Alcohols Enrichment Enables Their Electrooxidation Coupled with H2 Production at High Current Density

Electrochemical alcohols oxidation offers a promising approach to produce industrial-relevant chemicals and facilitate coupled H 2 production. However, the corresponding current density is very low at moderate cell potential that substantially limits the overall productivity. Here, we report enrichment of alcohols in local environment over a cooperative catalyst of Au nanoparticles supported on cobalt oxyhydroxide nanosheets (Au/CoOOH), enabling alcohols electrooxidation at high current density. Specically, the current density of benzyl alcohol electrooxidation can reach 523 mA cm − 2 at potential of 1.5 V vs. RHE. Experimental and theoretical results suggest that benzyl alcohol molecules are enriched on Au/CoOOH interface via strong d-π interaction. The enrichment has a broad substrate scope that involves alcohols with α-π bond including α-phenyl, C = C and C = O groups. Based on these ndings, we design an intermittent potential (IP) strategy for long-term alcohol enrichment, achieving electrooxidation with current density of > 250 mA cm − 2 over 24 hours and promoted productivity. benzyl alcohol. The substrate can be applied to a wide range of alcohols with α-π bond including α-phenyl, C = C and C = O groups. Based on the nding that Au can be readily reduced at open circuit with enhanced alcohol adsorption ability, we design an intermittent potential (IP) strategy for alcohol electrooxidation, which achieves long-term alcohol enrichment and thus high current density and productivities, and an energy-saving process is realized. This work may open up a promising avenue for anodic reaction coupling H 2 production at high current density via engineering adsorption sites for substrate enrichment. measurements was performed at the beamline 1W1B of the Beijing Synchrotron Radiation Facility (BSRF), Institute of High Energy Physics (IHEP), Chinese Academy of Sciences (CAS). Extended X-ray absorption ne structure spectra (EXAFS) were recorded at ambient temperature in transmission mode. The typical energy of the storage ring was 2.5 GeV with a maximum current of 250 mA; the Si (111) double crystal monochromator was used. Fourier transform of the EXAFS spectra were carried out in a K-range from 3.0 to 12.8 Å −1 . The in-situ FTIR spectra of benzyl alcohol were carried out in a Bruker Equinox 55 spectrometer, between 4000 and 400 cm −1 with a resolution of 4 cm −1 after 600 scans per spectrum. About 30 mg of the sample was pressed into a wafer with a diameter of 13 mm, which was then installed in an in-situ IR reactor with CaF 2 windows. The sample was pre-processing by He gas at r.t. for 1 h and then benzyl alcohol was owed into the cell for 30 min, then physically adsorbed benzyl alcohol was removed by owing He gas for 30 min. The FTIR spectra were in-situ collected during the He purging process 38 .


Introduction
Hydrogen (H 2 ) is a promising energy to replace fossil fuels that addresses the environmental problems associated with global warming and alleviates the energy crisis. Electrocatalytic water splitting powered by clean energy (e.g., solar energy, wind) represents a green approach to produce H 2 1,2 . However, this process still suffers from large overall electricity consumption stemming from high cell potential due to the sluggish four-electron transfer of anodic oxygen evolution reaction (OER) 3,4 . Developing anodic reactions with cell potential lower than OER could be a promising strategy for fundamentally lowering the energy requirements for electrocatalytic H 2 production. Such anode oxidation reactions have additional bene t of producing high value-added chemicals from inexpensive industrial byproducts or renewable biomass carbon sources 5,6 . For reactions of this kind, large positive potential unavoidably results in uncontrolled OER, therefore applying moderate potential is necessary to achieve high Faradaic e ciency (FE) toward target products 7 . Tremendous efforts have been devoted to developing electrocatalysts with improved catalytic activities [8][9][10][11][12][13] . Despite these efforts, the reported current density remains very low under moderate cell potential, such as the electrooxidation of alcohols 8,9 , 5-hydroxymethyfurfural (HMF) 10,11 , primary amines 12 and tetrahydroisoquinolines 13 are often operated at current density lower than 200 mA cm − 2 . The production of relevant target chemicals coupled with H 2 evolution at low current density would hamper the overall e ciency for industrial production.
