The 280 mV photovoltage is consistent with the many reported photovoltages using p-type Si photoelectrodes (0.2 to 0.6 V). Unfortunately, overpotential for photoelectrocatalytic CO2reduction using 3@SiOx-porSi cannot be estimated because the species that are present in the mixture generated from CO2 Et3N, andiPrOH in MeCN are not clear. The product distribution for CPE experiments after forty minutes using 3@SiOx-porSi was analyzed at various potentials ranging from -1.55 V to -2.15 V under standard conditions (FIG.20D, FIGs.63-64, FIGs.65A-65C, and FIGs.66A-66C). A strong dependence on current was observed. At less forcing potential, -1.55 V, only small amounts of formate (1.22 mmol), H2(1.10 mmol), and CO (0.83 mmol) were made with low FE (FIG.63). As described above, there is a significant amount of formate produced at -1.75 V. When the potential is increased further to -1.95 V, there is an almost doubling of the geometric current density after forty minutes, but there is a decrease in the FE for formate from 96% to 81%. At -2.15 V, there is a further linear increase in current density, accompanied by a decrease in the FE for formate to 60%. The maximum TOF of 11.7 h-1is observed at 2.15 V. For a surface electrocatalyst it might be expected that the geometric current density would rise exponentially with increasing applied potential. The observed linear increase suggests that the system does not behave ideally, which could be related to the porous nature of the material. To evaluate whether the SiOx-porSi material (and not 3) was responsible for absorbing light in the photoelectrocatalytic system, a CPE experiment was performed with 3@SiOx- porSi under standard conditions using only red-light excitation (between 600-700 nm) (FIGs. 78A-78D). At this wavelength 3 does not absorb. After 5 hours ~20 mmols of formate were produced, which is almost the same as under the standard conditions, and consistent with a - 69 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) small decrease in the geometric current density. The FE for formate was ~95%, with almost no detectable H2or CO generated. Thus, photoelectrocatalytic reduction of CO2using 3@SiOx-porSi results from the light absorption properties of Si, which facilitates charge separation, and multielectron catalysis. An additional experiment tested the effect of light intensity on CO2 reduction using 3@SiOx-porSi, by systematically decreasing the intensity of light used in CPE from 1 to 0.01 sun and then to dark under standard conditions (FIG.79). The catalytic current density was dependent on the intensity of light and after the irradiation was stopped, resubjecting the sample to one sun illumination returned the CPE to the original current. LSV scans with 3@SiOx-porSi also showed changes in current with variation in light intensity and established that the stirring speed impacted the measured current although under (FIGs.61A-61B). To complement these experiments with 3@SiOx-porSi, the product distribution from photoelectrocatalytic CO2 reduction was determined using 4@SiOx-porSi through a CPE experiment at -1.75 V under standard conditions (FIG.20C and FIGs.67A-67C). After forty minutes the FE for formate was 82%, and unlike for 3@SiOx-porSi a measurable amount of H2 was generated (FEH2 = 6%). The maximum TOF to formate of 4@SiOx-porSi at -1.75 V after forty minutes is 13.1 h-1, which is higher than for 3@SiOx-porSi. Other trends for 4@SiOx-porSi are the same as for 3@SiOx-porSi. For example, the catalytic activity of 4@SiOx-porSi decreased by a factor of 20% over seven hours at -1.75 V (compared to 25% for 3@SiOx-porSi) and the current density increases as the applied potential in photoelectrocatalysis is increased from -1.75 V to -2.15 V. Notably, although the FE for formate decreases with increasing applied potential for 4@SiOx-porSi, the decrease is smaller than for 3@SiOx-porSi (10% decrease for 4@SiOx-porSi vs 35% for 3@SiOx-porSi). The maximum TOF to formate for 4@SiOx-porSi was achieved at -2.15 V and is 23.3 h-1(FIGs. 69A-69C). Comparison of the selectivity for photoelectrocatalytic CO2 reduction of 3@SiOx- porSi and 4@SiOx-porSi with electrochemical reduction of CO2in solution by 3 and 4 under these standard conditions (without illumination in solution) reveals an increase in selectivity to formate. For example, after forty minutes under standard conditions at a potential of -1.75 V, 3 gives 82% FE to formate in solution, whereas 3@SiOx-porSi gives 96% (FIG.16E and FIG.20C). The change in selectivity is even more notable when CO2reduction is performed in the presence of 100 mM PhOH as the proton source. In solution both 3 and 4 generate CO as the major product (FEs for CO are 67 and 61%, respectively) at applied potentials of - 1.75 V (FIGs.39A-39D). In contrast, when immobilized, 3@SiOx-porSi and 4@SiOx-porSi - 70 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) generate formate as the major product under the standard conditions with FEs for formate of 54 and 41%, respectively (FIGs.71A-71C and FIG.72). Although significant amounts of CO are still produced in the immobilized systems (the FEs for CO are 37 and 27%, respectively), the trend