Electrolyte selection toward efficient photoelectrochemical glycerol oxidation on BiVO4

Glycerol, a primary by-product of biodiesel production, can be oxidized into various value-added chemicals, significantly enhancing the techno-economic value of photoelectrochemical (PEC) cells. Several studies have explored various photoelectrode materials and co-catalysts, but the influence of electrolytes on PEC glycerol oxidation has remained relatively unexplored despite its significance. Here, we explore the impact of various acidic (pH = 2) electrolytes, namely NaNO3, NaClO4, Na2SO4, K2SO4, and KPi, on PEC glycerol oxidation using nanoporous thin film BiVO4 as a model photoanode. Our experimental findings reveal that the choice of electrolyte anion and cation significantly affects the PEC performance (i.e., photocurrent, onset potential, stability, and selectivity towards value-added products) of BiVO4 for glycerol oxidation. To explain this interesting phenomenon, we correlate the observed performance trend with the ion specificity in the Hofmeister series as well as the buffering capacity of the electrolytes. Notably, NaNO3 is identified as the optimal electrolyte for PEC glycerol oxidation with BiVO4 when considering various factors such as stability and production rates for glycerol oxidation reaction (GOR) products, surpassing the previously favored Na2SO4. Glycolaldehyde emerges as the most dominant product with ∼50% selectivity in NaNO3. The general applicability of our findings is confirmed by similar observation in electrochemical (EC) GOR with a polycrystalline platinum anode. Overall, these results emphasize the critical role of electrolyte selection in enhancing the efficiency of EC/PEC glycerol oxidation.

The Raman shift scale was calibrated using the Si 520 cm -1 mode.The spectra have been recorded at about 22 °C, using a 2 mL optical grade-quartz cuvette.
 Figure S20.C-O stretching band region for the spectral references, 0.5 M glycerol (a) and 0.5 NaNO3 (b).The spectra have been recorded at about 22 °C, using a 2 mL optical grade-quartz cuvette, at a laser wavelength of 785 nm, and a total spectral power of 450 mW.

Tables for the Supporting Information
 Table S1.Water oxidation photocurrent values expressed in mA cm −2 at 1.23 VRHE, extracted from Figure S4.
 Table S3.The impact of adding 50 mM protons to the pH of the 0.5 M KPi + 0.1 M glycerol solution.
 Table S4.The impact of adding 50 mM protons to the pH of the 0.5 M K2SO4 + 0.1 M glycerol solution.
 Table S5.The impact of adding 50 mM protons to the pH of the 0.5 M Na2SO4 + 0.1 M glycerol solution.
 Table S6.The impact of adding 50 mM protons to the pH of the 0.5 M Na2SO4 + 0.1 M glycerol solution.
 Table S7.The impact of adding 50 mM protons to the pH of the 0.5 M NaNO3 + 0.1 M glycerol solution.

Sample preparation
First, 0.4 M of potassium iodide (Santa Cruz Biotechnology) was completely dissolved in 50 mL of deionized water (18.2MΩ), followed by adding 0.1 mL of nitric acid (>69.0%,Honeywell) and 0.04 M of bismuth nitrate pentahydrate (Acros Organics).The solution was stirred using a magnetic bar until the salts were fully dissolved.Second, a 20 mL ethanolic solution with 0.225 M of p-benzoquinone (Alfa Aesar) was prepared.Subsequently, the aqueous solution was slowly added to the ethanolic solution, resulting in a very dark red but clear solution.Next, BiOI nanosheet arrays were grown on FTO substrates with a sheet resistance of 7 Ω sq −1 (Sigma Aldrich) through electrodeposition.In the electrodeposition process, a platinum coiled wire with a diameter of 0.5 mm and an Ag/AgCl (saturated KCl) electrode (XR300, Radiometer Analytical) were employed as the counter electrode and the reference electrode, respectively.With the three electrodes immersed in the solution, a constant potential of −0.1 V vs. Ag/AgCl was applied until reaching a charge of 200 mC cm −2 , resulting in the formation of red-orange films.Subsequently, the BiOI films were coated with a 50 µL cm −2 solution of 0.2 M vanadyl acetylacetonate (Acros Organics) in dimethyl sulfoxide (VWR Life Science).The coated films were then annealed on a hot plate at 450 °C for 2 hours with a ramping rate of 5 K min −1 to induce the conversion into monoclinic BiVO4.Finally, excess V2O5 layers were removed by immersing the samples into a 1 M NaOH (Sigma Aldrich) solution for 15 minutes.