Among the value-added anodic reactions, alcohol oxidation reactions (AORs) are particularly important for their wide applications in commodity chemicals production 14 . For instance, oxidation of benzyl alcohol produces benzoic acid, which is an important ne chemical used in synthetic ber, resin, and antiseptic industries 15 . Oxidation of polyols (such as glycerol) afford corresponding ketones/aldehydes and formate, which can nd applications as degradable plastic monomer, dyestuffs and food additives [16][17][18] . However, their large-scale productions using electrochemical approach have yet to be realized because of the insu cient productivity derived from the low current density. One of the limitations might be the low alcohol concentration in the vicinity of the catalytically active sites at anode that hinders reaction occurring (Scheme 1a; Typically, water is used as the electrolyte for good alcohol solubility). In line with this rationale, enriching alcohol molecules at the anode via electrocatalyst surface engineering would possibly promote alcohol oxidation productivity, but this strategy has rarely been considered.
For electrocatalytic reaction of gaseous molecules, for instance, CO 2 , N 2 and CH 4 , it has been demonstrated that constructing a hydrophobic surface on the electrode is bene cial to their enrichment and therefore facilitate the reactivity [19][20][21] . However, this strategy may not be applicable for electrooxidation of alcohols often featuring hydrophilic property. Very recently, Sargent et al. reported porphyrin-based metallic complexes on metal electrode as adsorption sites for CO, a key intermediate during CO 2 reduction, thus increasing CO concentration on the electrode and achieving ethanol generation with high rate and selectivity 19 . Inspired by the catalyst design principle, we hypothesize that constructing adsorption site for alcohol on the AOR-active anode would increase local alcohol concentration and thereby promote the current density of AOR coupled with H 2 generation.
Herein, we report the enrichment of alcohols in local environment by synthesizing a cooperative catalyst of Au nanoparticles supported on cobalt oxyhydroxide nanosheets (Au/CoOOH), which greatly enhances the current density in electrocatalytic oxidation of alcohols coupled with H 2 production (Scheme 1b). In-situ infrared spectroscopy combined with density functional theory (DFT) calculations and ab initio molecular dynamics (AIMD) simulations demonstrate that benzyl alcohol molecules can be strongly adsorbed on the Au/CoOOH interface via d-π interaction between the d orbital of Au and π* orbital of benzyl alcohol, together with the hydrogen bonding between the surface terminated O of CoOOH and hydroxyl group of benzyl alcohol, leading to the enrichment of benzyl alcohol in local environment. The Au/CoOOH catalyst is also e cient to enrich diverse alcohols with α-π bond consisting of α-phenyl, C = C and C = O groups with 2 ~ 20-fold higher electrooxidation rate than CoOOH. We found that while Au gradually gets oxidized at anodic potential, it can readily undergo reduction at open circuit with revival of the enrichment ability. Based on these ndings, by deploying an intermittent potential (IP) strategy between anodic potential and open circuit, we achieve long-term alcohol enrichment, maintaining high current density (> 250 mA cm − 2 at 1.35 V vs. RHE) for benzyl alcohol electrooxidation over 24 h. As a result, the productivities of oxidation products (benzaldehyde and benzoic acid) and H 2 achieved 10-and 9-fold increment compared with constant potential (CP) strategy, together with electric energy saving of 43 and 33% at 200 and 300 mA cm − 2 , respectively.

