MaterialsThe catalyst support carbon black (BLACK PEARLS 2000 LOT-1366221) was purchased from Cabot Corporation. Pd/C was purchased from Adamas, Nickel (II) nitrate hexahydrate (Ni(NO3)2·6H2O, Analytical Reagent, AR), sodium hydroxide (NaOH, 99.9%), potassium hydroxide (KOH, 85%), anhydrous ethanol (C2H5OH, 99.5%), methanol (CH3OH, >99.9%), concentrated nitric acid (65%), ammonium formate (CH5NO2, 99%) were purchased from Aladdin Reagent Company. Sodium tetrachloropalladate (II) (Na2PdCl4, 99.9%) was supplied by Sino-Platinum Co. Ltd. 5-Hydroxymethyl (HMF, 99%), 5-Formylfuran-2-carboxylic acid (HMFCA, 98%) 2,5-Furandicarboxaldehyde (DFF, 98%), 5-Formylfuran-2-carboxylic acid (FFCA, 98%),2,5-Furandicarboxylic acid (FDCA, 99%) were obtained from Aladdin. The pure water used in all experiments was obtained using a pure water system (Milli-Q, 18.2 MΩ). The above-mentioned reagents were used as received without further purification. Electrolyte KOH and HMF were dissolved in pure water, respectively. To avoid HMF self-polymerization, the electrolyte was used immediately after preparation.Preparation of Ni(OH)2/CFirstly, Ni(NO3)2·6H2O (0.877 g) was sonically dissolved in pure water (82.5 mL) for 15 min, followed by vigorous magnetic stirring for 10 min. Then, anhydrous ethanol (12.5 mL) was added dropwise into the solution, followed by vigorous magnetic stirring for another 10 min. Next, carbon black (0.125 g) was added, and the resulting dispersion was sonicated for 15 min, followed by stirring for 1 h to adsorb the nickel ions. Subsequently, 3.6 M NaOH (12.5 mL) was quickly injected into the dispersion under vigorous magnetic stirring to nucleate Ni(OH)2 rapidly. The beaker was then covered with plastic wrap, and the dispersion stirred for another 20 h. Finally, the solid product was collected by filtration and then dried under vacuum at 60 °C. The obtained sample is denoted herein as Ni(OH)2/C.Preparation of Pd/Ni(OH)2
Ni(OH)2/C (0.084 g) was dispersed in pure water (200 mL) under vigorous magnetic stirring for 30 min. Then, a specific volume of Na2PdCl4 solution (19 mg/mL) was added to the dispersion under continuous stirring, with the resulting dispersion then stirred for 20 h. Next, the mixture was collected by filtration and dried under a vacuum (60 °C). The obtained sample was denoted herein as Pd2+/Ni(OH)2. By this approach, Pd2+/Ni(OH)2 with different Pd loadings was obtained by controlling the volume of the Na2PdCl4 solution. The Pd2+/Ni(OH)2 was electrochemically reduced by cyclic voltammetry between 0 V and 0.8 V (vs. RHE) at a scan rate of 200 mV s−1 for 100 cycles. This transformed the adsorbed Pd2+ to Pd0. The obtained catalysts are denoted herein as Pd/Ni(OH)2.Preparation of Pd/Ni(OH)2-etchingThe as-prepared Pd/Ni(OH)2 (0.035 g) was dispersed in a 0.5 M nitric acid solution (25 mL) under continuous stirring for 1.5 h. After filtration and drying in a vacuum (60 °C) overnight, a Pd/Ni(OH)2-etching catalyst was successfully synthesized (in which most of the Ni(OH)2 support had been removed).Inductively coupled plasma optical emission spectrometry (ICP-OES)The standardization curves of different Pd and Ni concentrations were collected before sample measurements. Moreover, the sample pretreatment was conducted by microwave digestion in aqua regia with hydrofluoric acid. Then, Pd and Ni loadings in the synthesized catalysts were determined by inductively coupled plasma optical emission spectrometry (ICP-OES) on an iCAP 7200 Duo (Thermofisher Scientific).X-ray diffraction (XRD)The catalysts’ X-ray diffraction (XRD) patterns were collected on an X-ray diffractometer (PANayltical Empyreanquipped with a Cu Kα source (60 mA and 60 kV, 5 ~ 90°).X-ray photoelectron spectroscopy (XPS)X-ray photoelectron spectroscopy (XPS) data were collected on a Thermo Fisher Scientific Escalab Xi+ XPS spectrometer (Al Ka radiation-1486.6 eV).Transmission electron microscopy (TEM)TEM images were obtained on a Talos F200X and JEM 2100 instrument operating at an acceleration voltage of 200 kV.Aberration-corrected scanning transmission electron microscopy (AC-STEM)AC-STEM images and energy-dispersive X-ray measurements (STEM-EDS elemental maps) were collected on an ARM-200CF (JEOL, Tokyo, Japan) operating at a voltage of 200 kV.X-ray absorption spectra (XAS)X-ray absorption spectra (XAS), including