Premium
Characterizing the pore size distribution of a chloride silt soil during freeze–thaw processes via nuclear magnetic resonance relaxometry
Author(s) -
Liu Jianpeng,
Yang Ping,
Li Lin,
Zhang Ting
Publication year - 2020
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.1002/saj2.20087
Subject(s) - relaxometry , silt , soil water , water content , moisture , pore water pressure , soil science , chemistry , soil test , mineralogy , materials science , analytical chemistry (journal) , geology , environmental chemistry , geotechnical engineering , composite material , medicine , paleontology , spin echo , magnetic resonance imaging , radiology
Soil pore size distribution (PSD) is typically used to predict the soil freezing characteristic curve and estimate the hydrological and mechanical properties during freeze–thaw cycles. However, direct measurements of frozen soil PSD remain a great challenge. This study proposed a method to determine the PSD of frozen soils based on nuclear magnetic resonance (NMR) relaxometry. Tests were performed on a saturated chloride silt soil at different salt contents (0.3, 1.0, 2.0, and 3.0%) and temperatures (between –30 and 0 °C) during a freeze–thaw cycle. The NMR‐detected PSD (only accounting for pores occupied by unfrozen water) varied with soil temperature, salt content, and freeze–thaw cycle. The sequence of the water–ice phase change and hysteresis were also identified in a freeze–thaw cycle. A regression analysis was performed on the cumulative NMR‐detected PSD via a variant van Genuchten model. The critical freezing pore radius and the thickness of unfrozen water film were computed and used to transform NMR‐detected PSD into the actual PSD of frozen soils on the basis of a pore radius transformation equation established in this paper. Notably, the actual PSD accounted for pores occupied by ice and unfrozen water. The actual PSD indicated that the water–ice phase change was more pronounced in macro‐ and mesopores, especially at lower temperatures and salt content. A comparison between the calculated average pore size and those presented in other studies showed that the proposed technique is a valuable alternative for the prediction of actual frozen soil PSD.