摘要：氫氣因其高能量密度、可持續(xù)性和燃燒后無污染等優(yōu)點，被認(rèn)為是取代傳統(tǒng)化石燃料的最具前途的新興能源載體之一。其中，電解水制氫技術(shù)因為其高效和綠色的特性而備受關(guān)注。然而電解水制氫過程通常受到陽極析氧反應(yīng)（OxygenEvolution Reaction，OER）的限制，因此這種方法的大規(guī)模應(yīng)用面臨重大挑戰(zhàn)?？朔@一難題的一個有前途的解決方法是在陽極上使用電催化甘油氧化反應(yīng)（Glycerol Oxidation Reaction，GOR）代替OER，這種替代反應(yīng)可以實現(xiàn)節(jié)能降耗的同時提高電解水制氫的效率，進(jìn)一步推動氫氣作為清潔能源的發(fā)展。然而，這一目標(biāo)的實現(xiàn)需要高效、低成本且高選擇性的GOR電催化劑。在這篇文章中，我們報告了一種新型的酸堿雙電解質(zhì)流電解器（AADEF-electrolyzer），用于在堿性陽極GOR耦合酸性陰極析氫反應(yīng)（Hydrogen Evolution Reaction，HER）。我們通過一種簡單的水熱煅燒方法制備了一種在鎳泡沫（NF）上原位生長的自支撐的NiCo2O4納米針電極材料（NiCo2O4/NF）。該電極在GOR中表現(xiàn)出優(yōu)異的電催化性能，在低電位下實現(xiàn)了高的電解電流密度，對甲酸鹽的生產(chǎn)表現(xiàn)出優(yōu)異的選擇性，法拉第效率超過85%。密度泛函理論計算表明，NiCo2O4對GOR具有較低的反應(yīng)能壘，Ni的存在有利于降低Co的電子態(tài)密度，從而實現(xiàn)NiCo2O4與中間體的高效解離，促進(jìn)甲酸的生成?；贜iCo2O4/NF出色的GOR性能和電化學(xué)中和能（ENE）理論，我們構(gòu)建了一個新型的AADEF-electrolyzer，利用NiCo2O4/NF作為GOR的陽極，配合酸性陰極進(jìn)行析氫反應(yīng)（HER）。實驗結(jié)果表明，AADEFelectrolyzer對GOR具有低過電位和高選擇性產(chǎn)甲酸的優(yōu)異性能，僅需0.36 V的電壓即可實現(xiàn)10 mA?cm?2的電流密度，平均產(chǎn)甲酸的法拉第效率為85%。同時該電解槽表現(xiàn)出良好的長期穩(wěn)定性和輔助產(chǎn)氫性能，陰極產(chǎn)氫的法拉第效率接近100%。這種低成本、易于制備的自支撐電極材料和新型酸堿雙電解質(zhì)流動電解器為促進(jìn)化學(xué)品的增值轉(zhuǎn)化和開發(fā)新型混合電解系統(tǒng)或其他相關(guān)電化學(xué)反應(yīng)的混合電解裝置提供了創(chuàng)新策略。

關(guān)鍵詞：制氫；甘油氧化；電催化；電解器

中圖分類號：O646

Hydrogen Generation Coupling with High-Selectivity Electrocatalytic Glycerol Valorization into Formate in an Acid-Alkali Dual-Electrolyte Flow Electrolyzer

Abstract： Owing to its high energy density， sustainability， and pollutionfreecombustion， hydrogen is considered one of the most promisingemerging energy carriers to replace conventional fossil fuels. Among thevarious hydrogen production technologies， electrolytic water splitting hasgained significant attention thanks to its high efficiency and environmentallyfriendly characteristics. However， the large-scale application of electrolyticwater splitting is often hindered by the limitations imposed by the anodicoxygen evolution reaction （OER）. To overcome this challenge， a promisingalternative approach is to replace the OER with the electrocatalytic glyceroloxidation reaction （GOR） at the anode. This substitution can lead to energysavings and enhanced efficiency of electrolytic water splitting for hydrogen production， thereby further promoting thedevelopment of hydrogen as a clean energy source. However， the application of the GOR at anode requires efficient， costeffective，and highly selective electrocatalysts. To this end， we report the development of a novel acid-alkaline dualelectrolyteflow electrolyzer （AADEF-electrolyzer） by coupling the GOR at the alkaline anode with the hydrogen evolutionreaction （HER） at the acidic cathode. A self-supported NiCo2O4 nanoneedle electrode material （NiCo2O4/NF） has been insitu grown on nickel foam （NF） using a simple hydrothermal-calcination method. The electrode demonstrates excellentelectrocatalytic performance for the GOR， achieving high electrolysis current density at low potentials and exhibiting highselectivity for formate production， with the Faraday efficiency exceeding 85%. Density functional theory （DFT） calculationsimply that NiCo2O4 has a lower energy barrier for the reaction and that the presence of Ni facilitates the reduction of theCo state density， thereby promoting the GOR. An innovative AADEF-electrolyzer was constructed by utilizing NiCo2O4/NFas the anode for the GOR and an acidic cathode for the HER. Experimental results indicate that the AADEF-electrolyzerexhibits excellent GOR performance with a low overpotential and high selectivity toward formate production. It requires avoltage of only 0.36 V to achieve a current density of 10 mA·cm?2 and long-term stability with a Faraday efficiency close to100% for hydrogen production. The low-cost and easily fabricated self-supported electrode material， together with theacid–alkaline dual-electrolyte flow electrolyzer， provide an innovative strategy for developing hybrid electrolysis systems.

