Gallium Nitride Based Electrode for High‐Temperature Supercapacitors

Abstract Gallium nitride (GaN) single crystal, as the representative of wide‐band semiconductors, has great prospects for high‐temperature energy storage, of its splendid power output, robust temperature stability, and superior carrier mobility. Nonetheless, it is an essential challenge for GaN‐based devices to improve energy storage. Herein, an innovative strategy is proposed by constructing GaN/Nickel cobalt oxygen (NiCoO2 ）heterostructure for enhanced supercapacitors (SCs). Benefiting from the synergy effect between the porous GaN network as a highly conductive skeleton and the NiCoO2 with massive active sites. The GaN/NiCoO2 heterostructure‐based SCs with ion liquids electrolyte are assembled and delivered an impressive energy density of 15.2 µWh cm−2 and power density, as well as superior service life at 130 °C. The theoretical calculation further explains that the reason for the energy storage enhancement of the GaN/NiCoO2 is due to the presence of the built‐in electric fields. This work offers a novel perspective for meeting the practical application of GaN‐based energy storage devices with exceptional performance capable of operation under high‐temperature environments.

(Shanghai). All the reagents were used as received.

S2.1 Preparation of porous GaN single crystal:
GaN crystal was grown by metal organic chemical vapor deposition. Porous GaN membrane single crystal was fabricated by a classical electrochemical etching processing. Specifically, two electrode system that the 0.3 M oxalic acid solution as electrolyte was applied. The GaN crystal and platinum wire as anode and cathode, respectively. The etching process was controlled by voltage parameter. Firstly, the pulse voltage is kept constant at 5 V with 10 min, and then the voltage is converted to 15 V with 3 min.

S2.2 Preparation of the GaN/NCO heterostructure:
The NiCo precursor in situ growth on the surface of porous GaN membrane by a strep hydrothermal method. A series of porous GaN based heterostructures were prepared by using the molar ratios of porous GaN to nickel/cobalt chloride hexahydrate were 1: 0: 0, 1: 1: 1, 1: 3: 3, 1: 6: 6 and 0: 1: 1, they are labeled as GaN, GaN/NCO-1, GaN/NCO-2, GaN/NCO-3 and NCO, respectively. For the GaN/NCO-2, cobalt chloride hexahydrate (0.065 g), nickel chloride hexahydrate (0.065 g) and urea (0.1 g) were dissolved in 40 mL deionized water during continuous stirring. Afterward, carefully transfer the homogeneous solution and a piece of porous GaN membrane (approximately 7 mg) to 50 ml autoclave for heating 130 ℃ with 6 hours. The GaN/NiCo-precursor was washed and dried at 80 ℃ in a vacuum oven. Finally, the GaN/NiCo-precursor sample was put into quartz boat and annealed in argon atmosphere at 350 ℃ for 2 hours with the heating rate of 5 ℃/min to obtain the porous GaN/NCO product.

S2.3 Fabrication of the porous GaN based heterostructure electrode and devices
with ionic liquid as electrolyte: The active materials (GaN/NCO heterostructure), PVDF binders, acetylene blacks with a weight ratio of 80: 10 :10 were mixed in N-methylpyrrolianone (NMP) to prepared the electrode. Then the evenly slurry was spreaded on the current collector (stainless steel cloth with 1 × 1 cm) that was cleaned and completely dried for 10 hours in a vacuum. The mass loading of the electrode was approximately 1 mg cm -2 .
GaN based heterostructure SCs were assembled in an argon filled glove box (< 0.01 ppm of oxygen and water). The porous GaN/NCO heterostructure served as anode and cathode respectively and the 1-ethyl 1-3 methimidazole (trifluoromethyl sulfonyl) imide (EMImNTf 2 ) used as electrolyte.
The convergence criteria of force and energy for the structural relaxation were set at 0.01 eV/Å and 10-5 eV. To obtain the accurate electronic structure, All the electronic properties of GaN, NCO, and GaN/NCO are determined using the screened Heyde-Scuseriae-Ernzerhof (HSE) hybrid functional. [4,5] Monkhorst-Pack k-points [6] were sampled using a 2×2×2 for electronic calculations and 4×4×1 for the relaxation and total energy. The adsorption energy (E ad ) for species X is defined as follows: E ad is the total energy of the fully relaxed surface/adsorbate system, E(surface) is the total energy of the relaxed substrate slab, and E(X) is the total energy of a free adsorbate species. Negative adsorption energy means effective adsorbate binding.

S3. Characterization Methods
The morphology of the as- The area specific capacity of the electrodes in three-electrode test system, is derived from the GCD curve, calculated according to equation 1.
Where C is the specific capacitance (F cm -2 ) base on the area of the electrode materials, I is the discharge current (A), Δt is the discharge time (s), S is the area (cm 2 ) of the active materials of the single electrode, ΔV is the working voltage window (V).
The area specific capacitance for GaN/NCO based SC is calculated from Where C is the area specific capacitance (F cm -2 ) of the device, I is the discharge current (A), Δt is the discharge time (s), S is the total area (cm 2 ) of active material on the SC, ΔV is the working voltage window (V).
The energy density of the SCs are calculated by equation 3 Where E is the energy density (mWh cm -2 ) of the device, C is the area specific capacitance (mF cm -2 ) of the device, ΔV is the working voltage window (V).

S5. Calculation of Capacitance Contribution
The b value from i=av b can be obtained by plotting current and sweep rate in logarithm (Equation 6), namely, the gradient of linear plots.

logi= b log v + log a
In order to determine the k1 value, Equation 7 can be reformulated as: