Thermoporometry, namely revealing pore sizes based on the shifts of the freezing and melting transitions of liquid in the pores, has proven a powerful approach for mesostructure characterization of porous solids. Complementary to gas adsorption, it has several advantages, such as applicability to wet samples. At the same time, while gas sorption characterization has largely profited from the recent advances in the understanding of the gas-liquid phase equilibria in confined spaces, the solid-liquid transitions have remained poorly covered. In this work, in an analogy to the advanced gas sorption analysis, we implement a novel approach in thermoporometry by introducing the transition kernels, i.e. the families of solid-liquid transition curves describing different transition scenarios in single pores. In this way, a variation of the non-frozen layer thickness with temperature and the effects of thermodynamic fluctuations on phase transitions become automatically encoded in the transition kernels. As a next step, we incorporate these kernels into a recently developed theoretical framework, the serially-connected pore model, allowing for the analysis of pore networks beyond collections of independent pores conventionally assumed in the literature. This combination yields a substantial improvement of the accuracy of the thermoporometry method which is further demonstrated by applying NMR cryoporometry to MCM-41 and SBA-15 materials.

热孔法，即根据孔中液体的冻结和熔融转变的位移揭示孔尺寸，已被证明是一种用于表征多孔固体的介观结构的有效方法。作为气体吸附的补充，它具有多种优势，例如适用于湿样品。同时，虽然气体吸收特性在很大程度上受益于对密闭空间中气液平衡的理解的最新进展，但对固液过渡的研究仍然很少。在这项工作中，类似于高级气体吸附分析，我们通过引入过渡核（即描述单个孔中不同过渡情况的固液过渡曲线族）在热孔法中实现了一种新方法。通过这种方式，非冻结层厚度随温度的变化以及热力学波动对相变的影响会自动编码在过渡核中。下一步，我们将这些内核整合到最近开发的理论框架中，即串行连接的孔隙模型，从而可以对孔隙网络进行分析，超出文献中通常假定的独立孔隙集合。这种结合大大提高了热孔法的准确性，这通过在MCM-41和SBA-15材料上应用NMR低温法进一步证明。我们将这些内核整合到最近开发的理论框架中，即串行连接的孔隙模型，从而可以对孔隙网络进行分析，而超出了文献中通常假设的独立孔隙的收集范围。这种结合大大提高了热孔法的准确性，这通过在MCM-41和SBA-15材料上应用NMR低温法进一步证明。我们将这些内核整合到最近开发的理论框架中，即串行连接的孔隙模型，从而可以对孔隙网络进行分析，而超出了文献中通常假设的独立孔隙的收集范围。这种结合大大提高了热孔法的准确性，这通过在MCM-41和SBA-15材料上应用NMR低温法进一步证明。