Key Words： Hydrogen production; Glycerol oxidation; Electrocatalysis; Electrolyzer

1 Introduction

The global energy shortage and climate crisis， caused mainlyby the overuse of fossil fuels， have motivated an extensivepursuit of alternative and sustainable energy sources as well asnew energy storage and conversion systems. With its remarkableenergy density， renewability， and zero-emission combustionproperties， hydrogen has been considered as a highly promisingenergy carrier to replace fossil fuels 1–3. Nonetheless， hydrogenproduction still mainly relies on fossil fuel-derived gasreforming processes， which consume substantial amounts ofenergy and emit large quantities of carbon dioxide into theatmosphere 4–6. Despite the significant advantages ofelectrochemical hydrogen production via water splitting， whichis recognized as a highly efficient and green technology， thelarge-scale application of this approach faces major challengesdue to the costly nature of electrocatalysts and the slow kineticsof the anodic oxygen evolution reaction （OER） 7–11.

Recent studies have indicated that using organic molecularoxidation in place of OER in the electrocatalytic anode is aneffective means of lowering the voltage required forelectrolyzing water. For instance， the oxidation of glucose，methanol， glycerol and 5-hydroxymethylfurfural can effectively reduce the potential required for the reaction and yield productswith greater added value compared to oxygen 12–22. Glycerol， anabundant and inexpensive resource 23，24， has a substantiallylower theoretical oxidation potential （0.003 V vs. RHE）compared to that of OER （1.23 V vs. RHE） 25，26. Moreover， theliquid-phase oxidation products of glycerol have higher addedvalue than oxygen 26–32. Therefore， using the glycerol oxidationreaction （GOR） in place of the OER not only reduces energyconsumption in hydrogen production systems but also facilitatesthe efficient collection of clean hydrogen while yielding liquidproducts with higher added value. Although the approachmentioned above has shown considerable success， it remainschallenging to use low-cost catalysts to simultaneously lower thepotential required for the reaction and produce value-addedproducts with high selectivity.

To date， a majority of the electrocatalysts used have beenprecious metals. However， their high cost and limitedavailability have hindered their widespread practicalapplication 33–36. To address this issue， researchers have beenfocused on using abundant transition metal oxide catalysts toconstruct more economical and efficient hydrogen productionsystems through electrolytic water. Cobalt-based oxides have received significant attention as potential catalysts due to theirrelative abundance and lower cost， as well as their rich redoxchemical properties 37–39. Additionally， cobalt oxide can exist invarious oxidation states， facilitating rapid redox charge transferand exhibiting excellent electrocatalytic activity 39–41. However，the low inherent conductivity of bare transition metal oxides canlimit their practical application. Recent research suggests that theconductivity of bimetal NiCo2O4 is at least twice as high as thatof single metals Co3O4 and NiO， indicating a potentialsynergistic effect between nickel and cobalt ions 42–45. It has beenobserved that adding Ni to Co3O4 can enhance its catalyticactivity by creating more active sites 44，46. In addition， theelectrolyzer system is also a critical factor affecting GORassistedwater electrolysis.

The concept of electrochemical neutralization energy （ENE）has provided a novel idea and method for a distinct class ofhydrogen production systems through electrolytic water 47. Arange of innovative electrolyzers based on ENE was built withthe cathode （pH = 0） and anode （pH = 14）， and the pH gradientbetween the dual chambers can generate a considerable voltage（0.059 × ΔpH）， which can significantly reduce the energyconsumption for electrolytic water cracking 25，48–51. Thisapproach provides an excellent solution for expanding the use ofelectrocatalysts and enriching the selection of electrolytes.

This study investigates the use of a self-supported NiCo2O4nanoneedle electrode on a nickel foam substrate as a catalyst forGOR in a novel hydrogen production system for waterelectrolysis. The NiCo2O4/NF electrode shows high catalyticactivity and selective formate production in a broad voltagerange. The acid-alkaline dual-electrolyte flow electrolyzer（AADEF-electrolyzer） is developed using NiCo2O4/NF as analkaline anode for the GOR， while Pt/C/CC （Carbon Cloth）serves as an acid cathode for the HER. At a voltage as low as0.36 V， the AADEF-electrolyzer can attain an electrolyticcurrent density of 10 mA?cm?2， exhibiting remarkable stability，high auxiliary hydrogen production， and efficient formateproduction.

2 Experimental section

2.1 Reagents and chemicals

Nickel nitrate hexahydrate （Ni（NO3）2?6H2O）， Cobaltousnitrate hexahydrate （Co（NO3）2?6H2O）， ammonium fluoride（NH4F）， and potassium hydroxide （KOH） were procured fromMacklin. Ethanol （C2H5OH） and urea （CH4N2O） were obtainedfrom Shanghai Titan Technology Co.， Ltd. hydrochloric acid（HCl） and Pt/C （20 wt%） were purchased from SinopharmChemical Reagent Co.， Ltd. （Shanghai， China）， while glycerol（C3H8O3） was sourced from Xilong Chemical Co.， Ltd. Thedeionized water （DI water） utilized in the experiment wasproduced in-house within the laboratory. The nickel foamutilized in this study was acquired from Suzhou Taili FoamMetal Factory. All chemicals are of analytical purity and can beused without further purification.

2.2 Preparation of catalyst

The NiCo2O4 nanoneedle self-supported electrode grown insitu on nickel foam （NF） was obtained via a facile hydrothermalcalcinationmethod. Prior to synthesis， a piece of 2 cm × 3 cmNF underwent pretreatment consisting of ultrasonic cleaningwith 3 mol?L?1 HCl， ethanol， and DI water for 10 min each.Subsequently， the NF was dried in a vacuum drying oven andserved as the base material for the next steps. A solution wasprepared by dissolving Co（NO3）2?6H2O （2 mmol） andNi（NO3）2?6H2O （1 mmol） in 30 mL of DI water containing 6mmol of urea and 3 mmol of NH4F. The resulting solution wasthen stirred for 2 h. The pre-treated nickel foam was placed intoa 50 mL PTFE-lined stainless steel autoclave， and the mixedsolution was added. The reaction was conducted at 120 °C for 6h， followed by natural cooling to normal temperature. Theresulting NiCo-pre/NF was washed three times with DI waterand ethanol， then dried under vacuum at 60 °C for 12 h. Finally，the NiCo-pre/NF was heated at a rate of 2 °C?min?1 in a furnaceuntil it reached 350 °C and held for 2 h to form NiCo2O4/NF.Additionally， Co3O4/NF and NiOx/NF were also synthesized viaa similar process.