Page 4/23
Catalyst synthesis and characterizations. The cooperative catalyst of Au nanoparticles supported on CoOOH nanosheets (Au/CoOOH) was prepared via a two-step electrochemical method, in which Co(OH) 2 was initially synthesized on nickel (Ni) foam, then Au nanoparticles were electrodeposited onto Co(OH) 2 . The as-prepared Au/Co(OH) 2 was electro-oxidized in 1 M KOH solution to structurally transform to Au/CoOOH. For catalytic comparison, CoOOH nanosheets and Au nanoparticles on Ni foam were also synthesized following similar methods. Scanning electron microscope (SEM) images ( Fig. 1a and Supplementary Fig. 1) show that the as-synthesized Au/CoOOH nanosheets (with an average thickness of ~10 nm and diameter of 200-300 nm) are vertically grown on Ni foam and form array structure. High-resolution transmission electron microscope (HRTEM) images reveal that Au nanoparticles are well dispersed on the CoOOH nanosheets ( Fig. 1b and supplementary Fig. 2) with average diameter of 4.3 nm (Fig. 1c). The HRTEM image of an individual Au nanoparticle displays fringe distances of 0.203 nm ( Fig. 1b;  Chronoamperometric (CA) measurement was then carried out. As shown in Fig. 2c, the current-time (I-t) curve of Au/CoOOH shows a high initial current density of ~330 mA cm −2 at 1.3 V vs. RHE in 1 M KOH with 0.1 M benzyl alcohol at r.t., much higher than that of CoOOH (with an initial current density of ~20 mA cm −2 ). Although the Au catalyst exhibits an initial current density of 152 mA cm −2 , it decays rapidly to lower than 6 mA cm −2 within 600 s. In addition, the current densities are very weak in the absence of benzyl alcohol ( Supplementary Fig. 4), indicating that the high current density over Au/CoOOH is caused by the oxidation of the benzyl alcohol, rather than the reduction of Co, OH − adsorption or other effects on the surface.   Fig. 7). The calculated adsorption energy of benzyl alcohol is much lower on Au/CoOOH (Fig. 3c), suggesting that the adsorption of benzyl alcohol on Au/CoOOH is thermodynamically more preferable. By analyzing the density of states and Hirshfeld charge of Au/CoOOH, together with the frontier orbitals of benzyl alcohol, it can be deduced that the electron in occupied Au 5d orbital can transfer to the unoccupied π* orbital of benzyl alcohol (Fig. 3d, Supplementary Fig. 8a and Supplementary Note 1) 43 . This d-π interaction endows Au/CoOOH with stronger adsorption ability to benzyl alcohol than CoOOH. Meanwhile, the interaction between hydroxyl group of benzyl alcohol and terminated-O on CoOOH via hydrogen bonding also contributes to the enhanced adsorption energy (Supplementary Fig. 8b).
To consider the enrichment of benzyl alcohol on Au/CoOOH in real reaction system, Au/CoOOH for benzyl alcohol enrichment in real reaction system, and the interface between Au and CoOOH is the preferential adsorption site.
To understand the reaction mechanism of benzyl alcohol oxidation, Gibbs free energy diagrams of benzyl alcohol oxidation to benzoic acid were calculated over different catalysts ( Fig. 4a; Supplementary Figs. 9-11, Table 3 and Note 2). The reaction begins with benzyl alcohol adsorption, then it is oxidized to benzaldehyde and finally to benzoic acid. compared with CoOOH, together with higher production rates of alcohol-oxidized products (entries 1-6, Table 1; and Supplementary Fig. 13) and comparably high FE towards them. Among these alcohol molecules, benzyl alcohol substituted with −C(CH 3 ) 3 shows inferior activity (entry 6, Table 1), which may be due to the steric hindrance that inhibits benzene ring adsorption on Au/CoOOH. We further investigated the effect of phenyl and hydroxyl groups and their proximity on the current density. As shown in Supplementary Fig. 14, there is no current density enhancement for alcohols without phenyl group (ethanol and cyclohexanemethanol) or aromatics without hydroxyl group (toluene), suggesting the importance of hydroxyl and phenyl groups in benzyl alcohol for current density enhancement over Au/CoOOH. Moreover, we found that the Au/CoOOH shows very weak current density enhancement for alcohol without α-phenyl group (β-phenethyl alcohol; entry 7, Table 1), which is attributed to the relatively weak adsorption of β-phenethyl alcohol on Au/CoOOH demonstrated by DFT calculation (Supplementary Fig. 15).