X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) data of the catalysts, were collected at the XAS station (BL14W1) of the Shanghai Synchrotron Radiation Facility (SSRF). The Ni K-edge (~8333 eV) and Pd K-edge (~24350 eV) XAS spectra were collected in transmission mode, using a metallic Ni foil and a Pd foil for energy scale calibration reference. The electron storage ring was operated at 3.5 GeV, and a Si (111) crystal was used as the monochromator.Electrochemical experiments in three-electrode H-type CellAn electrochemical workstation (BioLogic EC-Lab) with a traditional three-electrode H-cell was used for electrochemical measurements. HMFOR performance test was conducted in a three-electrode H-type Cell, with a Pt-foil as the counter electrode and Hg/HgO as the reference electrode, and the voltages were 95% iR-corrected. Catalyst inks were prepared by dispersing a catalyst (Pd/Ni(OH)2-5.0 mg, Pd/Ni(OH)2-etching-5.0 mg, Pd/C-11.11 mg) in a mixture of isopropanol (0.5 mL) and pure water (0.45 mL) and Nafion (0.05 mL) followed by sonication for 0.5 h. Subsequently, 300 µL of catalyst ink was dropped onto the carbon paper (1 × 0.5 cm2, 3.0 mg cm2). After drying at room temperature, the catalyst-covered carbon paper was utilized as the working electrode in HMFOR measurements. The pH of the electrolytes of 1 M KOH + 50 mM HMF solution was 13.8 ± 0.2 (determined by pH meter). Chronoamperometry tests were performed at different working potentials in an Ar-saturated 1 M KOH + 5 mM HMF solution (10 mL). The cathode and anode were separated by a proton exchange membrane (PEM, Nafion 117). PEM was suitable for a wide pH range (0 ~ 14), and we found through controlled experiments that it was more effective than anion exchange membranes (AEM) in preventing HMF and products from crossing the membrane to the cathode. Moreover, the theoretical maximum transfer coulomb amount was 28.95 C and would not cause a change in pH value. The curves of charge consumption (Q) versus reaction time (t) at each working potential were collected until the reaction finished. An inert gas (Ar) was purged through the electrolyte during the chronoamperometry test to avoid the oxygen reduction reaction (ORR) side reaction. The volume of the reaction solution was measured and collected after the performance test.Two-electrode flow cellAn electrochemical workstation (BioLogic EC-Lab) was used for two-electrode flow cell measurements. For the HMFOR performance test in the two-electrode flow cell reaction, the cathode and anode were separated by a proton exchange membrane (Nafion 117), and the voltages were 95% iR-corrected. Commercial 20 wt% Pt/C (10.0 mg) was sprayed on carbon paper (2.0 × 2.0 cm2, 2.5 mg/cm2) and used as the cathode electrode. An inert gas of Ar was purged through the cathodic electrolyte to avoid unwanted competition from the ORR. The electrolyte of 1.0 M KOH was circulated through the cathode chamber by a peristaltic pump. The Pd/Ni(OH)2 catalyst loaded on a carbon felt was utilized as the anode in the two-electrode flow cell. Catalyst inks were prepared by dispersing Pd/Ni(OH)2 catalyst (100.0 mg) in a mixture of isopropanol (5.0 mL), pure water (5.0 mL), and Nafion (0.3 mL). Subsequently, carbon felt was immersed in catalyst ink and dried at room temperature. The anodic electrolyte of 5 mM HMF in 1.0 M KOH was prepared by mixing 10 mM HMF aqueous solution with 2.0 M KOH before pumping into the anode chamber with a flow rate of 1.0 mL/min. The flow cell voltage was fixed at 0.85 V to drive the cathodic HER and anodic HMFOR. The electrolyte pumping out from the anode was collected, and the product was analyzed using HPLC.Cyclic voltammetry (CV)Before the measurements, the reference electrode (Hg/HgO) was calibrated with the standard saturated calomel electrode (SCE) at open-circuit voltage. The wire of the working electrode was connected to the Hg/HgO electrode, and then the wire of the reference and counter electrode were connected to the SCE, the open-circuit voltage was tested and was closed (<10 mV) to the differentials of the standard electrode potential of the Hg/HgO and SCE electrodes. Catalyst inks were prepared by dispersing catalyst (Pd/Ni(OH)2-5.0 mg, Pd/Ni(OH)2-etching-4.6 mg, Pd/C-9.6 mg, the catalyst mass was calculated according to the Pd loading determined by ICP-OES, to maintain