Non-destructive X-ray computed tomography (µCT) scanning was used to characterise changes in pore architecture as influenced by the proportion of cations (Na, K, Mg, or Ca) bonded to soil particles. These observed changes were correlated with measured saturated hydraulic conductivity, clay dispersion, and zeta potential, as well as cation ratio of structural stability (CROSS) and exchangeable cation ratio. Pore architectural parameters such as total porosity, closed porosity, and pore connectivity, as characterised from µCT scans, were influenced by the valence of the cation and the extent it dominated in the soil. Soils with a dominance of Ca or Mg exhibited a well-developed pore structure and pore interconnectedness, whereas in soil dominated by Na or K there were a large number of isolated pore clusters surrounded by solid matrix where the pores were filled with dispersed clay particles. Saturated hydraulic conductivities of cationic soils dominated by a single cation were dependent on the observed pore structural parameters, and were significantly correlated with active porosity (R2 = 0.76) and pore connectivity (R2 = 0.97). Hydraulic conductivity of cation-treated soils decreased in the order Ca > Mg > K > Na, while clay dispersion, as measured by turbidity and the negative charge of the dispersed clays from these soils, measured as zeta potential, decreased in the order Na > K > Mg > Ca.
The results of the study confirm that structural changes during soil–water interaction depend on the ionicity of clay–cation bonding. All of the structural parameters studied were highly correlated with the ionicity indices of dominant cations. The degree of ionicity of an individual cation also explains the different effects caused by cations within a monovalent or divalent category. While sodium adsorption ratio as a measure of soil structural stability is only applicable to sodium-dominant soils, CROSS derived from the ionicity of clay–cation bonds is better suited to soils containing multiple cations in various proportions.
Estimating the macroscopic capillary length from Beerkan infiltration experiments and its impact on saturated soil hydraulic conductivity predictions
