Oxygen‐Bridged Cobalt–Chromium Atomic Pair in MOF‐Derived Cobalt Phosphide Networks as Efficient Active Sites Enabling Synergistic Electrocatalytic Water Splitting in Alkaline Media

Abstract Electrochemical water splitting offers a most promising pathway for “green hydrogen” generation. Even so, it remains a struggle to improve the electrocatalytic performance of non‐noble metal catalysts, especially bifunctional electrocatalysts. Herein, aiming to accelerate the hydrogen and oxygen evolution reactions, an oxygen‐bridged cobalt–chromium (Co‐O‐Cr) dual‐sites catalyst anchored on cobalt phosphide synthesized through MOF‐mediation are proposed. By utilizing the filling characteristics of 3d orbitals and modulated local electronic structure of the catalytic active site, the well‐designed catalyst requires only an external voltage of 1.53 V to deliver the current density of 20 mA cm−2 during the process of water splitting apart from the superb HER and OER activity with a low overpotential of 87 and 203 mV at a current density of 10 mA cm−2, respectively. Moreover, density functional theory (DFT) calculations are utilized to unravel mechanistic investigations, including the accelerated adsorption and dissociation process of H2O on the Co‐O‐Cr moiety surface, the down‐shifted d‐band center, a lowered energy barrier for the OER and so on. This work offers a design direction for optimizing catalytic activity toward energy conversion.

(China).All the reagents and chemicals were of analytical grade and used directly without further purification.

Preparation of Co-MOF precursor
The cobalt-based metal-organic framework supported on nickel foam (NF) was prepared according to the previous report. [1]In a typical process, the commercial NF was pre-washed successively with ethanol, deionized water, and ethanol, each for 15 min, to ensure a clean surface.The bare Ni foam was first immersed into 2-methylimidazole aqueous solution (0.4 M, 40mL) overnight and then immersed into the mix solution of Co 2+ and 2-methylimidazole for another 6 hours.In the end, a uniform purple layer could be observed on Ni foam.The growth mechanism of Co-MOF is therefore believed to be as follows: 2-methylimidazole molecules are initially adsorbed on the Ni foam surface and then coordinate with Co 2+ ions in the aqueous solution to form ZIF-Co.

Fabrication of Cr-Co-MOF precursor
A piece of as-fabricated Co-MOF-LN were immersed into an aqueous ethanol solution (100 mL) containing Co(NO3)2 and Cr(NO3)3 and kept stationary to allow an in-situ conversion from Co-MOF to Cr-Co-MOF composite on the surface of NF substrate, where Cr(NO3)3 accounts for 0~20% of the total number of moles of Co(NO3)2 and Cr(NO3)3.After a reaction of 15~45 min under room temperature, the resulting product was washed with deionized water and absolute ethanol for three times, respectively, and then dried at 60 o C to provide a Cr-Co-MOF sample.

Preparation of Co-P, Cr-Co-P
To synthesize Co-P, Cr-Co-P, the corresponding precursors and 800 mg of NaH2PO2 were added into two crucibles, respectively, and NaH2PO2 was placed at the upstream side of gas flow.They are then heated for a phosphorization at 300 °C in Ar flowing for 1 h with a heating rate of 2 °C min -1 .After cooling down in an Ar atmosphere, the Co-P, Cr-Co-P were collected.

Characterizations
Field emission scanning electron microscopy (FE-SEM) images were obtained using a FEI Quattro S (ThermoFisher Scientific).TEM images were taken using a FEI Talos F200S (ThermoFisher Scientific) scanning transmission electron microscope with an operating voltage of 200 kV.XPS analysis was carried out using an ESCALAB250Xi system (ThermoFisher Scientific).XRD patterns were obtained from a PANalytical X'Pert Powder using Cu Kα radiation (λ = 1.5406Å).Precision balance instrument (FA224TC, Lichen Instrument Technology Co., Ltd) was used to weigh the loading mass of different catalysts.Raman spectra were carried out on a Raman spectrometer (HORIBA LabRam HR 800).The nitrogen adsorption-desorption isotherm of the sample was measured on a Quantachrome QuadraSorb Station 3 instrument at 77 K.