To broaden the substrate scope, we explored the alcohols with α-C=C (methallyl alcohol) and α-C=O group (hydroxyacetone), because these groups contain π bond that resembles benzene ring. To our delight, Au/CoOOH catalyst exhibits higher current density and product yields for electrooxidation of these alcohols compared with CoOOH (entries 8 and 9, Table 1; and Supplementary Fig. 16). This can be explained by the calculated negative adsorption energies for their adsorption on Au/CoOOH, which is due to the strong interaction between α-C=C/-C=O groups and Au via Au(d)-π interaction (see Supplementary Fig. 17 and Supplementary Note 3). Furthermore, we found that the Au/CoOOH catalyst shows enhanced current density as well as higher product yields for electrooxidation of polyols with vicinal hydroxyl groups, including ethylene glycol, 1,2propanediol and glycerol (entries 10-12, Table 1; and Supplementary Fig. 18), among which glycerol is an important biomass derivative produced in large amount in biodiesel production. The enhancement can be explained by the oxidation of -CH 2 −OH to -C=O under the reaction conditions 16,17 , which leads to a similar enrichment effect as in the case of alcohol with α-C=O group.
[Please see the supplementary files section to view table 1.] Understanding of the decay and revival of current density. We found that the current density decays gradually during the electrooxidation of benzyl alcohol (Fig. 5a), indicating a deactivation process may be involved. However, the initial high current density can be restored after potential cut-off for approximately 100 s at open circuit. To unveil the underlying mechanism for the decay and revival of current density, operando extended Xray absorption fine structure spectroscopy was conducted (operando EXAFS). Fig. 5b shows the Au L 3 -edge XANES spectra of Au/CoOOH when anodic potential and open circuit were alternatively applied in 1 M KOH with 0.1 M benzyl alcohol. These operations are denoted as potential ON and OFF, respectively. When potential ON (1.4 V vs. RHE), the white line intensity enhances gradually (from line_a to line_d), which is assigned to Au oxidation possibly by the electrolyte, as verified by operando Au L 3 -edge XANES spectra ( Supplementary Fig. 19). This process is accompanied by the gradual decay of current density, suggesting that Au oxidation may has a negative effect on alcohol enrichment. This was confirmed by pre-oxidizing Au/CoOOH prior to introducing benzyl alcohol into the electrolyte, which shows very low catalytic activity (Supplementary Fig. 20). Subsequently, potential OFF was exerted and it causes weakening of white line intensity (from line_d to line_f), indicating that Au is reduced at open circuit. It should be noted that Au cannot be reduced in 1 M KOH electrolyte alone ( Supplementary Fig. 19), suggesting that Au is more likely reduced by benzyl alcohol. Indeed, alcohols can be used as reducing agent. 44,45 This process is accompanied by rapid revival of current density, suggesting that the reduced Au can again adsorb and enrich benzyl alcohol molecules. The Au oxidation/reduction processes was demonstrated reversible when further anodic potential (from line_f to line_i) and open circuit (from line_i to line_k) were applied (Fig. 5b), suggesting that we could readily regain the high current density by exerting potential OFF. It should be mentioned that Co species in Au/CoOOH is oxidized more rapidly compared with current density decay ( Supplementary Fig. 21), suggesting that Co oxidation may not be the main reason for the current decay.
To investigate the structure of Au species in Au/CoOOH when it is oxidized or reduced, Fourier-transform Au L 3 -edge EXAFS spectra was performed. As shown in Fig. 5c Fig. 5d, the current density decays instantly when methionine was introduced to the electrolyte, which is attributed to the strong interaction between methionine and Au via Au-S bond that prevents benzyl alcohol from adsorption. The above results indicate that when Au gradually gets oxidized to form Au-O(H) species at anodic potential with attenuation of alcohol enrichment, it can reversibly undergo reduction at open circuit with revival of the enrichment ability (Fig. 5e).
Intermittent potential strategy. In light of the findings that the alcohol enrichment could be reversible, we sought to develop an intermittent potential (IP) strategy to maintain the high current density in long-term electrooxidation of benzyl alcohol over Au/CoOOH, via alternatively switching between anodic potential (potential ON) and open circuit (potential OFF). Specifically, the reaction was initially conducted at 1.35 V vs. RHE in 1 M KOH with 0.1 M benzyl alcohol to get high current density (~400 mA cm −2 ). When the current density decays to <250 mA cm −2 , we cut off the potential for about 100 s to regain the benzyl alcohol enrichment. Then we switched the potential back to 1.35 V vs. RHE with the recovery of initial high current density. By repeating the above operations, the current density can be maintained at high level (250-400 mA cm −2 ) over 24 h (Fig. 6a). In contrast, the current density of Au/CoOOH decays from 400 to 60 mA cm −2 within 2 h using traditional constant potential (CP) strategy. As a result, the Au/CoOOH with IP strategy exhibits larger benzyl alcohol consumption, higher productivities of benzaldehyde, benzoic acid and H 2 , and comparably high FE of total anodic products (Fig. 6b). The Au/CoOOH maintains its original structure after the reaction ( Supplementary Fig. 22), demonstrating its stability. The electric energy consumption using IP and CP strategies were also compared. The Au/CoOOH with the IP strategy requires much lower potential to reach the same current density (Fig. 6c), together with 33-43 % energy saving (Fig. 6d).