is striking. As the applied potential is changed, the combination of 3@SiOx-porSi with PhOH, display a different dependence on selectivity compared to 3@SiOx-porSi with NEt3andiPrOH. From -1.75 V to -1.95 V, there is essentially no change in selectivity, but at - 2.15 V, CO becomes the major product with a FE of 53%. There is no increase in the FE to H2, which stays constant at around 17% from -1.75 V to -2.15 V. A common problem in using molecular catalysts that are immobilized on solid supports is catalyst leaching into solution, which in some cases is responsible for catalysis. As independently shown herein, leaching of 3 and 4 into solution is unlikely to be responsible for CO2reduction. Further, FTIR spectroscopy of 3@SiOx-porSi and 4@SiOx-porSi before and after CPE under standard conditions for seven hours showed that ~86% and ~91%, respectively of the Mn was still present on the surface after electrolysis (FIG.8A, FIGs.86A- 86H, FIG.87, FIGs.88A-88H, and FIGs.89-91). The presence of clear CO stretches in the FTIR spectrum is also consistent with complexes of the type (Rbpy)Mn(CO)3X (X = solvent or atom from support; see FIG.21A, FIGs.86A-86H, and FIG.87) still being present, which is further supported by XPS (FIG.21B). ICP-MS was used to more accurately quantify the amount of Mn attached to the surface after CPE (Table 2). A sample with a geometric surface area of 1.2 cm2contained 792 nmol of Mn on SiOx-porSi after CPE, while ICP-MS analysis of the reaction solution after CPE determined that there were 108 nmol of Mn in solution. Thus, ICP-MS indicates that 88% of the Mn was still present on the SiOx-porSi photocathode, with only 12% of the Mn in the solution for a sample of 3@SiOx-porSi after CPE. Cross- sectional SEM confirmed that the internal pore structures of 3@SiOx-porSi and 4@SiOx- porSi after CPE were similar to samples of 3@SiOx-porSi and 4@SiOx-porSi before CPE, showing that CO2 reduction did not impact the structure of the photocathode. Additionally, EDS elemental mapping spectra of 3@SiOx-porSi and 4@SiOx-porSi after CPE showed that the Mn and Si oxide are still relatively uniformly distributed in the pores. These results suggest that the immobilization of molecular catalysts on SiOx-porSi results in a relatively small amount of catalyst leaching or surface restructuring. Table 2. ICP-MS data used to calculate the amount of Mn in 3@SiOx-porSi, 4@SiOx-porSi after roughly 7 hours of CPEaMn Catalyst Measured conc. ( ^g / L) Dilution factor Total Mn nmol - 71 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) 3@SiOx-porSi After CPE62.248 20 792P t CPE l ti n The performance of 3@SiOx-porSi and 4@SiOx-porSi for photoelectrochemical CO2 reduction to formate makes them some of the best performing immobilized systems in terms of selectivity, current density per geometric area, TOF, and stability. For example, previous systems give FEs to formate which are lower than 85% and are completely inactive after seven hours. Another attractive feature of 3@SiOX-porSi is that reproducible catalytic results are obtained with very similar kinetics, FEs, and current densities achieved over three trials (FIGs.73A-73C). Without wishing to be bound by any theory, it has been proposed herein that the reproducibility is, in part, due to the ability to gauge the quality of the initial H-porSi wafer, the oxidized SiOx-porSi wafer, and the surface coverage of 3 using transmission FTIR spectroscopy. This degree of control is rare in the immobilization of molecular catalysts on solid supports. It is notable that immobilization of 3 or 4 on SiOx-porSi increases the selectivity for formate. There is significantly less CO formed from the immobilized system compared to the molecular catalyst in solution. Although at this stage it is unclear why the selectivity to formate increases when the catalyst is immobilized, one possible explanation is that the porous structure leads to variation in pH in different places along the pores, which could result in preferential formate production. There are also non-intuitive changes in selectivity of the immobilized systems as the applied potential of the photoelectrocatalytic system. For example, less formate and more CO is formed at higher potentials with PhOH. However, these trends typically mirror the behavior of the molecular catalyst in solution, so are likely not related to catalyst immobilization. The stability studies described herein indicate that although there is relatively little leaching (~10%) of 3 or 4 from SiOx-porSi, there is a decrease in geometric current density during photoelectrolysis (~33% over 7 hours for 3@SiOx-porSi). This is coupled both with a decrease in FE efficiency to formate and a decline in the overall FE from essentially 100% to 75% over 7 hours. Control experiments indicate that formate is stable under the reaction conditions and that it is not trapped in the pores of the SiOx-porSi support. Hence, an unknown process, is leading to a loss of catalytic performance by consuming electrons. - 72 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Nevertheless, the stability of 3@SiOX-porSi and 4@SiOX-porSi is higher than other immobilized molecular