Electrochemical experiments
Electrochemical measurements were conducted using a potentiostat (VersaSTAT 3F, Princeton Applied Research).The same reference electrode and counter electrode used during the electrodeposition process were employed.An AM1.5G solar simulator (WACOM WXS-50S-5H Class AAA) with an irradiance of 100 mW cm −2 was employed as the light source.In the PEC measurements, the light was illuminated from the backside (i.e., the rear of the sample through the FTO-substrate side).The scan rate in LSV was set at 20 mV s −1 .
Chronoamperometry was performed at 1.23 VRHE under the same AM1.5Gillumination.The applied potential with respect to the Ag/AgCl (VAg/AgCl) reference electrode was converted to the VRHE scale using the following Nernst equation: where V 0 Ag/AgCl is the standard potential of the reference electrode (0.197 V).No iR correction was performed to present the data, considering the relatively small current (maximum current ~2 mA) and cell impedance (~10 Ω) during all electrochemical measurements.

Product analysis using high-performance liquid chromatography (HPLC)
Liquid samples, collected after 12 hours of photoelectrolysis at a constant potential of 1.23 VRHE in various electrolyte solutions, were analyzed using an HPLC system (UltiMate 3000, Thermo Scientific) for quantifying glycerol oxidation reaction (GOR) products.The system was equipped with a single column (HyperREZ XP H+, Thermo Scientific) and utilized both a wavelength-variable UV detector (UltiMate 3000, Thermo Scientific) and a refractive index (RI) detector (RefractoMax 520, Thermo Scientific).The flow rate was maintained at 0.5 mL min −1 , and the column temperature was held constant at 60°C.A 5 mM H2SO4 aqueous solution was employed as the mobile phase.
The following chemicals were used as reference GOR products: dihydroxyacetone (DHA, for synthesis, Sigma Aldrich), formic acid (FA, 98−100%, Sigma Aldrich), DL-glyceraldehyde (GLAD, >90%, Sigma Aldrich), glycolaldehyde dimer (GCAD, Sigma Aldrich), glycolic acid (GCA, 98%, Thermo Scientific), DL-glyceric acid (GA, ~2 M in water, Chem Cruz), and lactic acid (LA, 85%, Sigma Aldrich).For calibration, five aqueous solutions with varying concentrations (100−500 mM in increments of 100 mM for glycerol and 1−5 mM in increments of 1 mM for GOR products) were analyzed using the RI detector and the UV detector at 200 nm and 210 nm, as depicted in Figure S9−S11.We integrated the areas under their peaks to establish a linear relationship between the concentration and the peak area, also shown in where moli represents the amount of product i in moles, and moltotal represents the total amount of all products, also in moles.For example, since GCAD, GLAD, DHA, and FA were the only products produced in our case, moltotal was calculated using the following formula: Faradaic efficiency (FE) was calculated using the following formula: where Qtotal is the total charge passed during the photoelectrolysis, and QGOR represents the charge used to oxidize glycerol.QGOR can be calculated using the following formula: where qi represents the molar charge (C mol −1 ) that is used to produce product i (e.g., qGCAD).

