Abstract: The objective of this study was to develop an efficient non-enzymatic electrochemical sensor for H2O2 detection in milk using the catalytic reduction of H2O2 on ZnFeCoO4. Firstly, ZnFeCoO4 derived from ZnCo2O4 by the partial substitution of Co with Fe was synthesized via sol-gel combustion. In ZnCo2O4, 1 mmol Zn(NO3)2•6H2O and 2 mmol of Co(NO3)2·6H2O were dissolved in a nitric acid solution (2 mL of nitric acid, and 30ml of distilled water). The sticky gel was subsequently obtained by adding 15 mmol citric acid and stirring for 5 h in the oil bath at 90℃. After that, the gel was placed in a 170℃ oven (10℃/min heating rate) for 12 h. Then, the moisture-removed gel was crushed with a mortar. Finally, the powder was calcined in a 600 ℃ tubular furnace (5℃/min heating rate) for 6 hours to obtain the spinel phase ZnCo2O4 powder. The preparation process of ZnFeCoO4 was consistent with that of ZnCo2O4. The only difference was that the 2 mmol of Co(NO3)2·6H2O was replaced with 2 mmol of Fe(NO3)2·6H2O. Transmission electron microscopy, X-ray diffraction, and X-ray photoelectron spectroscopy were used to characterize the crystal structure and morphology of the catalyst, while cyclic voltammetry (CV) was used to study the electrochemical performance of the electrode. In order to verify the effectiveness of the ZnFeCoO4 electrode in detecting H2O2 content in milk, the milk from the local supermarket was centrifuged and 1 mL supernatant was dissolved in 10 mL PBS (pH value 7.4). Standard addition was used for the recovery experiment. The results showed that a similar morphology was obtained after the partial or complete replacement of Co with Fe. ZnFeCoO4 presented a better crystal shape and high consistency with the standard database card JCPDS No.23-1390. The mass ratios of Zn, Co, and Fe were measured as 31.46%, 35.77%, and 27.15%, respectively, and the molar ratio was about 1:1.2:1, which roughly met the chemical expression of the compound. The XRD patterns of ZnCo2O4 and ZnFeCoO4 showed that the diffraction peaks of ZnFeCoO4 were basically the same as those of ZnCo2O4, both of which were conformed to the standard database card JCPDS No.23-1390, indicating the spinel phases. In addition, the diffraction peak of ZnFeCoO4 was shifted to a smaller angle than that of ZnCo2O4, because the lattice radius of Fe3+ was slightly larger than that of Co3+. This proved the successful synthesis of spinel phase ZnFeCoO4. The CV analysis showed that the ZnFeCoO4 presented better catalytic activity than the ZnCo2O4, where the electrochemical sensor using ZnFeCoO4 shared the better electrocatalytic performance for H2O2 with a theoretical detection limit of 0.5 μmol/L. There was a linear relationship between the concentration of hydrogen peroxide and the peak reduction current in the range of 0.5-2.5 mmol/L, where the determination coefficient (R2) is 0.989. The stability and selectivity of ZnFeCoO4 electrode analysis showed that there was no significant change in the current density when the electrode was exposed to 1 mmol/L H2O2 solution for 30 days. After one month, the observed current density was about 92.7% of the original current density, with a relative standard deviation of 13.2%, indicating the better stability of the ZnFeCoO4 electrode. The amperage method was implemented to evaluate the effect of ascorbic acid, glucose, sucrose, citric acid, urea, and H2O2 on the current density response. Except for H2O2, there was no significant change in the current density, indicating the better selectivity of the sensor. In addition, the sensor also presented excellent reproducibility for H2O2. Excellent recoveries from 98.60% to 101.23% were obtained for the H2O2-spiked milk. The as-presented sensor can be expected to serve as a promising candidate for the H2O2 determination in food safety.