catalysts for formate production. It is also significant that there are differences in catalytic performance, in terms of initial FE to formate and stability, between immobilized 3 and 4. This indicates that further tuning of the molecular catalyst and attachment group may lead to systems with even greater FEs, current densities, and stability. Here, the lower loading of 4 compared to 3 is likely related to the lower solubility of 4 as there is some precipitation during immobilization. An interesting characteristic of the photoelectrocatalytic performance of 3@SiOx- porSi is that in solution 3 is unstable when it is illuminated with one sun irradiation, likely due to dimerization of a reduced form of the complex, as has been observed previously for related complexes. Given that 3@SiOX-porSi is still photoelectrochemically producing formate after seven hours, these results suggest that immobilization prevents this type of light-induced decomposition and that the molecular catalysts are not sufficiently close on the surface to dimerize. It may be the case that other molecular electrocatalysts which cannot be converted into photochemical systems in solution due to instability to light will be active photoelectro- or photocatalysts when immobilized on SiOx-porSi. This observation could widen the range of molecular catalysts available for photoelectro- or photocatalytic CO2 reduction. In terms of improving the system described herein, several control experiments indicate that both the light intensity and mass transport (evaluated through stirring speed) can impact the current density of 3@SiOx-porSi for CO2reduction. Likely the relatively moderate inherent activity of 3 and diffusion of species in and out of the pores are also impacting the current density. Another factor that influences catalytic performance is the length of time for which the H-porSi was oxidized. A sample which was oxidized for eighteen hours before the immobilization of 3, generated less formate with a lower FE (~60% after 45 minutes) than under the standard conditions (FIGs.76A-76D and FIGs.77A-77D). A significant amount of H2(~25% FE after 45 minutes) was also formed. This suggests that when coatings are applied to H-porSi for catalytic applications it will be crucial to carefully optimize thickness. Enumerated Embodiments The following exemplary embodiments are provided, the numbering of which is not to be construed as designating levels of importance: Embodiment 1 provides a heterogeneous catalyst composition comprising at least one transition metal catalyst complex covalently attached to a porous silicon substrate. - 73 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Embodiment 2 provides the heterogenous catalyst composition of Embodiment 1, wherein the transition metal catalyst complex is a photoelectrocatalyst. Embodiment 3 provides the heterogeneous catalyst composition of Embodiment 1 or 2, wherein each transition metal catalyst complex is covalently attached to a Si atom of the porous silicon substrate. Embodiment 4 provides the heterogenous catalyst composition of any one of Embodiments 1-3, wherein the porous silicon substrate is an oxidized porous silicon substrate, optionally wherein the oxidized porous silicon substrate is a thermally oxidized porous silicon substrate. Embodiment 5 provides the heterogenous catalyst composition of Embodiment 4, wherein the covalent attachment comprises a covalent bond formed by a reaction between a silatrane moiety of the transition metal catalyst complex and a surface Si-OH moiety of the oxidized porous silicon substrate. Embodiment 6 provides the heterogenous catalyst composition of any one of Embodiments 1-3, wherein the porous silicon substrate is a hydrogen-terminated porous silicon substrate. Embodiment 7 provides the heterogenous catalyst composition of Embodiment 6, wherein the covalent attachment comprises a covalent bond formed by a reaction between a non-aromatic, at least partially unsaturated moiety of the transition metal catalyst complex and a surface Si-H moiety of the hydrogen-terminated porous silicon substrate. Embodiment 8 provides the heterogeneous catalyst composition of any one of Embodiments 1-7, wherein the transition metal catalyst complex comprises a transition metal center selected from the group consisting of Pd, Cu, Ag, Au, Ni, Pt, Mn, Re, Co, Rh, Ir, Fe, Ru, and Os. Embodiment 9 provides the heterogeneous catalyst composition of any one of Embodiments 2-8, wherein the covalent attachment occurs via a linker, optionally wherein the linker comprises an optionally substituted C1-C12alkylenyl moiety. Embodiment 10 provides the heterogeneous catalyst composition of any one of Embodiments 1-9, which is a composition of formula (I): , wherein: Spis a porous silicon L is selected from the group consisting of a bond, optionally substituted C1-C6 alkylenyl, optionally substituted C1-C6heteroalkylenyl, optionally substituted C2-C6- 74 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) alkenylenyl, optionally substituted C2-C6 heteroalkenylenyl, optionally substituted C3-C8 cycloalkylenyl, optionally substituted C2-C8heterocycloalkylenyl, optionally substituted