2.3 Materials characterization

The structure and crystal morphology of the material werecharacterized by X-ray diffraction （XRD） using an X-raydiffractometer （Cu Kα） （D8ADVANCE-A25， Bruker， Germany）.Using a Nova NanoSEM 450 （FEI， USA）， scanning electronmicroscope （SEM） pictures were taken. The images werecaptured on a Talos F200X （FEI， USA） using transmissionelectron microscopy （TEM） and energy-dispersive X-rayspectroscopy （EDS）. X-ray photoelectron spectroscopy （XPS）was used to determine the surface compositions of the catalystsby an ESCALAB 250Xi （Thermo Scientific， USA）.Additionally， superconducting nuclear magnetic resonancespectroscopy （NMR） on an AVANCE 400 （Jobin Yvon， France）was used to characterize the products of electrocatalytic glyceroloxidation.

2.4 Reagents and chemicals

The electrochemical measurements were conducted using athree-electrode system with a CHI 760E ElectrochemicalWorkstation in 1.0 mol?L?1 potassium hydroxide （KOH）electrolyte. To minimize the influence of dissolved oxygen onthe results， the electrolyte was purged with nitrogen for 30 minuntil it reached saturation before the test. The working electrodescomprised as-made nanofiber-based （NF-based） compositeswith dimensions of 1 cm × 1 cm. A carbon rod served as thecounter electrode， and Hg/HgO （1 mol?L?1 KOH） was used asthe reference electrode. The measured potentials were calibratedto the reversible hydrogen electrode （RHE） using the equation：E（RHE） = E（Hg/HgO） + 0.05916 × pH + 0.098. Prior to linearsweep voltammetry （LSV）， cyclic voltammetry （CV）measurements were conducted several times to activate thesystem until the cycle stabilized. Both LSV and CV werescanned at a rate of 5 mV?s?1 without iR-compensation.Electrochemical impedance spectroscopy （EIS） was carried outwith the working electrode at a potential of 0.5 V vs. Hg/HgO，while the frequency was swept from 10 kHz to 0.1 Hz with asinusoidal voltage amplitude of 5 mV. To determine theelectrochemical double layer capacitance （Cdl） of the material，CV measurements were performed at various scan rates （20， 40，60， 80， 100 and 120 mV?s?1） in the non-Faraday reaction interval（?0.3 to ?0.2 V vs. Hg/HgO）. The Cdl value was calculated fromthe slope of the current density difference （Δj = ja ? jc， where jcand ja are the cathodic and anodic current densities， respectively）plotted against the scan rate. The Cdl data were used to determinethe electrochemically active surface area （ECSA）.

2.5 Assembly and tests of the AADEF-electrolyzer

An AADEF-electrolyzer was assembled using a NF compositematerial as the anode， 1 mol?L?1 KOH solution as the anolyte， aplatinum-carbon catalyst loaded on carbon cloth （Pt/C/CC） as thecathode， and 0.5 mol?L?1 H2SO4 as the catholyte. To preparePt/C/CC electrodes， Pt/C ink was prepared by dispersing 5.0 mgof 20 wt% Pt/C catalyst into a mixture of 50 μL Nafion， 50 μLethanol and 400 μL deionized water and sonicating for 15 min. 50μL of Pt/C ink was then applied to 1 cm?2 of CC and dried naturallyat normal temperature to obtain a Pt/C/CC electrode with a loadingof approximately 2 mg?cm?2. The anode and cathode chamberswere separated by a cation exchange membrane （CEM）， with theelectrolyte circulated by a flow pump （Approx. 16.5 mL?min?1，NKCP-B08B， Kamoer）. The electrochemical performance of theelectrolyzer was assessed by LSV with a scan rate of 5 mV?s?1.Additionally， the stability of the electrolyzer was assessed throughchronopotentiometry （CP）.

2.6 Product analysis

The liquid products obtained by glycerol oxidation wereanalyzed through nuclear magnetic resonance （NMR） using theAVANCE 400 system manufactured by Jobin Yvon in France.The internal standard technique was implemented for everysolution analyzed， utilizing dimethyl sulfoxide （DMSO） servingas the internal standard. To be specific， a 2 mL sample of theelectrolyte to be tested was combined with 400 μL of D2O and3.4 μL of DMSO for NMR analysis.

Moreover， the standard solution of glycerol and formate was subjected to the same conditions.

The Faraday efficiency （FE） for the production of formate and hydrogen can be calculated by utilizing Eqs. （1） and （2）：

FE（formate yield） =N（fomate yield）／Q1/（z1 × F）× 100% （1）

FE（H2 Production） =N（H2 Production）／Q2/（z2 × F）× 100% （2）

where Q1 and Q2 are the total charges passed through theelectrodes， Z1 = 8/3 represents the number of electrons requiredto form a mole of formate， Z2 = 2 represents the number ofelectrons required to produce a molecule of H2， and F is theFaraday constant （96， 485 C?mol?1） 18，52，53.

2.7 Reagents and chemicals

The calculations were performed using PWSCF codes included in the Quantum ESPRESSO distribution 54. Weemployed spin-polarized density functional theory （DFT）calculations to compute periodic supercells using the Perdew-Burke-Ernzerhof （PBE） functional of the generalized gradientapproximation （GGA） for exchange-correlation and ultrasoftpseudopotentials for both nuclei and core electrons. The Kohn-Sham orbitals were expanded using a plane-wave basis set witha kinetic energy cutoff of 30 Ry and a charge-density cutoff of300 Ry. To account for Fermi-surface effects， we employed theMethfessel and Paxton smearing technique with a smearingparameter of 0.02 Ry. For NiCo2O4 （001）， a 2 × 2 supercell andfour-layer slab are utilized. For Co3O4 （100）， a 2 × 1 supercelland four-layer slab are utilized. The bottom layer is fixed tomodel NiCo2O4 and Co3O4 bulk. To eliminate the interaction ofadjacent atomic slabs in the z direction， a vacuum layer ofapproximately 15 ? was utilized. During the geometricoptimization process， every atom， including the adsorbates，underwent relaxation until the Cartesian force components oneach atom fell below 10?3 Ry/Bohr and the total energyconverged to within 10?5 Ry. The Brillouin zones were sampledusing 1 × 1 × 1 k-point mesh.