the same quality of Pd on the surface of the working electrode) in a mixture of isopropanol (0.5 mL) and pure water (0.45 mL) and Nafion (0.05 mL) followed by sonication for 0.5 h. Subsequently, 300 µL of catalyst ink was dropped onto the carbon paper (1 × 0.5 cm2) as the working electrode. After complete drying at room temperature, the catalyst-covered carbon paper was utilized as the working electrode in HMFOR measurements, and the voltages were 95% iR-corrected. Potentials were corrected to the RHE using the following equation:$${{{{{\rm{E}}}}}}({{{{{\rm{RHE}}}}}})={{{{{\rm{E}}}}}}({{{{{\rm{Hg}}}}}}/{{{{{\rm{HgO}}}}}})+0.0591\times {{{{{\rm{pH}}}}}}+0.098$$
(1)
The pH value of the HMFOR electrolyte was measured to 13.8 using a pH meter. Ar gas was purged through the electrolyte to avoid unwanted competition from the ORR. HMFOR polarization curves were obtained in 1 M KOH + 0.05 M HMF by continuous CV at a sweep rate of 5 mV s−1 by using a BioLogic EC-Lab workstation.In-situ electrochemical impedance spectroscopy (in-situ EIS)An electrochemical workstation (BioLogic EC-Lab) with a traditional three-electrode H-cell was used for in-situ electrochemical impedance spectroscopy measurements. Work electrodes for in-situ EIS measurements were obtained using the same method as cyclic voltammetry measurements. An Ar gas was purged through the electrolyte to avoid unwanted competition from the ORR. The in-situ electrochemical impedance spectroscopy (in-situ EIS) data for the as-prepared catalysts were collected in the frequency range of 0.1–100,000 Hz in 1 M KOH solution with/without 0.05 M HMF or HMFCA.CO-strippingThe electrochemical surface areas (ECSAs) of the different catalysts were estimated using CO-stripping experiments. The CO-stripping measurements involved forming a monolayer of adsorbed CO on the electrocatalysts by holding the electrode potential at 0.1 V (vs. RHE) while bubbling 100% CO through the 1 M KOH electrolyte for 30 min. After that, the working electrode was characterized by cyclic voltammetry for two continuous cycles from 0 V to 1.20 V (vs. RHE) in a pure 1 M KOH solution without added CO, and the voltages were 95% iR-corrected by employing BioLogic EC-Lab workstation.In-situ FTIR spectraElectrochemical in-situ Fourier transform infrared (FTIR) reflection spectroscopy measurements used a liquid-nitrogen-cooled MCT-A detector (Nicolet-8700 spectrometer). The IR beam was passed through a thin solution layer between the working electrode and a CaF2 window, allowing both adsorbed and dissolved species to be detected. The catalyst (2.0 mg) was dispersed in an ethanol-water mixture (1:1) under sonication for 0.5 h. 10.0 μL of the obtained ink was applied to a carbon paper working electrode, after which 5.0 μL of a 0.25 wt% Nafion solution was applied. A carbon rod was used as the counter electrode and a Hg/HgO as the reference electrode, respectively. Before the electrochemical in-situ FTIR measurements in 1 M KOH + 0.05 M HMF, the working electrode was first electrochemically cleaned until stable in a N2-saturated 1 M KOH solution. Multi-stepped FTIR spectroscopy (MS-FTIRS) was used to collect spectra from 0.1 V to 1.1 V (vs. RHE) at 0.1 V intervals by using a 263 A potentiostat/galvanostat (EG&G) workstation.Product analysisCatalyst inks were prepared by dispersing a catalyst (Pd/Ni(OH)2-10.0 mg, Pd/Ni(OH)2-etching-9.2 mg, Pd/C-19.2 mg (the catalyst mass was calculated according to the Pd loading determined by ICP-OES, to maintain the same quality of Pd on the surface of the working electrode) in a mixture of isopropanol (1.0 mL) and pure water (0.90 mL) and Nafion (0.10 mL) followed by sonication for 0.5 h. Subsequently, 2 mL of catalyst ink was dropped onto the carbon paper (2 × 1.5 cm2) as the working electrode. After drying at room temperature, the catalyst-covered carbon paper was utilized as the working electrode in HMFOR measurements. A continuous inert gas (Ar) was purged through the electrolyte to avoid unwanted competition from the oxygen reduction reaction (ORR). Moreover, suppose the oxygen was not removed cleanly. In that case, the oxygen activation oxidation of HMF on the Pd surface will occur, rather than the electrocatalytic oxidation process, which will cause problems in product detection results and Faraday efficiency calculations. The HMFOR