C6- C10 arylenyl, and optionally substituted C2-C10 heteroarylenyl; ; independently an L-type ligand, L1, L2, L3, L4, or L5groups can combine to form a bidentate L-type ligand; X1is an X-type ligand; and m is an integer is greater than 1; wherein one of L1, L2, L3, L4, L5, and X1, or a combination thereof, is covalently linked to Sp via L. Embodiment 11 provides the heterogeneous catalyst composition of Embodiment 10, wherein at least one of the following applies: (a) L1, L2, L3, L4, and L5are each independently selected from the group consisting of carbon monoxide (CO), P(RA)(RB)(RC), N(RA)(RB)(RC), O(RA)(RB), optionally substituted C2-C12 alkene, and optionally substituted C5-C12cycloalkene, and optionally substituted C2-C8heteroaryl, wherein RA, RB, and RCare each independently selected from the group consisting of H, optionally substituted C1-C6alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8 cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10aryl, and optionally substituted C2-C10 heteroaryl; and (b) two adjacent groups selected from L1, L2, L3, L4, and L5combine to form a bidentate phosphine ligand or bidentate amine. Embodiment 12 provides the heterogeneous catalyst composition of Embodiment 10 or 11, wherein at least one of the following applies: (a) L1, L2, L3, L4, and L5are each independently CO; and (b) two adjacent groups selected from L1, L2, L3, L4, and L5combine to form an optionally substituted bipyridine. Embodiment 13 provides the heterogeneous catalyst composition of any one of Embodiments 10-12, which is a composition selected from the group consisting of formula - 75 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) (Ia) and (Ib): , of optionally substituted C1-C6alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8 cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C2-C6alkenyl, optionally substituted C6-C10aryl, optionally substituted C2-C10heteroaryl, halogen, CN, NO2, ORD, N(RD)(RE), C(=O)RD, C(=O)N(RD)(RE), C(=O)ORD, N(RE)S(=O)2RD, and S(=O)2N(RD); RDand REare each independently selected from the group consisting of H, optionally substituted C1-C6alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10 heteroaryl; n is selected from the group consisting of 0, 1, 2, and 3; and o is selected from the group consisting of 0, 1, 2, 3, and 4. Embodiment 14 provides the heterogeneous catalyst composition of Embodiment 13, which is a composition selected from the group consisting of formula (Ia-1), (Ib-1), (Ia-2), and (Ib-2): , 2), - 76 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) wherein: Spois an oxidized porous silicon substrate, wherein bond L is a covalent bond formed by a reaction between a silatrane moiety of the transition metal catalyst complex and a surface Si-OH moiety of the oxidized porous silicon substrate; and Sphis a hydrogen-terminated porous silicon substrate, wherein bond L is a covalent bond formed by a reaction between a non-aromatic, at least partially unsaturated moiety of the transition metal catalyst complex and a surface Si-H moiety of the hydrogen-terminated porous silicon substrate. Embodiment 15 provides the heterogeneous catalyst composition of any one of Embodiments 10-14, wherein X1is selected from the group consisting of RF, halogen, OS(=O)2RF, OC(=O)RF, N(C(=O)RF)2, tetracoordinate boronate, hexacoordinate phosphorus, optionally substituted C6-C10aryl, and optionally substituted C2-C10heteroaryl, wherein RFis selected from the group consisting of H, optionally substituted C1-C6 alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10 heteroaryl. Embodiment 16 provides the heterogeneous catalyst composition of any one of Embodiments 10-15, wherein X1is selected from the group consisting of Cl and Br. Embodiment 17 provides the heterogeneous catalyst composition of any one of Embodiments 10-16, wherein n is 0 or 1. Embodiment 18 provides the heterogeneous catalyst composition of any one of Embodiments 10-17, wherein o is 0 or 1. Embodiment 19 provides the heterogeneous catalyst composition of any one of Embodiments 10-18, wherein R1aand R1b, if present, are each independently CH3. Embodiment 20 provides the heterogeneous catalyst composition of any one of Embodiments 10-19, wherein L is -(CH2)4-. Embodiment 21 provides the heterogeneous catalyst composition of any one of Embodiments 1-20, wherein the porous silicon substrate is selected from the group consisting of a wafer, bead, nanotube, and nanorod. Embodiment 22 provides the heterogeneous catalyst composition of any one of Embodiments 1-21, wherein the porous silicon substrate comprises p-type silicon or n-type silicon. Embodiment 23 provides the heterogeneous catalyst composition of any one of Embodiments 1-22, wherein the porous silicon substrate has a pore size ranging from about 2 - 77 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) nm to about 120 nm, optionally wherein the porous silicon substrate has a pore size of about 70 nm. Embodiment 24 provides the heterogeneous catalyst composition of any one of Embodiments 1-23, wherein the porous silicon substrate has a surface area of about 50 m2 / g to about 400 m2 / g, optionally wherein the porous silicon substrate