To calculate the Gibbs free energy differences， the adsorption free energy of the adsorbates was determined using Eq. （3）：

GA = EA + ZPE ? TS + ∫CpdT （3）

where EA is the total energy of a particular molecule or adsorbateA*. If A represents a molecule， the total energies can becalculated directly. However， if A represents an adsorbate， it iscalculated as the difference between the DFT-basedsubstrate with （EA*DFT） and without adsorbate A （E*DFT）， asshown in Eq. （4）：

EA = EA*DFT ? E*DFT （4）

The corrections from zero-point energy， entropy， and heat capacity are denoted as ZPE， TS and ∫CpdT， respectively.

3 Results and discussion

3.1 Material characterization

The superior catalytic oxidation characteristics of Ni and Co，as well as their synergistic influence on electrochemicalprocesses， have been widely recognized. Based on thisknowledge， our research focused on the development of a selfsupportedNiCo2O4 electrode grown in situ on a nickel foam（NF） substrate （NiCo2O4/NF）. Fig. S1 （Supporting Information）demonstrates the procedure of synthesizing the electrodematerials. Initially， the precursor material （NiCo-pre/NF） （Fig.S2） was synthesized via an in situ hydrothermal growth method，utilizing a pretreated NF substrate （Fig. 1a） possessing a smoothsurface， followed by hydrothermal calcination to produce theself-supported NiCo2O4/NF electrode （Fig. 1b，c）. It is noteworthythat the surface of the NF after hydrothermal calcination exhibitsa dense and uniform nanoneedle morphology， which is moreuniform than that of the precursor material （Fig. S2）. The surfacecolor of the NF after calcination changes from brown to black， asshown in the digital photo （Fig. S3）.

Transmission electron microscopy （TEM） images ofNiCo2O4/NF （Fig. 1d） confirm the presence of the nanoneedlestructure. Furthermore， the lattice stripes at a lattice distance of0.245 nm， shown in Fig. 1e， correspond to the （311） crystal planeof NiCo2O4， as observed by high-resolution TEM （HRTEM）imaging. The uniform distribution of the three elementsthroughout the nanoneedles is demonstrated by the energydispersion spectroscopy （EDS） element mapping analysis， aspresented in Fig. 1f.

The crystal structure of NiCo2O4/NF was thoroughlyexamined using X-ray diffraction （XRD） analysis. As presentedin Fig. 2a， the primary diffraction peaks observed in the XRDpattern perfectly match those of NiCo2O4 （JCPDS No. 20-0781）and Ni （JCPDS No. 04-0850）， suggesting that the primaryconstituent of NF composite is composed of the NiCo2O4 phaseand Ni phase. In addition to the diffraction peaks at 44.5°， 51.8°and 76.4° of the NF substrate， diffraction peaks at 18.9°， 31.1°，36.7°， 59.1°， and 65.0° are ascribed to the （111）， （220）， （311），（511）， and （440） crystal planes of NiCo2O4 （JCPDS No. 20-0718）. XRD analysis also was conducted to determine the crystalstructures of the NiCo-pre/NF， Co3O4/NF， and NiOx/NFmaterials prepared using the same experimental approach （Fig.S4）. The results indicate that the Co3O4/NF material exhibiteddistinct diffraction peaks of Co3O4 and Ni. On the other hand，both NiCo-pre/NF and NiOx/NF show only a single metal Nipeak due to the strong peak of nickel.

X-ray photoelectron spectroscopy （XPS） measurements wereused to analyze the surface elemental composition ofNiCo2O4/NF. The presence of Ni， Co， and O elements inNiCo2O4/NF is confirmed by the XPS spectra （Fig. 2b）， with thevalence states of these elements examined in further detail.Therefore， further analysis of Ni 2p， Co 2p， and O 1s spectra wascarried out. The Ni 2p spectrum （Fig. 2c） exhibits two satellitepeaks and two spin-orbit peaks， with binding energies of 854.2and 871.8 eV. The fitting peaks of 855 and 871.8 eV are assignedto Ni2+ 55. The Co 2p spectrum （Fig. 2d） shows two satellitepeaks and two spin-orbit peaks with binding energies of 779.5and 795 eV， which are attributed to Co3+. Additionally， two otherfitting peaks at 780.5 and 796.4 eV are assigned to Co2+ 56，57.Finally， The O 1s profile （Fig. 2e） reveals the contributions ofthree types of oxygen： O1 （529.9 eV）， O2 （531.5 eV）， and O3（533.4 eV）， which are associated with metal-oxygen bonds，hydroxyl groups， and chemically adsorbed water， respectively 43.The appearance of C 1s in the spectra （Fig. 2f） is attributed to thepresence of ubiquitous carbon. The surface elementalcomposition of Co3O4/NF and NiOx/NF are also measured andanalyzed using XPS. （Figs. S5–6）.