Discussion
In summary, we achieve the enrichment of alcohols in local environment to enable their electrooxidation coupled with H 2 production at high current density by using a cooperative Au/CoOOH catalyst.
Experimental studies, DFT calculations and AIMD simulations together reveal that Au/CoOOH interface has a strong enrichment ability for benzyl alcohol that derives from d-π interaction between the d orbital of Au and π* orbital of benzyl alcohol, together with hydrogen bonding between CoOOH and hydroxyl group of benzyl alcohol. The substrate can be applied to a wide range of alcohols with α-π bond including α-phenyl, C = C and C = O groups. Based on the nding that Au can be readily reduced at open circuit with enhanced alcohol adsorption ability, we design an intermittent potential (IP) strategy for alcohol electrooxidation, which achieves long-term alcohol enrichment and thus high current density and productivities, and an energy-saving process is realized. This work may open up a promising avenue for anodic reaction coupling H 2 production at high current density via engineering adsorption sites for substrate enrichment.

General information
Except noted, all chemicals were purchased and used without further purification. Page 10/23 Catalyst preparation The Au nanoparticles supported on CoOOH nanosheet catalyst (Au/CoOOH) was prepared via a two-step electrochemical method, in which Co(OH) 2 nanosheets was initially grown on nickel (Ni) foam, then Au nanoparticles were electrodeposited onto Co(OH) 2 .
Subsequently, the as-synthesized Au/Co(OH) 2 was electro-oxidized in 1 M KOH solution to enable the structural transformation to Au/CoOOH.
Preparation of Co(OH) 2 nanosheets: The nickel foam (10 × 20 × 1.5 mm) was used as matrix for growing Co(OH) 2 nanosheets array. Initially, the nickel foam (10 × 20 × 1.5 mm) was sequentially washed with dilute HCl Preparation of Au/Co(OH) 2 : The electrochemical deposition of Au nanoparticles onto Co(OH) 2 was carried out in a three-electrode con guration as above described. Speci cally, Au nanoparticles were deposited on Co(OH) 2 by stepping the potential to −0.6 V vs. SCE for 10 s, followed by stepping back to −0.2 V vs. SCE for 10 s for three cycles, using aqueous electrolyte with 0.1 M NaCl and 5 mM HAuCl 4 47 . The pure Au catalyst was prepared via the similar electrodeposition method by directly using Ni foam as the working electrode.
Preparation of Au/CoOOH: The Au/CoOOH was obtained from the as-prepared Au/Co(OH) 2 via a simple cyclic voltammetry (CV) method in a three-electrode con guration, using Ag/AgCl (with saturated KCl) and Pt foil as the reference and counter electrodes respectively. The electrochemical oxidization process was performed at a scan rate of 100 mV s −1 from 0 V to 0.8 V vs. Ag/AgCl for approximately 20 cycles in 1 M KOH solution 46 . The pure CoOOH catalyst was prepared via the similar electrochemical oxidization method by using as-obtained Co(OH) 2 as the working electrode.