has a surface area of about 200 m2 / g. Embodiment 25 provides a method for preparing the heterogeneous catalyst composition of any one of Embodiments 1-3 and 6-24, the method comprising: (a) contacting an etched, hydrogen-terminated porous silicon substrate and a transition metal catalyst complex to provide a mixture, wherein the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one non-aromatic, at least partially unsaturated moiety; and (b) heating the mixture to a temperature of about 160 °C. Embodiment 26 provides the method of Embodiment 25, wherein at least one of the following applies: (i) the transition metal catalyst has a concentration of about 0.4 mM during the contacting in (a); (ii) each contacting occurs in a solvent, optionally wherein the solvent is mesitylene; (iii) the contacting in (a) occurs for a period of about 18 h; and (iv) the contacting in (b) occurs for a period of about 4 hours. Embodiment 27 provides the method of Embodiment 25 or 26, wherein each contacting occurs in a solvent, optionally wherein the solvent is mesitylene. Embodiment 28 provides the method of any one of Embodiments 25-27, wherein the non-aromatic, at least partially unsaturated moiety is selected from the group consisting of optionally substituted C2-C6alkenyl, optionally substituted C2-C6alkynyl, optionally substituted C5-C10 cycloalkylenyl, optionally substituted C8-C12 cycloalkynylenyl, optionally substituted C2-C6heteroalkenyl, optionally substituted C2-C6heteroalkynyl, optionally substituted C5-C10 heterocycloalkylenyl, optionally substituted C8-C12 heterocycloalkynylenyl. Embodiment 29 provides the method of any one of Embodiments 25-27, wherein the transition metal catalyst complex comprises a bipyridyl moiety substituted with at least one - (CH2)0-12CH=CH2 moiety and is optionally further substituted with one or more substituents. - 78 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) Embodiment 30 provides a method for preparing the heterogeneous catalyst composition of any one of Embodiments 1-5 and 8-24, the method comprising: (a) contacting an at least partially surface oxidized porous silicon substrate and a transition metal catalyst complex to provide a mixture, wherein the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one silatrane moiety; and (b) heating the mixture to a temperature of about 80 °C. Embodiment 31 provides the method of Embodiment 30, wherein at least one of the following applies: (i) the transition metal catalyst has a concentration of about 0.2 mM during the contacting in (a); (ii) each contacting occurs in a solvent, optionally wherein the solvent is acetonitrile; and (iii) the contacting in (a) occurs for a period of about 18 h. Embodiment 32 provides the method of Embodiment 30 or 31, wherein the transition metal catalyst complex comprises at least one bipyridyl moiety substituted with at least one optionally further substituted with one or more substituents. 33 provides an electrode comprising the heterogeneous catalyst composition of any one of Embodiments 1-24. Embodiment 34 provides a method for reducing carbon dioxide (CO2), the method comprising: (a) contacting a medium comprising CO2with an electrochemical cell comprising an anode and a cathode, wherein the cathode comprises the heterogeneous catalyst composition of any one of Embodiments 1-24; (b) applying an electric potential between the anode and the cathode; and (c) irradiating the medium. Embodiment 35 provides the method of Embodiment 34, wherein the irradiation of the medium comprising irradiation with a solar simulator, optionally wherein the solar simulator comprises about 1 sun of illuminance (100 mW / cm2), and optionally wherein the wavelength ranges from about 380 nm to about 1000 nm. Embodiment 36 provides the method of Embodiment 34 or 35, wherein a potential of - 79 - 53095228.4 Attorney Docket No.047162-7475WO1(02446) about -1.75 V is applied. Embodiment 37 provides the method of any one of Embodiments 34-36, wherein at least one selected from the group consisting of formic acid (HC(=O)OH), formate (HC(=O)O-), carbon monoxide (CO), and hydrogen (H2) is produced. Embodiment 38 provides the method of any one of Embodiments 34-37, wherein the anode comprises graphite. Embodiment 39 provides the method of any one of Embodiments 34-38, wherein the medium comprises a liquid. Embodiment 40 provides the method of any one of Embodiments 34-39, wherein the medium further comprises at least one electrolyte. Embodiment 41 provides the method of any one of Embodiments 34-40, wherein the electrolyte is tetrabutylammonium hexafluorophosphate (TBAPF6). The terms and expressions employed herein are used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the embodiments of the present application. Thus, it should be understood that although the present application describes specific embodiments and optional features, modification and variation of the compositions, methods, and concepts herein disclosed may be resorted to by those of ordinary skill in the art, and that such modifications and variations are considered to be within the scope of embodiments of the present application. - 80 - 53095228.4