3.2 Electrocatalytic performance of glycerol oxidationreaction

To investigate the catalytic activity of the NiCo2O4/NFelectrode towards anodic electrocatalytic oxidation of glycerol，electrochemical tests were conducted in a three-electrodesystem. Fig. 3a demonstrates the LSV curves of the NiCo2O4/NFelectrode at OER and GOR. The LSV curves （Figs. 3a and S7）indicate that the NiCo2O4/NF electrode displays thedistinguished catalytic activity for GOR. In the absence ofglycerol， the NiCo2O4/NF electrode displays normal OERcatalytic activity， the catalytic current density of 10 mA?cm?2 canonly be achieved at a high potential of 1.51 V vs. RHE. It is worthnoting that with the mixing of 0.5 mol?L?1 glycerol in 1.0mol?L?1 KOH， the NiCo2O4/NF electrode requires only 1.15 Vvs. RHE to reach 10 mA?cm?2， indicating its superior GORcatalytic activity. Furthermore， Tafel slope derived from the LSVcurve shows that GOR has a slope of 128.9 mV?dec?1 comparedwith OER’s 145.7 mV?dec?1， suggesting faster catalytic reactionkinetics for GOR （Fig. 3b）. Notably， the GOR activity of theNiCo2O4/NF electrode outperform those of Co3O4/NF， NiOx/NF，NiCo-pre/NF， and bare NF （Figs. 3c and S8）， this may be relatedto the fact that NiCo2O4/NF has a higher ECSA and lowerimpedance. In addition， the influence of glycerol concentrationin the electrolyte on GOR catalytic activity was studied， and theoptimal glycerol concentration was found to be 0.5 mol?L?1 （Fig.S9）. The results of EIS indicate that the NiCo2O4/NF electrodehas a lower charge transfer resistance than the other electrodesand a fast charge transfer rate （Fig. S10）. The NiCo2O4/NFelectrode’s ECSA was determined through the calculation of double-layer capacitance （Cdl） from the CV curve of the non-Faraday interval. The Cdl value of NiCo2O4/NF （49.34 mF?cm?2）is found to be considerably higher than that of Co3O4/NF （33.42mF?cm?2） and NiOx/NF （1.08 mF?cm?2） （Fig. S11）. Thesefindings suggest that the NiCo2O4/NF electrode has the largestECSA and exposes more active sites， further confirming itsexcellent GOR catalytic activity.

By 1H NMR analysis of the oxidation products of glycerol onthe NiCo2O4/NF electrode， we added 0.5 mol?L?1 glycerol toa 1 mol?L?1 KOH electrolyte and carried out a 2 hchronoamperometry （i–t） test at various potentials （1.2–1.5 V vs.RHE） （Fig. S12）. The results show that the NiCo2O4/NFelectrode is capable of producing formate with exceptionalselectivity across a broad range of potential （1.2 to 1.5 V vs.RHE）， with an FE exceeding 85%. Moreover， the highestformate yield （91.2%） is achieved at 1.4 V vs. RHE （Figs. 3d andS13）. The calculated FE for formate production was less than100%， presumably due to the generation of gaseous byproductssuch as CO2 and O2 during the electrolysis process. Notably，during electrolysis at a constant current density of 10 mA?cm?2，the NiCo2O4/NF electrode remains active for 12 h without anysignificant changes in the LSV curves before and after thestability test （Figs. 3e and S14） and also that SEM testing of thematerial after the stability test shows no significant change in thematerial （Fig. S15）. Qualitative and quantitative analysis of theelectrolyte before and after the stability test using 1H NMRrevealed that only formate was present as the anodic glycerolelectrolytic product （Fig. 3f）.

3.3 Theoretical computation

In order to elucidate the differences between Co3O4 andNiCo2O4 in GOR， we employed DFT calculations. Specifically，the computational hydrogen electrode （CHE） method was utilized to investigate the activities 58. The structures of NiCo2O4and Co3O4 are presented in Fig. 4a，b， with the corresponding freeenergy diagram （FED） plotted in Fig. 4c （see calculation detailsin Sections 1–2， Supporting Information）. Our results suggestthat the reaction energy barrier of NiCo2O4 is less than that ofCo3O4， suggesting that it possesses superior catalytic activitytoward GOR. Furthermore， we calculated the projected densityof states of Co atoms on the surface of Co3O4 and NiCo2O4 to be?1.50 and ?1.65 eV， respectively （Fig. 4d，e）. According to the dbandcenter theory， this indicates a decrease in the antibondingenergy state， which weakens the interaction between theintermediate and surface 59. The different positions of theNiCo2O4（001） cut surfaces are shown in Fig. S16. The freeenergy calculations for the Co surface and the Co/Ni surface areshown in Fig. S17. Their calculated structures are shown in Figs.S18–20. The results show that the Co surface of NiCo2O4（001）has a smaller decision speed step （RDS）， which is morefavourable for the GOR reaction. The Ni/Co surface of NiCo2O4（001） has a better Co as active centre compared to Ni. Takentogether， these findings indicate that intermediates adsorbed onthe surface of NiCo2O4 are more likely to desorb， resulting in theproduction of formic acid， which is consistent with ourexperimental results （Fig. 3c）.

