Abstract: Abstract: Salt expansion and frost heave have posed a great threat on the engineering buildings and structures that constructed on the saline soil in cold and arid regions. The ice formation and salt crystallization often occur in the pore solution of saline soil, particularly in the large engineering projects. The phase transition of pore solution can greatly contribute to fully understanding the contribution of frost heave and salt expansion in cryogenic deformation. However, it is still lacking on the key eutectic temperature of pore solution for the phase transition of saline soil. Taking the saline loess in Datong basin as an experiment object, this study aims to investigate the temporal variation in the eutectic temperature of pore solution in the saline soil during cooling. Two types of salt (sodium chloride and sodium sulfate) were then chosen in the experiment. A series of cooling tests were conducted to explore the eutectic temperature of saline under different salt contents and water contents, according to the salt properties. The loess was firstly desalinized by the distilled water to remove the artifacts of initial salt. After dried and crushed, the desalinized soil was mixed with the solution to obtain the remolded saline soil with different water and salt contents. The range of salt content was 2.5% to 7%, and the initial water content was taken as 10% to 30%. A cold bath (model: TMS 8035-R40) was used to control the soil temperature. The soil temperature was measured by a thermal-susceptible resistance sensor with a precision of ±0.05oC, and then collected by CR300 at intervals of 10 s during cooling. As such, the phase transition temperature was obtained using the temperature mutation in the cooling curve. Combined with the phase diagram of solution, the physical meaning of temperature mutation was determined to analyze the eutectic temperature. The results show that the eutectic temperatures of sodium chloride soil and sodium sulfate soil were lower than -24℃ and -1.25℃, respectively, which were varied with the water content and salt content. In sodium chloride soil, the eutectic temperature increased with the increase of salt content, while the same trend was found under different water content conditions. In sodium sulfate soil, the eutectic temperature decreased with the increase of salt content, where the eutectic temperature varied more outstanding under the condition of low water content. Since the eutectic temperature of saline soil was related to the liquid water content at the eutectic point, both the precipitation of ice and salt hydrate crystal were posed a great impact on the eutectic temperature. The more ice crystals or hydrated salt precipitates in the first phase transition stage were, the greater temperature deviation of the eutectic temperature between the pore solution and bulk solution was. Moreover, a theoretical model was proposed to predict the variation of the eutectic temperature using the physical chemistry and porous medium. A compelling mechanism was then established for the variation in the eutectic temperature under different salt contents and water contents, in order to clarify the effects of ice formation and hydrated salt crystallization on the eutectic temperature. An influence mechanism was also proposed for the supersaturation degree. The finding can be of great theoretical significance to further understanding the phase transition mechanism of pore solution in saline soil. Anyway, an effective way can also be offered to determine the contribution of frost heave and salt expansion of saline in the cooling process.