Claims

Attorney Docket No.047162-7475WO1(02446) CLAIMS What is claimed is:

1. A heterogeneous catalyst composition comprising at least one transition metal catalyst complex covalently attached to a porous silicon substrate.

2. The heterogenous catalyst composition of claim 1, wherein the transition metal catalyst complex is a photoelectrocatalyst.

3. The heterogeneous catalyst composition of claim 1 or 2, wherein each transition metal catalyst complex is covalently attached to a Si atom of the porous silicon substrate.

4. The heterogenous catalyst composition of any one of claims 1-3, wherein the porous silicon substrate is an oxidized porous silicon substrate, optionally wherein the oxidized porous silicon substrate is a thermally oxidized porous silicon substrate.

5. The heterogenous catalyst composition of claim 4, wherein the covalent attachment comprises a covalent bond formed by a reaction between a silatrane moiety of the transition metal catalyst complex and a surface Si-OH moiety of the oxidized porous silicon substrate.

6. The heterogenous catalyst composition of any one of claims 1-3, wherein the porous silicon substrate is a hydrogen-terminated porous silicon substrate.

7. The heterogenous catalyst composition of claim 6, wherein the covalent attachment comprises a covalent bond formed by a reaction between a non-aromatic, at least partially unsaturated moiety of the transition metal catalyst complex and a surface Si-H moiety of the hydrogen-terminated porous silicon substrate.

8. The heterogeneous catalyst composition of any one of claims 1-7, wherein the transition metal catalyst complex comprises a transition metal center selected from the group consisting of Pd, Cu, Ag, Au, Ni, Pt, Mn, Re, Co, Rh, Ir, Fe, Ru, and Os.

9. The heterogeneous catalyst composition of any one of claims 2-8, wherein the - 81 - 53095228.4Attorney Docket No.047162-7475WO1(02446) covalent attachment occurs via a linker, optionally wherein the linker comprises an optionally substituted C1-C12alkylenyl moiety.

10. The heterogeneous catalyst composition of any one of claims 1-9, which is a composition of formula (I): (I), wherein: Spis a porous siliconL is selected from the group consisting of a bond, optionally substituted C1-C6 alkylenyl, optionally substituted C1-C6heteroalkylenyl, optionally substituted C2-C6alkenylenyl, optionally substituted C2-C6 heteroalkenylenyl, optionally substituted C3-C8 cycloalkylenyl, optionally substituted C2-C8heterocycloalkylenyl, optionally substituted C6- C10 arylenyl, and optionally substituted C2-C10 heteroarylenyl; 15X L L1;independently an L-type ligand, wherein two adjacent L1, L2, L3, L4, or L5groups can combine to form a bidentate L-type ligand; X1is an X-type ligand; and m is an integer is greater than 1; wherein one of L1, L2, L3, L4, L5, and X1, or a combination thereof, is covalently linked to Sp via L.

