Promotion of Overall Water Splitting Activity Over a Wide pH Range by Interfacial Electrical Effects of Metallic NiCo‐nitrides Nanoparticle/NiCo2O4 Nanoflake/graphite Fibers

Abstract Many efforts have been made to develop bifunctional electrocatalysts to facilitate overall water splitting. Here, a fibrous bifunctional 3D electrocatalyst is reported for both the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER) with high performance. The remarkable electrochemical performance is attributed of the catalysts to a number of factors: the metallic character of the three components (i.e., Ni3N, CoN, and NiCo2O4); the electronic structure, nanoflake‐nanosphere network with abundant electroactive sites, and the electric field effect at the interfaces between different components. The oxide–nitride/graphite fibers have the lowest overpotential requirements of 71 and 183 mV at 10 mA cm−2 for HER and OER in alkaline medium, respectively. These values are comparable to those of commercial Pt/C (20 wt%) and RuO2. The electrodes also show a response to HER and OER in both neutral and acid media. Furthermore, the 3D structure can be highlighted by all‐round electrodes for overall water splitting. The calculations on the changes in electrons transfer and the Femi level from oxides to oxides/nitrides reveal that the observed superb electrocatalytic performance can be attributed to the presence of Ni3N and CoN derived from the in situ nitridation of NiCo2O4.


Electrochemical tests
The electrochemical measurements were performed on a CHI660E electrochemical workstation with a three-electrode cell.

Electrochemical active surface area
The capacitive currents are measured in a potential range where no faradic processes are observed. We sweep the potential between 0.1~0.2 V vs RHE at different scan rates. The difference in current density variation( j=j a -j c ) at the potential of 0.15 V vs RHE plotted against scan rate are fitted to estimate the electrochemical double-layer capacitances (C dl ), which can be used to estimate the electrochemically active surface area (EASA).

Faradic efficiency
The Faradic efficiency reflects the utilization efficiency of electron in HER and OER process. For OER (or HER) process, the Faradic efficiency can be obtained by calculating the radio of the experimentally produced O 2 (H 2 ) amount(n O2 ) to the theoretical produced O 2 amount(n O2' ). Specifically, under a constant oxidation current (I) within a certain time (t), the experimentally produced O 2 amount can be measured by Automatic online trace gas analysis system (Labsolar 6A supplied by Beijing Perfectlight Technology Co., Ltd.)-gas chromatography (Perfect Light GC7806). Thus, the Faradic efficiency can be calculated as following: Faradic efficiency= n O2 / n O2' =4F n O2 /It (or Faradic efficiency= n H2 / n H2' =4F n H2 /It

5.Computational formulas:
Our first-principles calculations were performed within density-functional theory (DFT) using the Vienna ab initio simulation package known as the VASP code. [1] The projector augmented wave method (PAW) [2] was used to describe the electronic-ion interaction. The energy cutoff of the plane waves was set to 450 eV with an energy precision of 10 -5 eV. The electron exchange-correlation function was treated using a generalized gradient approximation (GGA) in the form proposed by Perdew, Burke, and Ernzerhof (PBE). [3] The method of local density approximation (LDA)+U, The U value for Co and Ni taken from the previous work are 6.7 and 7.1 respectively. [4] , The Monkhorst-Pack [5] k-point meshes for the Brillouin zone (BZ) sampling are well converged for each system . Both atomic positions and lattice vectors were fully optimized using the conjugate gradient (CG) algorithm until the maximum atomic forces were less than 0.01 eV/Å. The equilibrium lattice constant for each system is donated in Figure S18.

S1 SEM image and XRD pattern of bulk CoN
Figure S1a shows that the XRD pattern of prepared CoN was well indexed to the characteristic peaks of CoN (JPCDS No.83-0831). Furthermore, the morphology of CoN consisted of nanoparticles investigated by SEM in Figure S1b. The prepared CoN was used as standard reference sample in XAFS measurements Figure S2a shows that the XRD pattern of prepared Ni 3 N was well indexed to the characteristic peaks of Ni 3 N (JPCDS No.89-5144). Furthermore, the morphology of Ni 3 N consisted of nanoparticles investigated by SEM in Figure S2b. The prepared Ni 3 N was used as standard reference sample in XAFS measurements. Figure S3 The Raman spectrum of bare graphite fibers. As is shown in Figure S3 Figure S5 shows the SEM image of NiCo 2 O 4 on graphite fibers after nitrogenation for 4h. Clearly, the nanoflakes mostly melt after 4h in NH 3 atmosphere and fell off the graphite fibers. Thus, we choose nitrogenation for 2h reasonably.

S6 The cross-sectional views after electrodeposition.
From the Figure S6, we can see clearly that the nanoflakes were tightly attached on the graphite fibers forming 3D nanostructures. Furthermore, the thickness of ~500 nm was demonstrated here.

S7 TEM image of NiCo 2 O 4 derived from the NiCo 2 O 4 /GF sample.
The

S9 XPS peak of N1s spectra of NiCo-nitrides/ NiCo 2 O 4 /GF on graphite fibers
For N1s spectrum in Figure S9, the characteristic peak was located at around 399 eV, which can be assigned to the nitrogen in a metal nitride.

Figure S10
Chronopotentiometry of the NiCo-nitrides/NiCo 2 O 4 /GF and NiCo 2 O 4 /GF for HER. Figure S10 show the results of a stability test conducted for the

S11 SEM image of NiCo-nitrides/NiCo 2 O 4 /GF after 40h by continuous chronoamperometric response (i-t) under alkaline condition
The morphology of the NiCo-nitrides/NiCo 2 O 4 /GF hardly changed, which indicates the good stability of the NiCo-nitrides/NiCo 2 O 4 /GF.

S12 Equivalent circuit of EIS
The equivalent circuit consisted of a resistor (Rs) in series with two parallel combinations of a resistor (R1, Rct) and a constant phase element (CPE1, CPE2). This was used to fit the EIS data that can be evidenced by the semicircles in the high-and low-frequency range. R s and R ct were determined by electrocatalytic kinetics, may result from the Ohmic resistance arising from the electrolyte as well as the all contact, and charge transfer resistance at the interface between the catalysts and the electrolyte, respectively. It is known to us that small values of R s correspond to close contact between current collector and catalysts, and that small values of R ct endow the electrocatalysts with rapid charge transfer kinetics.

S13 Chronopotentiometry of the NiCo-nitrides/NiCo 2 O 4 /GF and NiCo 2 O 4 /GF for OER
The stability test of the NiCo-nitrides/NiCo 2 O 4 /GF and the NiCo 2 O 4 /GF in Figure S13 was also tested by continuous chronoamperometric response (i-t) under alkaline condition at the applied potential of 1.41 V (vs RHE) and 1.60V (vs RHE), respectively. They show almost negligible degradation during 40 h of continuous operation, which confirms the excellent durability.

NiCo-nitrides/NiCo 2 O 4 /GF for overall water splitting in 1M KOH
The durability test was maintained at 10mA/cm 2  As we can see in Figure S15, the NiCo-nitrides/NiCo 2 O 4 /GF delivered a higher C dl (114.4 mF cm -2 ) than NiCo 2 O 4 /GF (90.13 mF cm -2 ), which is proportional ECSA. As a function of previous reports [6] , the activity enhancement arose from the increased It is well accepted that the Femi level would change in the effect of electric potential. [1][2][3][4][5][6]  Interface electric field effect at the interface of Ni 3 N/CoN and NiCo 2 O 4 will benefit for electrons transfer during the electrochemical reactions. As can be seen from the