measurements under different oxidation potentials were conducted, and the voltages were 95% iR-corrected by employing a BioLogic EC-Lab workstation. Cycle stability of Pd/C and Pd/Ni(OH)2-etching for HMFOR at 0.75 V versus RHE. were repeated independently three times, and Pd/Ni(OH)2 was performed independently five times at 0.75 V versus RHE. After a certain period of electrocatalysis, the concentrations of organic compounds in the electrolytes were analyzed by HPLC (LC-20AD, Shimadzu) equipped with a photo-diode array (PDA) detector with a detector wavelength of 265 nm. Samples were prepared for HPLC: 50 µL of electrolyte and 950 µL of an aqueous 5 mM ammonium formate solution (ammonium formate/methanol = 7:3) were mixed. A 5 µm C18 column (WondaSil C18-WR, 4.6 × 250 nm) was used to separate different organic compounds, with the analysis requiring 10 min/sample. The mobile phase component A (70%) was a 5 mM ammonium formate aqueous solution, and component B (30%) was methanol with a total flow rate of 0.6 mL/min (Supplementary Fig. S56).The Faradaic efficiency (FE) of HMFOR was calculated as follows:$${{{{{\rm{FE}}}}}}(\%)=100\%\times \frac{{charge\; for\; product}}{{total\; charge\; passed}}=100\%\times \frac{n\times m\times 96485{{{{{\rm{C}}}}}}\,{{{{{{\rm{mol}}}}}}}^{-1}}{{total\; charge\; passed}}$$
(2)
where n is the number of electron transfers from HMF electrooxidation to each product for DFF and HMFCA (n = 2), FFCA (n = 4), FDCA (n = 6); m is the mole of each product; 96485 C mol−1 is the Faraday constant.$${{{{{\rm{Selectivity}}}}}}(\%)=100\%\times \frac{{mol}{es\; of}a{certain\; product}}{{moles\; of\; all\; the\; detected\; products}}$$
(3)
$${{{{{\rm{Y}}}}}}{{{{{\rm{ield}}}}}}(\%)=100\%\times \frac{{mol}{es\; of}a{certain\; product}}{\,{initial\; moles\; of\; HMF}}$$
(4)
DFT calculationsAb initio density functional theory calculations were carried out using the Vienna ab initio simulation program (VASP)41. All calculations were carried out using a plane-wave cutoff of 400 eV. The exchange and correlation energies were described by the Perdew–Burke–Ernzerhof (PBE) form of the generalized gradient approximation42, with dispersive interactions modeled by Grimme’s D3 correction43. The self-consistent total energy convergence criteria was less than 10−5 eV per atom, and atom positions were optimized until the Hellman–Feynman force on each atom was smaller than 0.03 eV/Å. A vacuum layer of 15 Å between periodically repeated slabs was set to avoid interactions among the surface and its periodic images. Kinetic barriers were computed using climbing-image nudged elastic band (CI-NEB) and dimer methods with the same optimization criteria44,45. Transition states were confirmed through frequency analysis to ensure only one imaginary frequency existed.The Pd/Ni(OH)2 interface was constructed using periodic boundary conditions based on optimized Pd and Ni(OH)2 bulk structures. A (5 × 5) supercell of Ni(OH)2 (001) slab model with one layer was adopted as the support for loading a Pd20 cluster with exposed (111) surface facets (Supplementary Fig. S40). The Pd20 cluster and Ni(OH)2 were fully relaxed during Pd/Ni(OH)2 optimization. The Pd(111) surface slab model was built with a (5 × 5) supercell and four layers. The upper two Pd(111) layers were relaxed, and the bottom two were fixed at the bulk geometry in structure optimization. The Brillouin zone was sampled using a 3 × 3 × 1 k-point mesh in the Monkhorst–Pack setups for the Pd/Ni(OH)2 and Pd(111) calculations46.The Gibbs free energy of the species involved in HMFOR was defined as$$\Delta G={\Delta E}_{{{{{{\rm{DFT}}}}}}}+{\Delta E}_{{{{{{\rm{ZPE}}}}}}}-T\varDelta S$$
(5)
where ∆EDFT, ∆EZPE, and ∆S denote the DFT-calculated electronic energy difference between reactants and products of reactions, the zero-point energy correction, and the vibrational entropy at room temperature (T = 298.15 K), respectively. The steps involving an electron transfer from OH− to e− in HMFOR, the corresponding free energy change was calculated based on the standard hydrogen electrode method proposed by ref. 47. The free energy of OH− was derived as G(OH−) = G(H2O(l)) − G(H+), where G(H+) = 1/2 G(H2(g)) − kBTln 10 × pH. The pH was set to 14, and kB is the Boltzmann constant.
Enhanced electrocatalytic biomass oxidation at low voltage by Ni2+-O-Pd interfaces
