•A Co-N-graphene catalyst composed of CoN4-macrocyclic-like (CoNx) structure is synthesized.•Co-Nx-Graphene has effective electrocatalytic activity for Li/SOCl2 batteries.•The storage stability of the catalyst is attributed to its insolubility in electrolyte.A mixture of cobalt phthalocyanine (CoPc) and graphene is thermally decomposed at 800 °C to synthesize a novel catalyst. Scanning electron microscopy (SEM) and transmission electron microscope (TEM) show that the catalyst retains the lamellar structure of graphene. X-ray diffraction (XRD) reveals that the catalyst is no longer composed of CoPc and high-resolution TEM (HRTEM), X-ray photoelectron spectra (XPS) prove that Co and N elements have entered the graphene molecular structure, thus forming a Co-Nx-graphene (Co-Nx-G) catalyst composed of a CoN4-macrocyclic-like structure. This catalyst serves as an excellent catalyst of thionyl chloride (SOCl2) reduction. Cyclic voltammetry and battery discharge tests reveal that Co-Nx-G-800 substantially increases the discharge voltage and capacity of a Li/SOCl2 battery. Moreover, Co-Nx-G-800 exhibits stable catalytic activity during battery storage. Ultraviolet–visible spectroscopy shows that CoPc is soluble in a SOCl2 electrolyte solution, whereas Co-Nx-G-800 is not, this characteristic contributes to the stable catalytic property of Co-Nx-G.

•A mathematical life prediction model for lithium thionyl chloride (Li/SOCl2) batteries was established.•The model was parameterized by means of multiple non-linear curve fitting method based on Universal Global Optimization.•Battery’s storage lifetime, self-discharge rate and acceleration factor can be calculated on the basis of the model.Battery life prediction is critical for lithium/thionyl chloride cells with a long storage life. The objective of this study was to develop models for rapidly estimating the storage life of Li/SOCl2 cells using the semiempirical approach. An accelerated degradation test involving numerous cells stored at various temperatures (room temperature or RT, 40, 50, 60, and 70 °C) was conducted to investigate the effect of the storage time and temperature on capacity degradation. The degradation law can be summarized on the basis of the test data for constructing the semiempirical equation; this law demonstrates that the residual capacity of aging cells exponentially changes with the storage time and temperature. The degradation data are also used for parameterization with the multiple nonlinear curve fitting method based on the universal global optimization algorithm. According to the simulation and comparison between the experimental data and prediction curve, the fitting prediction curve accurately fits the experimental data. This finding indicates that the semiempirical model is useful because of its satisfactory ability to approximate the measured data. In addition, characteristic values of the battery, including the storage life under various storage conditions, average self-discharge rate, and acceleration factor, can be calculated on the basis of the mathematical model.

•Accelerated degradation test was performed on Li/SOCl2 cells.•The prediction model for residual capacity was established by using nonlinear curve fitting method based on the least square calculation.•Empirical capacity prediction model was found valid for a wide region of aging condition.The empirical prediction model of residual capacity (Cap) for D-size Li/SOCl2 cells has been developed and validated based on the accelerated degradation test (ADT) data. In this experiment, a series of constant storage temperatures (25 °C, 55 °C, 70 °C, and 85 °C) was selected and the residual capacity of each cell was monitored continuously during the aging test. The model was established by fitting twice. Firstly, time dependence of Cap (t, T) was investigated. Secondly, the generalized model of residual capacity was built. The prediction model, as a function of storage time and temperature, can precisely predict the value of residual capacity. The generalized empirical model of Cap, involving two aging processes, is valid for the degradation condition of temperatures from 25 °C to 70 °C. The first aging process completed rapidly within 7 days. The second aging process was accelerated by temperature with time1/2 kinetics. For the cells stored at 85 °C, another failure mechanism may exist based on the departure of linear fitting coefficients.

•Micro-spherical FeS2/CNT sample is prepared by one-step hydrothermal technique for the first time.•CNT can restrain the growth of marcasite during the hydrothermal process.•CNT can help FeS2 form dispersed micro-spherical particles and form a conductive matrix on the FeS2 particle surface.•FeS2/CNT electrode shows superior rate capability and cycling performance compared with FeS2.Micro-spherical FeS2/CNT powder was synthesized through one-step hydrothermal method for the first time. Carbon nanotube (CNT) had a great influence on the structure, morphology and electrochemical performance of FeS2/CNT powder. X-ray diffraction (XRD) patterns have demonstrated that the existence of CNT can restrain the growth of marcasite during hydrothermal process. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) test have shown that CNT help to form monodisperse micro-sphere particles. Comparing with untreated FeS2, FeS2/CNT electrode exhibits superior rate and cycling performance. The FeS2/CNT powders can sustain 491 and 370 mAh g− 1, much higher than FeS2 prepared without CNT (265 and 74 mAh g− 1) after 50 cycles at 0.1 and 1 C (1 C = 890 mA g− 1).