11. The heterogeneous catalyst composition of claim 10, wherein at least one of the following applies: (a) L1, L2, L3, L4, and L5are each independently selected from the group consisting of carbon monoxide (CO), P(RA)(RB)(RC), N(RA)(RB)(RC), O(RA)(RB), optionally substituted C2-C12alkene, and optionally substituted C5-C12 cycloalkene, and optionally substituted C2-C8 heteroaryl, wherein RA, RB, and RCare each independently selected from the group consisting of H, optionally substituted C1-C6 alkyl, optionally substituted C1-C6 heteroalkyl, - 82 - 53095228.4Attorney Docket No.047162-7475WO1(02446) optionally substituted C3-C8 cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10aryl, and optionally substituted C2-C10 heteroaryl; and (b) two adjacent groups selected from L1, L2, L3, L4, and L5combine to form a bidentate phosphine ligand or bidentate amine.

12. The heterogeneous catalyst composition of claim 10 or 11, wherein at least one of the following applies: (a) L1, L2, L3, L4, and L5are each independently CO; and (b) two adjacent groups selected from L1, L2, L3, L4, and L5combine to form an optionally substituted bipyridine.

13. The heterogeneous catalyst composition of any one of claims 10-12, which is a composition selected from the group consisting of formula (Ia) and (Ib): (R1a) 1n, wherein:each occurrence of R1aand R1bis independently selected from the group consisting of optionally substituted C1-C6alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8 cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C2-C6alkenyl, optionally substituted C6-C10aryl, optionally substituted C2-C10heteroaryl, halogen, CN, NO2, ORD, N(RD)(RE), C(=O)RD, C(=O)N(RD)(RE), C(=O)ORD, N(RE)S(=O)2RD, and S(=O)2N(RD); RDand REare each independently selected from the group consisting of H, optionally substituted C1-C6alkyl, optionally substituted C1-C6heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8 heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10heteroaryl; n is selected from the group consisting of 0, 1, 2, and 3; and o is selected from the group consisting of 0, 1, 2, 3, and 4. - 83 - 53095228.4Attorney Docket No.047162-7475WO1(02446) 14. The heterogeneous catalyst composition of claim 13, which is a composition selected from the group consisting of formula (Ia-1), (Ib-1), (Ia-2), and (Ib-2): , 2),an porous wherein bond L is a covalent bond formed by a reaction between a silatrane moiety of the transition metal catalyst complex and a surface Si-OH moiety of the oxidized porous silicon substrate; and Sphis a hydrogen-terminated porous silicon substrate, wherein bond L is a covalent bond formed by a reaction between a non- aromatic, at least partially unsaturated moiety of the transition metal catalyst complex and a surface Si-H moiety of the hydrogen-terminated porous silicon substrate.

15. The heterogeneous catalyst composition of any one of claims 10-14, wherein X1is selected from the group consisting of RF, halogen, OS(=O)2RF, OC(=O)RF, N(C(=O)RF)2, tetracoordinate boronate, hexacoordinate phosphorus, optionally substituted C6-C10 aryl, and optionally substituted C2-C10heteroaryl, wherein RFis selected from the group consisting of H, optionally substituted C1-C6 alkyl, optionally substituted C1-C6 heteroalkyl, optionally substituted C3-C8cycloalkyl, optionally substituted C2-C8heterocycloalkyl, optionally substituted C6-C10 aryl, and optionally substituted C2-C10 heteroaryl.

16. The heterogeneous catalyst composition of any one of claims 10-15, wherein X1is selected from the group consisting of Cl and Br. - 84 - 53095228.4Attorney Docket No.047162-7475WO1(02446) 17. The heterogeneous catalyst composition of any one of claims 10-16, wherein n is 0 or 1.

18. The heterogeneous catalyst composition of any one of claims 10-17, wherein o is 0 or 1.

19. The heterogeneous catalyst composition of any one of claims 10-18, wherein R1aand R1b, if present, are each independently CH3.

20. The heterogeneous catalyst composition of any one of claims 10-19, wherein L is - (CH2)4-.

21. The heterogeneous catalyst composition of any one of claims 1-20, wherein the porous silicon substrate is selected from the group consisting of a wafer, bead, nanotube, and nanorod.

22. The heterogeneous catalyst composition of any one of claims 1-21, wherein the porous silicon substrate comprises p-type silicon or n-type silicon.