3.4 Performance of the AADEF-electrolyzer

Considering the favorable electrochemical catalytic activityexhibited by the NiCo2O4/NF electrode and the underlyingprinciples of electrochemical neutralization energy （ENE）theory， we have designed an AADEF-electrolyzer， as illustratedin Figs. 5a and S21. In this electrolyzer， the anode was composedof NiCo2O4/NF， while the anode electrolyte comprises 0.5mol?L?1 glycerol and 1.0 mol?L?1 KOH. To separate the anodeand cathode chambers， a cation exchange membrane （CEM） was utilized， with the cathode comprising Pt/C and 0.5 mol?L?1H2SO4. The anodic oxidation of glycerol generated electrons，which then passed through the external circuit to the cathode，where H+ ions received these electrons and underwent reductionto form H2， while K+ ions migrated through the CEM into thecathode chamber to form a complete circuit. In the absence ofglycerol in the anode solution， a significant voltage （0.059 ×ΔpH） can be produced by the pH gradient between the anodechamber （pH = 14） and the cathode chamber （pH = 0） due to thepresence of ENE. As such， an AADEF-electrolyzer can generatea current density of 10 mA?cm?2 when the applied voltage was0.54 V （Fig. 5b）， which was significantly lower than thetheoretical voltage required for hydrolysis （1.23 V）. As expected，the introduction of 0.5 mol?L?1 glycerol resulted in a markedreduction in the required applied voltage for 10 mA?cm-2， with avalue of 0.36 V observed. This is notably lower than what istypically observed in traditional alkaline electrolytic cells （Figs.5b and S22）. Moreover， bare NF was also examined as the anodeof the electrolyzer， further highlighting the modification andelectrocatalysis of NF by NiCo2O4 （Fig. S23）. Subsequentexamination of the electrolyte products following the anodicreaction revealed the presence of exclusively liquid formateproducts， exhibiting an average Faradaic efficiency ofapproximately 85%. Meanwhile， the cathodic HER resulted inthe highly efficient production of hydrogen with nearly 100%Faraday efficiency， as depicted in Fig. 5c. Furthermore， thechronopotentiometry technique was employed to assess theenduring stability of the AADEF-electrolyzer at a constantcurrent density of 50 mA?cm?2. The results demonstrated that theAADEF-electrolyzer exhibited exceptional stability，maintaining steady electrolysis for 150 h with some minorpotential fluctuations （Fig. 5d）. Replenishing the electrolyte canreverse the rise in potential induced by electrolysis， which canbe attributed to the depletion of H+， OH?， and glycerol in theelectrolyte.

4 Conclusions

In summary， we develop a self-supported NiCo2O4/NFelectrode through a simple hydrothermal-calcination method，which shows excellent electrocatalytic performance for GOR.The electrode achieves a high current density at low potentialsand displays remarkable selectivity towards formate production，with a Faraday efficiency over 85%. DFT calculations indicatethat the lower GOR reaction energy barrier of NiCo2O4/NFenhances its catalytic activity. We also develop an AADEFelectrolyzerthat is capable of producing hydrogen from water ata voltage as low as 0.36 V， while exhibiting excellent stability，high-efficiency formate production， and auxiliary hydrogenproduction performance. Our work offers a novel approach todesigning electrocatalysts with exceptional performance tofacilitate the value-added conversion of chemicals and thedevelopment of novel hybrid electrolytic systems or devices forrelated electrochemical reactions.

Author Contributions： Xin Feng： Ideas， Conceptualization，Methodology， Data collection and curation， Data analysis andinterpretation， Writing-original draft. Kexin Guo： Graphicanalysis， Writing， Revision. Chunguang Jia： Theoreticalcalculations， Graphic analysis. Bowen Liu： Discussion， Graphicanalysis. Suqin Ci： Supervision， Editing， discussion andResources. Junxiang Chen： Guide Theoretical calculations，Revision. Zhenhai Wen： Writing： Review amp; Editing，Supervision.

Supporting Information： available free of charge via theinternet at http：//www.whxb.pku.edu.cn.

References

（1） Su， H.; Jiang， J.; Song， S.; An， B.; Li， N.; Gao， Y.; Ge， L. Chin. J.Catal. 2023， 44， 7. doi： 10.1016/s1872-2067（22）64149-4

（2） Chen， H.; Chen， J.; He， H.; Chen， X.; Jia， C.; Chen， D.; Liang， J.; Yu，D.; Yao， X.; Qin， L.; et al. Appl. Catal. B 2023， 323， 122187. doi：10.1016/j.apcatb.2022.122187

（3） Yin， F.; Qin， P.; Xu， J.; Cao， S. Acta Phys. -Chim. Sin. 2023， 39 （11），2212062. [殷方鑫， 秦品權(quán)， 許景三， 曹少文. 物理化學(xué)學(xué)報，2023， 39 （11）， 2212062.] doi： 10.3866/PKU.WHXB202212062

（4） Moreira， R.; Bimbela， F.; Gandía， L. M.; Ferreira， A.; Sánchez， J. L.;Portugal， A. Renew. Sust. Energ. Rev. 2021， 148， 111299.doi： 10.1016/j.rser.2021.111299

（5） Li， X.; Hao， X.; Abudula， A.; Guan， G. J. Mater. Chem. A 2016， 4（31）， 11973. doi： 10.1039/c6ta02334g

（6） Li， X.; Yu， J.; Jaroniec， M. Chem. Soc. Rev. 2016， 45 （9）， 2603.doi： 10.1039/c5cs00838g

（7） Sun， L.; Dai， Z.; Zhong， L.; Zhao， Y.; Cheng， Y.; Chong， S.; Chen， G.;Yan， C.; Zhang， X.; Tan， H.; et al. Appl. Catal. B 2021， 297， 120477.doi： 10.1016/j.apcatb.2021.120477

（8） Yang， S.; Du， R.; Yu， Y.; Zhang， Z.; Wang， F. Nano Energy 2020， 77，105057. doi： 10.1016/j.nanoen.2020.105057

（9） Ding， M.; Chen， J.; Jiang， M.; Zhang， X.; Wang， G. J. Mater. Chem. A2019， 7 （23）， 14163. doi： 10.1039/c9ta00708c

（10） Liu， Y.; Wang， Y.; Liu， B.; Amer， M.; Yan， K. Acta Phys. -Chim. Sin.2023， 39 （2）， 2205028. [劉瑤鈺， 王宇辰， 劉碧瑩， Mahmoud Amer，嚴(yán)凱. 物理化學(xué)學(xué)報， 2023， 39 （2）， 2205028.]doi： 10.3866/PKU.WHXB202205028

（11） Lü， L.; Zhang， L.; He， X.; Yuan， H.; Ouyang， S.; Zhang， T. ActaPhys. -Chim. Sin. 2021， 37 （7）， 2007079. [呂琳， 張立陽， 何雪冰，原弘， 歐陽述昕， 張鐵銳. 物理化學(xué)學(xué)報， 2021， 37 （7）， 2007079.]doi： 10.3866/PKU.WHXB202007079

（12） Zhou， B.; Dong， C.-L.; Huang， Y.-C.; Zhang， N.; Wu， Y.; Lu， Y.; Yue，X.; Xiao， Z.; Zou， Y.; Wang， S. J. Energy Chem. 2021， 61， 179.doi： 10.1016/j.jechem.2021.02.026