Electrochemical measurements
Electrochemical measurements were carried out using CHI660E electrochemistry work station in a standard three-electrode system.The working electrode was the Cr-Co-P (and other control samples) with an effective geometric area of 2 cm 2 .Hg/HgO electrode and graphite rod were used as reference electrode and counter electrode, respectively.All electrochemical data was with 90% iR-correction.1M KOH were used as electrolytes.All measured potentials were converted to a reversible hydrogen electrode (RHE) scale using Nernst equation: E(RHE) = E(Hg/HgO) + 0.098V + 0.059 × pH.For HER testing, the linear sweep voltammetry (LSV) measurements were performed at a scan rate of 5 mV•s −1 .Electrochemical impedance spectroscopy (EIS) measurements were conducted at the frequency range from 10 6 to 10 -2 Hz in -0.1V.The cyclic voltammetry (CV) was conducted at various scan rates (10-100 mV•s -1 ).
The electrochemically active surface area (ECSA) was estimated by measuring the capacitive current associated with double-layer charging from scan-rate dependence of CVs in a non-Faradaic region and the scan rates were 10-60 mV s -1 .The multi-step chronopotentiometric curves were obtained by changing the current densities from -10 and 100 mA cm -2 .The stability of the catalyst films was assessed by repeated CV scans at 100 mV•s -1 for 3000 sweeps.For HER processes, the potential ranges were set in 0 V ~ -0.20 V vs. RHE.

Computational studies
The first principles calculations are performed using the Cambridge Serial Total Energy Package (CASTEP) program code based on the plane-wave pseudo-potential method within the framework of density func-tional theory (DFT). [2]The generalized gradient approximation (GGA) is adopted along with the exchange-correlation function realized by Per-dow-Burke-Emzerhof (PBE). [3]A plane wave basis was set up to an energy cutoff of 500 eV for the surface of all catalysts.The convergence tolerance for the residual force on each atom during structure relaxation was set to 0.03 eV/Å and for the energy difference between two consecutive selfconsistent calculations is less than 10 -5 eV, respectively.A sufficiently large vacuum region of 20 Å was used for all the systems to ensure the periodic images were well separated, and the atoms relax during the geometry optimizations.The calculated minimum distance between the different moiety surface was 3 Å.The computational hydrogen electrode (CHE) was used to calculate the free energy of each intermediate state.The H2O absorption energy was calculated using the following equation, ΔEH2O = EsurfH2O -Esurf -EH2O, where EsurfH2O and Esurf are the total energies of the surface covered with and without H2O molecule, EH2O is the energy of free H2O molecule.The overall HER pathway includes two steps: first, adsorption of hydrogen on the catalytic site (*) from the initial state (H + + e -+ *), and second, release of the product hydrogen (1/2H2).The total energies of H + + e − and 1/2H2 are equal.Therefore, the Gibbs free energy of the adsorption of the intermediate hydrogen on a catalyst (GH) is the key descriptor of the HER activity of the catalyst and is obtained by equation: ∆G=∆E+∆EZPE-T∆S, where the ∆E is the adsorption energy of a specific step, and ∆EZPE and T∆S are the difference of zero point energy and the entropy between the adsorbed state and the free-standing state, respectively.In addition, to evaluate the OER activity of the materials, we calculate the Gibbs free energy of coordinate elementary steps and overpotential for OER based on the following 4e --mechanism proposed by Norskov for water oxidation.was previously reported by us [4] ).Table S1.The system resistance (Rs) and charge transfer resistance (Rct) for four samples.

Figure S3 .
Figure S3.(a) XRD patterns of synthesized catalysts, (b) XRD patterns of synthesized Co-MOF peeled off from NF, (c) XPS full spectra of samples (where the result of N,Ni-Co2P@TC

Figure S5 .
Figure S5.HER performance of Co-MOF with (a) different etching times and (b) different Cr doping levels.

Figure S10 .
Figure S10.Dissolution concentration of Co, Cr and Ni in the electrolyte after OWS catalysis for 24 h (analyzed using ICP-OES).

Figure S11 .
Figure S11.The side view and top view of the optimized structure of (a) Co-P and (b) Cr-Co-P (Obri) (Substitution for Co-P site with Cr-O atoms on the outermost surface of Co-P).

Figure S13 .
Figure S13.Adsorption configurations of *M-OH, *M-O and *M-OOH on Co centers from the basal plane of Cr-Co-P (Obri) and Co-P (Obri).