Characterizations X-ray diffraction patterns were collected on a Shimadzu XRD-6000 diffractometer using a Cu Kα source, with a scan range of 3−70° and scan step of 5° min −1 . X-ray photoelectron spectra (XPS) were performed on a Thermo VG ESCALAB 250 X-ray photoelectron spectrometer at a pressure of about 2×10 −9 Pa using Al Kα X-rays as the excitation source. Scanning electrode microscope (SEM) imaging was performed using a Zeiss SUPRA with energy dispersive X-ray spectroscopy (EDX) for the determination of metal composition. Metal contents in catalysts were determined by ICP-AES on a Thermo ICAP6300 Radial. The operando Au and Co XAFS measurements was performed at the beamline 1W1B of the Beijing Synchrotron Radiation Facility (BSRF), Institute of High Energy Physics (IHEP), Chinese Academy of Sciences (CAS). Extended X-ray absorption ne structure spectra (EXAFS) were recorded at ambient temperature in transmission mode. The typical energy of the storage ring was 2.5 GeV with a maximum current of 250 mA; the Si (111) double crystal monochromator was used. Fourier transform of the EXAFS spectra were carried out in a K-range from 3.0 to 12.8 Å −1 . The in-situ FTIR spectra of benzyl alcohol were carried out in a Bruker Equinox 55 spectrometer, between 4000 and 400 cm −1 with a resolution of 4 cm −1 after 600 scans per spectrum. About 30 mg of the sample was pressed into a wafer with a diameter of 13 mm, which was then installed in an in-situ IR reactor with CaF 2 windows. The sample was pre-processing by He gas at r.t. for 1 h and then benzyl alcohol was owed into the cell for 30 min, then physically adsorbed benzyl alcohol was removed by owing He gas for 30 min. The FTIR spectra were insitu collected during the He purging process 38 .

Electrochemical measurement
All electrochemical measurements for alcohols oxidation were performed in 1 M KOH electrolyte at r.t. or 60 ºC on an electrochemical workstation (CHI 760E, CH Instruments, Inc.). The electrochemical tests were performed in a three-electrode system, using Ag/AgCl electrode (with saturated KCl) and Pt foil as reference and counter electrode, respectively.
Linear scan voltammetry (LSV) curves of catalysts were acquired from −0.2 V to 0.6 V vs.
Ag/AgCl at a scan rate of 10 mV s −1 . All potentials measured against Ag/AgCl were converted to the reversible hydrogen electrode (RHE) scale using: . The detailed energy balance calculations were also conducted to analyse the consumption of electric energy. The value of electric energy was determined using: . The space-time yield of H 2 in Fig. 2d was obtained by gascollecting method of drainage water. For long-term electrochemical measurement by intermittent potential (IP) and constant potential (CP) strategies, the productivity of H 2 (in Fig. 6b)  The model of bulk γ-CoOOH was built according to the experimental X-ray diffraction pattern 48 . The space group of bulk γ-CoOOH was R mH, with the lattice parameters of a = b = 2.82 Å, c = 20.65 Å, α = β = 90º, γ = 120º. After that, the model of CoOOH was built by cleaving the (001) facet of bulk γ-CoOOH since the (001) facet was the preferably exposed facet according to the X-ray diffraction pattern. The model CoOOH contained one layer of were performed with the Cambridge Sequential Total Energy Package (CASTEP) 52 . The ionic cores were described by the ultrasoft pseudopotentials to decrease the number of plane waves required in the expansion of the Kohn-Sham orbitals 53 . The cutoff energy was set as 380 eV. The Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm was used to search the potential energy surface in geometry optimization 54 . The k-point meshes were set as 3 × 3 × 1. During the geometry optimization, three convergence criteria were used as follows: (1) energy tolerance of 1.0 × 10 −5 eV per atom, (2) force tolerance of 3.0 × 10 −2 eV/Å, (3) displacement tolerance of 1.0 × 10 −3 Å.
The ab initio molecular dynamics (AIMD) simulations were performed in isothermalisobaric (NPT) ensemble, with a temperature of 298.15 K and a pressure of 0.1 MPa. The temperature and pressure were controlled by the Nose method 55  The electronic structures of benzyl alcohol, methallyl alcohol, and hydroxyacetone were calculated at the B3LYP level 57 in the Gaussian 09 program 58 .

Data Availability
The data that support the plots within the manuscript and other findings of this study are available from the corresponding author upon reasonable request.  Reaction mechanism of benzyl alcohol oxidation. a, Gibbs free energy diagrams of benzyl alcohol oxidation to benzoic acid on CoOOH, Au, and Au/CoOOH. b, ΔG of RDS for benzyl oxidation on Au/CoOOH with one and ve benzyl alcohol molecules adsorption, respectively. Inset displays the associated optimized geometries.