（13） Song， Y.; Xie， W.; Song， Y.; Li， H.; Li， S.; Jiang， S.; Lee， J. Y.; Shao， M.Appl. Catal. B 2022， 312， 121400. doi： 10.1016/j.apcatb.2022.121400

（14） Xie， Y.; Zhou， Z.; Yang， N.; Zhao， G. Adv. Funct. Mater. 2021， 31（34）， 2102886. doi： 10.1002/adfm.202102886

（15） Li， J.; Wei， R.; Wang， X.; Zuo， Y.; Han， X.; Arbiol， J.; Llorca， J.;Yang， Y.; Cabot， A.; Cui， C. Angew. Chem. Int. Ed. 2020， 59 （47），20826. doi： 10.1002/anie.202004301

（16） Zhou， B.; Li， Y.; Zou， Y.; Chen， W.; Zhou， W.; Song， M.; Wu， Y.; Lu，Y.; Liu， J.; Wang， Y.;et al. Angew. Chem. Int. Ed. 2021， 60 （42），22908. doi： 10.1002/anie.202109211

（17） Zheng， D.; Li， J.; Ci， S.; Cai， P.; Ding， Y.; Zhang， M.; Wen， Z. Appl.Catal. B 2020， 277， 119178. doi： 10.1016/j.apcatb.2020.119178

（18） Xiang， K.; Wu， D.; Deng， X.; Li， M.; Chen， S.; Hao， P.; Guo， X.; Luo，J. L.; Fu， X. Z. Adv. Funct. Mater. 2020， 30 （10）， 1909610.doi： 10.1002/adfm.201909610

（19） Liu， W. J.; Xu， Z.; Zhao， D.; Pan， X. Q.; Li， H. C.; Hu， X.; Fan， Z. Y.;Wang， W. K.; Zhao， G. H.; Jin， S.; et al. Nat. Commun. 2020， 11 （1），265. doi： 10.1038/s41467-019-14157-3

（20） Han， X.; Sheng， H.; Yu， C.; Walker， T. W.; Huber， G. W.; Qiu， J.;Jin， S. ACS Catal. 2020， 10 （12）， 6741. doi： 10.1021/acscatal.0c01498

（21） Vo， T.-G.; Ho， P.-Y.; Chiang， C.-Y. Appl. Catal. B 2022， 300， 120723.doi： 10.1016/j.apcatb.2021.120723

（22） Talebian-Kiakalaieh， A.; Amin， N. A. S.; Rajaei， K.; Tarighi， S. Appl.Energy 2018， 230， 1347. doi： 10.1016/j.apenergy.2018.09.006

（23） Jamil， F.; Al-Haj， L.; Al-Muhtaseb， A. H.; Al-Hinai， M. A; Baawain，M.; Rashid， U.; Ahmad， M. N. M. Rev. Chem. Eng. 2018， 34 （2）， 267.doi： 10.1515/revce-2016-0026

（24） Deng， C.-Q.; Deng， J.; Fu， Y. Green Chem. 2022， 24 （21）， 8477.doi： 10.1039/d2gc03235j

（25） Fan， L.; Ji， Y.; Wang， G.; Chen， J.; Chen， K.; Liu， X.; Wen， Z. J. Am.Chem. Soc. 2022， 144 （16）， 7224. doi： 10.1021/jacs.1c13740

（26） Fan， L.; Liu， B.; Liu， X.; Senthilkumar， N.; Wang， G.; Wen， Z. EnergyTechnol. 2020， 9 （2）， 2000804. doi： 10.1002/ente.202000804

（27） Duan， Y.; Liu， Z.; Zhao， B.; Liu， J. RSC Adv. 2020， 10 （27）， 15769.doi： 10.1039/d0ra00564a

（28） Bai， J.; Huang， H.; Li， F.-M.; Zhao， Y.; Chen， P.; Jin， P.-J.; Li， S.-N.;Yao， H.-C.; Zeng， J.-H.; Chen， Y. J. Mater. Chem. A 2019， 7 （37），21149. doi： 10.1039/c9ta08806g

（29） Brix， A. C.; Morales， D. M.; Braun， M.; Jambrec， D.; Junqueira， J. R.;Cychy， S.; Seisel， S.; Masa， J.; Muhler， M.; Andronescu， C.; et al.ChemElectroChem 2021， 8 （12）， 2336. doi： 10.1002/celc.202100739

（30） Kim， H. J.; Kim， Y.; Lee， D.; Kim， J.-R.; Chae， H.-J.; Jeong， S.-Y.;Kim， B.-S.; Lee， J.; Huber， G. W.; Byun， J.; et al. ACS Sustain Chem.Eng. 2017， 5 （8）， 6626. doi： 10.1021/acssuschemeng.7b00868

（31） Dodekatos， G.; Schünemann， S.; Tüysüz， H. ACS Catal. 2018， 8 （7），6301. doi： 10.1021/acscatal.8b01317

（32） Alaba， P. A.; Lee， C. S.; Abnisa， F.; Aroua， M. K.; Cognet， P.; Pérès，Y.; Wan Daud， W. M. A. Rev. Chem. Eng. 2021， 37 （7）， 779.doi： 10.1515/revce-2019-0013

（33） Yang， F.; Ye， J.; Yuan， Q.; Yang， X.; Xie， Z.; Zhao， F.; Zhou， Z.;Gu， L.; Wang， X. Adv. Funct. Mater. 2020， 30 （11）， 1908235.doi： 10.1002/adfm.201908235

（34） Lam， C. H.; Bloomfield， A. J.; Anastas， P. T. Green Chem. 2017，19 （8）， 1958. doi： 10.1039/c7gc00371d

（35） Lee， S.; Kim， H. J.; Lim， E. J.; Kim， Y.; Noh， Y.; Huber， G. W.; Kim，W. B. Green Chem. 2016， 18 （9）， 2877. doi： 10.1039/c5gc02865e

