Salt weathering is a common deterioration phenomenon that affects outdoor cultural properties, and it is important to precisely predict the heat, moisture, and salt transfer in porous materials to suppress salt weathering. Osmosis and osmotic pressure were considered in the field of soil research, especially in clay research, but not in the field of outdoor cultural properties and building materials, which are the main target of salt weathering. Osmosis in clay is supposed to be caused by its surface charge. However, it has been suggested that sandstones and bricks that constitute cultural properties and buildings also have surface charge as clay. Thus, osmosis and osmotic pressure can occur in building materials, which may lead to materials degradation. In this study, we derive basic equations, based on nonequilibrium thermodynamics, for the simultaneous heat, dry air, water vapor, liquid water, cation, and anion transfer in building materials by considering osmosis. This equation was compared with existing model for heat and moisture transfer equations as well as models that considered the salt transfer. Based on the previous research for osmosis in clay, we summarized conditions under which osmosis occurs in building materials and presented an outlook for modeling the physical properties of materials related to osmosis.

Cations in soil solutions affect soil structural stability and thus soil quality and health. Ca2+ and Mg2+ could alleviate soil clay dispersion by replacing Na+, which was the main driver. However, Mg2+ could also cause soil disaggregation and weaken aggregate integrity. Currently, most studies on Ca2+ and Mg2+ mainly examine soil hydraulic characteristics and clay particle dispersion, rather than macropore geometry. We analyzed the impact of Ca2+ and Mg2+ on soil macropore morphology by measuring macropore length, aspect ratio, and area indices. An indoor soil column experiment was set up, and irrigation water was prepared with the same electrical conductivity (4 dS m(-1)) and different cation compositions (Na+-Ca2+ (NC), (Na)(+)-Ca2+-Mg2+ (NCM), Na+-Mg2+ (NM) and Na+-only (N) were added), and deionized water as the control (CK). The results indicated that N had the highest soil suspension turbidity among all treatments, with NM being higher than NC. The highest percentage of soil macropore aspect ratio 2.0. The macropore anisotropy of NM was closer to 1.0, and anisotropy of NC was closer to 0. The soil macropore morphology of NC developed towards a spherical shape, while the macropores of NM might extend along one or several similar directions. For Na+-Ca2+ dominated soil, Ca2+ mainly affected the macropore area. However, for Na+-Mg2+ dominated soil, Mg2+ primarily influenced the number of macropores. Ca2+ inhibited the negative effects of Na+, but Mg2+ promoted unidirectional extension of macropores, posing a risk of soil cracking. This study provided a better understanding to explore the differences in the effects of different cations on soil pore structure and helped to provide guidance for field water management.

Cation ratio of soil structural stability (CROSS) can replace sodium adsorption ratio (SAR) to evaluate the effects of base cations on soil structure. It is generally accepted that in saline water with the same electrical conductivity (EC), a higher CROSS reflected a greater reduction in soil infiltration rate. However, we hypothesized that once the CROSS reached a certain value, soil cracks developed, alleviating the decrease in infiltration rate. We set up an indoor one-dimensional soil column infiltration experiment, using saline water with the same EC (4.0 dS m(-1)) but varying CROSSopt (Optimal CROSS) values (100.8 (NM100), 67.3 (NKM67), 37.9 (NCM38), 27.8 (NC28), and 9.3 (mmol(c) L-1)(0.5)) (KC9), and deionized water as the control (CK). The results demonstrated that Ksat decreased as CROSSopt increased, but there was no significant difference between NC28, NCM38, and NKM67 (P > 0.05). NM had the lowest Ksat, around 50 % of CK's. There was a positive correlation between CROSSopt and soil salination rate, with NM100 having about a 30 % higher salination rate than KC9. Only KC9 reduced the crack number compared with CK, and NKM67 had the most cracks. Compared with CK, NM100, NC28, and KC9 reduced the soil crack aspect ratio, with KC9 having the smallest ratio. The anisotropy of NKM67, NCM38 and KC9 was closer to 1.0 compared with CK, while NM100 was closer to 0. Based on EC and CROSSopt, we propose categorizing water samples into three types: no permeability problem expected, severe permeability problem expected, and no severe permeability problem expected. This study provides valuable theoretical support for assessing saline water quality and protecting soil quality.