Characterizations
X-ray diffraction (XRD) was conducted using an X-ray diffractometer (X'Pert, PANalytical).A Cu Kα radiation with a wavelength of 1.5406 Å was employed, and the incident angle of the X-ray was set to 2°.X-ray photoelectron spectroscopy (XPS) analyses were performed using a monochromatic Al Kα X-ray source (Focus 500, Specs), with a photon energy of 1486.84 eV, and an electron analyzer (Phoibos 100, Specs).Binding energy (BE) calibration was performed by referencing the peak position of Au 4f7/2 at 84.0 eV.Scanning electron microscopy (SEM) was carried out using a GeminiSEM 360 instrument (ZEISS).UV-Vis spectroscopy was performed using a Lambda 950 spectrophotometer (PerkinElmer).The electrical conductivity of the electrolyte solutions was measured using a Crison Basic 30 conductivity meter.Raman measurements were performed using a customized fiber-coupled system (Wasatch Photonics WP785ER), operating at NIR with a laser wavelength of 785 nm, and at a total spectral power tunable between 5 and 450 mW.The spectrometer has an average spectral resolution of about calibrated using the Si 520 cm -1 mode, while the total spectral power at the focal length (about 11 mm from the objective lens) was calibrated using a Si photodiode (Thorlabs PM16-121).
The spectra of the pristine liquid samples were recorded at about 22 °C, using a 2 mL optical grade-quartz cuvette.The temperature was measured in close proximity to the sample and logged throughout the measurements using a local temperature probe (Thorlabs TSP01-TH).
The integration time for all the measurements reported in this work was set to 2000 ms.To obtain a satisfactory statistics on the collected spectra, multiple spectra (about 25) were collected and averaged.NH3 solutions were prepared by diluting a 25% assay aqueous NH3 solution with deionized water at ratios of 1:1 (red curve) and 1:9 (blue curve); for instance, in the 1:9 dilution, 10 mL of ammonia solution was mixed with 90 mL of deionized water.Signals that peak at 6.9 min with onset of < 6.   NaNO3 (b).The spectra have been recorded at about 22 °C, using a 2 mL optical grade-quartz cuvette, at a laser wavelength of 785 nm, and a total spectral power of 450 mW.

Figure S1 .
Figure S1.SEM image of the nanoporous BiVO4 film used in our study.

Figure
Figure S2.(a) XRD pattern of the BiVO4 film and FTO substrate.(b) Tauc plot of the BiVO4

Figure S9 .
Figure S9.X-ray photoelectron spectroscopy (XPS) O 1s and V 2p core-level spectra of the (a)

Figure S10 .
Figure S10.Calibration data for glycerol used in high-performance liquid chromatography

Figure S11 .
Figure S11.Calibration data for dihydroxyacetone (DHA) used in high-performance liquid

Figure S13 .
Figure S13.Product analysis using high-performance liquid chromatography (HPLC) in the

Figure S14 .
Figure S14.Product analysis using high-performance liquid chromatography (HPLC) in the

Figure S15 .
Figure S15.Product analysis using high-performance liquid chromatography (HPLC) in the

Figure S16 .
Figure S16.Product analysis using high-performance liquid chromatography (HPLC) in the

Figure S17 .
Figure S17.Mass spectroscopy (MS) result obtained from PEC measurement with BiVO4

Figure S18 .
Figure S18.Chromatograms of aqueous ammonia (NH3) solutions and pH 2 NaNO3 solutions 5 min are attributed to NH3.The magenta curve represents the chromatogram of the pH 2 NaNO3 solution, and the olive curve represents the chromatogram of the NaNO3 solution, initially containing 0.5 M glycerol, obtained after the photoelectrochemical (PEC) chronoamperometry measurement conducted at 1.23 VRHE for 12 hours.If nitrate reduction reaction (NRR) occurs at the cathode during our experiments, signals of NH3 are therefore expected in the chromatogram of NaNO3 solutions after the PEC measurements.The fact that there is no feature observed until ~6.7 min (the onset of NaNO3 signal that peaks at ~7.2 min)suggests that NRR cannot be detected in our experiments.

Figure S19 .
Figure S19.Full range Raman scattering spectra taken on different liquid samples, at a laser

Figure S20 .
Figure S20.C-O stretching band region for the spectral references, 0.5 M glycerol (a) and 0.5