（36） De Souza， M. B.; Vicente， R. A.; Yukuhiro， V. Y.; Pires， C. T.;Cheuquepán， W.; Bott-Neto， J. L.; Solla-Gullón， J.; Fernández， P. S.ACS Catal. 2019， 9 （6）， 5104. doi： 10.1021/acscatal.9b00190

（37） Zhou， Z.; Chen， C.; Gao， M.; Xia， B.; Zhang， J. Green Chem. 2019，21 （24）， 6699. doi： 10.1039/c9gc02880c

（38） Lu， Y.; Liu， T.; Dong， C. L.; Yang， C.; Zhou， L.; Huang， Y. C.; Li， Y.;Zhou， B.; Zou， Y.; Wang， S. Adv. Mater. 2022， 34 （2）， e2107185.doi： 10.1002/adma.202107185

（39） Wang， Y.; Zhu， Y.-Q.; Xie， Z.; Xu， S.-M.; Xu， M.; Li， Z.; Ma， L.;Ge， R.; Zhou， H.; Li， Z.; et al. ACS Catal. 2022， 12432.doi： 10.1021/acscatal.2c03162

（40） Lv， J.; Wang， L.; Li， R.; Zhang， K.; Zhao， D.; Li， Y.; Li， X.; Huang，X.; Wang， G. ACS Catal. 2021， 14338.doi： 10.1021/acscatal.1c03960

（41） Zhou， H.; Zheng， M.; Pang， H. Chem. Eng. J. 2020， 38， 127884.doi： 10.1016/j.cej.2020.127884

（42） Wei， T. Y.; Chen， C. H.; Chien， H. C.; Lu， S. Y.; Hu， C.C. Adv. Mater.2010， 22 （3）， 347. doi： 10.1002/adma.200902175

（43） Liu， Q.; Xie， L.; Liang， J.; Ren， Y.; Wang， Y.; Zhang， L.; Yue， L.;Li， T.; Luo， Y.; Li， N.; et al. Small 2022， 18 （13）， e2106961.doi： 10.1002/smll.202106961

（44） Jo， H. J.; Shit， A.; Jhon， H. S.; Park， S. Y. J. Ind. Eng. Chem. 2020，89， 485. doi： 10.1016/j.jiec.2020.06.028

（45） Hu， L.; Wu， L.; Liao， M.; Hu， X.; Fang， X. Adv. Funct. Mater. 2012，22 （5）， 998. doi： 10.1002/adfm.201102155

（46） Zhao， Q.; Yan， Z.; Chen， C.; Chen， J. Chem. Rev. 2017， 117 （15），10121. doi： 10.1021/acs.chemrev.7b00051

（47） Ding， Y.; Cai， P.; Wen， Z. Chem. Soc. Rev. 2021， 50 （3）， 1495.doi： 10.1039/d0cs01239d

（48） Zhang， C.; Ci， S.; Peng， X.; Huang， J.; Cai， P.; Ding， Y.; Wen， Z.J. Energy Chem. 2021， 54， 30. doi： 10.1016/j.jechem.2020.04.073

（49） Wang， G.; Chen， J.; Cai， P.; Jia， J.; Wen， Z. J. Mater. Chem. A 2018，6 （36）， 17763. doi： 10.1039/c8ta06827e

（50） Zhang， M.; Chen， J.; Li， H.; Cai， P.; Li， Y.; Wen， Z. Nano Energy2019， 61， 576. doi： 10.1016/j.nanoen.2019.04.050

（51） Liu， B.; Wang， G.; Feng， X.; Dai， L.; Wen， Z.; Ci， S. Nanoscale 2022，14， 12841. doi： 10.1039/d2nr02689a

（52） Li， Y.; Wei， X.; Chen， L.; Shi， J.; He， M. Nat. Commun. 2019， 10 （1），5335. doi： 10.1038/s41467-019-13375-z

（53） Xu， Y.; Liu， M.; Wang， S.; Ren， K.; Wang， M.; Wang， Z.; Li， X.;Wang， L.; Wang， H. Appl. Catal. B 2021， 298， 120493.doi： 10.1016/j.apcatb.2021.120493

（54） Giannozzi， P.; Baroni， S.; Bonini， N.; Calandra， M.; Car， R.;Cavazzoni， C.; Ceresoli， D.; Chiarotti， G. L.; Cococcioni， M.; Dabo，I.; et al. J. Phys. Condens. Matter 2009， 21 （39）， 395502.doi： 10.1088/0953-8984/21/39/395502

（55） Yan， X.; Zhang， W.-D.; Hu， Q.-T.; Liu， J.; Li， T.; Liu， Y.; Gu， Z.-G.Int. J. Hydrogen Energy 2019， 44 （51）， 27664.doi： 10.1016/j.ijhydene.2019.09.004

（56） Sun， B.; Miao， F.; Tao， B.; Wang， Y.; Zang， Y.; Chu， P. K. J. Phys.Chem. Solids 2021， 158， 110255. doi： 10.1016/j.jpcs.2021.110255

（57） Qian， L.; Luo， S.; Wu， L.; Hu， X.; Chen， W.; Wang， X. Appl. Surf. Sci.2020， 503， 144306. doi： 10.1016/j.apsusc.2019.144306

（58） Norskov， J. K.; Rossmeisl， J.; Logadottir， A.; Lindqvist， L.; Kitchin， J.R.; Bligaard， T.; Jonsson， H. J. Phys. Chem. B 2004， 108 （46）， 17886.doi： 10.1021/jp047349j

（59） Norskov， J. K.; Bligaard， T.; Rossmeisl， J.; Christensen， C. H. Nat.Chem. 2009， 1 （1）， 37. doi： 10.1038/nchem.121

國家自然科學(xué)基金（22168025）和江西省自然科學(xué)基金（20192BAB203013， 20202ACBL203003）資助項目