23. The heterogeneous catalyst composition of any one of claims 1-22, wherein the porous silicon substrate has a pore size ranging from about 2 nm to about 120 nm, optionally wherein the porous silicon substrate has a pore size of about 70 nm.

24. The heterogeneous catalyst composition of any one of claims 1-23, wherein the porous silicon substrate has a surface area of about 50 m2 / g to about 400 m2 / g, optionally wherein the porous silicon substrate has a surface area of about 200 m2 / g.

25. A method for preparing the heterogeneous catalyst composition of any one of claims 1-3 and 6-24, the method comprising: (a) contacting an etched, hydrogen-terminated porous silicon substrate and a transition metal catalyst complex to provide a mixture, wherein the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one non- - 85 - 53095228.4Attorney Docket No.047162-7475WO1(02446) aromatic, at least partially unsaturated moiety; and (b) heating the mixture to a temperature of about 160 °C.

26. The method of claim 25, wherein at least one of the following applies: (i) the transition metal catalyst has a concentration of about 0.4 mM during the contacting in (a); (ii) each contacting occurs in a solvent, optionally wherein the solvent is mesitylene; (iii) the contacting in (a) occurs for a period of about 18 h; and (iv) the contacting in (b) occurs for a period of about 4 hours.

27. The method of claim 25 or 26, wherein each contacting occurs in a solvent, optionally wherein the solvent is mesitylene.

28. The method of any one of claims 25-27, wherein the non-aromatic, at least partially unsaturated moiety is selected from the group consisting of optionally substituted C2-C6 alkenyl, optionally substituted C2-C6 alkynyl, optionally substituted C5-C10 cycloalkylenyl, optionally substituted C8-C12 cycloalkynylenyl, optionally substituted C2-C6 heteroalkenyl, optionally substituted C2-C6heteroalkynyl, optionally substituted C5-C10heterocycloalkylenyl, optionally substituted C8-C12 heterocycloalkynylenyl.

29. The method of any one of claims 25-27, wherein the transition metal catalyst complex comprises a bipyridyl moiety substituted with at least one -(CH2)0-12CH=CH2moiety and is optionally further substituted with one or more substituents.

30. A method for preparing the heterogeneous catalyst composition of any one of claims 1-5 and 8-24, the method comprising: (a) contacting an at least partially surface oxidized porous silicon substrate and a transition metal catalyst complex to provide a mixture, wherein the transition metal catalyst complex comprises at least one L-type or X-type ligand which is substituted with at least one silatrane moiety; and (b) heating the mixture to a temperature of about 80 °C. - 86 - 53095228.4Attorney Docket No.047162-7475WO1(02446) 31. The method of claim 30, wherein at least one of the following applies: (i) the transition metal catalyst has a concentration of about 0.2 mM during the contacting in (a); (ii) each contacting occurs in a solvent, optionally wherein the solvent is acetonitrile; and (iii) the contacting in (a) occurs for a period of about 18 h.

32. The method of claim 30 or 31, wherein the transition metal catalyst complex comprises at least one bipyridyl moiety substituted with at least is optionally further substituted with one or more substituents.

33. An electrode comprising the heterogeneous catalyst composition of any one of claims 1-24.

34. A method for reducing carbon dioxide (CO2), the method comprising: (a) contacting a medium comprising CO2 with an electrochemical cell comprising an anode and a cathode, wherein the cathode comprises the heterogeneous catalyst composition of any one of claims 1-24; (b) applying an electric potential between the anode and the cathode; and (c) irradiating the medium.

35. The method of claim 34, wherein the irradiation of the medium comprising irradiation with a solar simulator, optionally wherein the solar simulator comprises about 1 sun of illuminance (100 mW / cm2), and optionally wherein the wavelength ranges from about 380 nm to about 1000 nm.

36. The method of claim 34 or 35, wherein a potential of about -1.75 V is applied.

37. The method of any one of claims 34-36, wherein at least one selected from the group - 87 - 53095228.4Attorney Docket No.047162-7475WO1(02446) consisting of formic acid (HC(=O)OH), formate (HC(=O)O-), carbon monoxide (CO), and hydrogen (H2) is produced.

38. The method of any one of claims 34-37, wherein the anode comprises graphite.

39. The method of any one of claims 34-38, wherein the medium comprises a liquid.

40. The method of any one of claims 34-39, wherein the medium further comprises at least one electrolyte.

41. The method of any one of claims 34-40, wherein the electrolyte is tetrabutylammonium hexafluorophosphate (TBAPF6). - 88 - 53095228.4

Images