大学物理实验中常用稳态法、准稳态法、常功率热源法测量材料常温与高温条件下的导热系数，而对于低温条件下材料的导热系数，大学物理实验中并未涉及.基于第一类边界条件下的一维半无限大物体的非稳态导热理论，以液氮为恒温冷源，设计了一套在宽低温范围内连续测量材料导热系数的装置，为材料低温导热系数的测量提供思路.以测量纯水所制冰柱的低温导热系数为例，详细阐述实验原理、方法与装置并分析了测量数据，经过修正测得冰在-10.49 ℃至-126.70 ℃范围内的导热系数. 由于修正模型的偏差、装置的漏热等问题，导致导热系数随温度的变化率高于参考值，后续希望通过实验方法与装置的优化，以提升测量的精确度.

In college physics experiments, steadystate method, quasisteadystate method, and constant power heat source method are commonly used to measure the thermal conductivity of materials under normal and hightemperature conditions. However, the measurement of thermal conductivity of materials under lowtemperature conditions is not typically covered in college physics experiments. Based on the unsteady heat conduction theory for onedimensional semiinfinite objects under the firsttype boundary condition, using liquid nitrogen as a constanttemperature cold source, a device was designed to continuously measure the thermal conductivity of materials at low temperatures within a wide range，which provide a new approach for measuring the thermal conductivity of materials at low temperatures. Taking the measurement of the lowtemperature thermal conductivity of ice columns made of pure water as an example, the experimental principle, method, and device were elaborated in detail, and the measurement data were analyzed. The thermal conductivity of ice was measured within the temperature range of 10.49°C to 126.70°C. Due to the deviation of the correction model and the heat leakage of the device, the rate of change of thermal conductivity with temperature is higher than the reference value. It is hoped that further optimization of the experimental methods and devices will be carried out to improve the accuracy of experimental measurements.