Chao Zhang

Water ice in lunar polar regolith is both a scientific target and a potential resource for sustained exploration, but its abundance cannot be inferred reliably unless its physical morphology is also constrained. Here we measure the thermal conductivity of a lunar regolith simulant under low‐temperature and low‐pressure conditions relevant to the lunar polar environment, with systematic variations in dry density, ambient pressure, temperature, ice saturation, and pore‐ice morphology. Two contrasting specimen‐preparation routes were used to generate ice‐cemented (IC) and inter‐mixed fabrics. Low‐field nuclear magnetic resonance measurements and soil‐freezing characteristic analysis show that unfrozen water is negligible under the thermal conductivity test conditions, so the observed response can be interpreted primarily in terms of solid ice content and ice morphology. The results show that the thermal conductivity of the dry simulant is governed jointly by density, pressure, and temperature through changing contributions from particle‐contact conduction, pore‐gas conduction, and radiative transfer. For ice‐bearing specimens, thermal conductivity increases with ice saturation in both fabrics, but the conductivity‐saturation trajectories differ fundamentally because the pore ice is distributed differently. Inter‐mixed specimens, in which ice occurs mainly as discrete pore‐filling particles, show a gradual increase close to lower‐bound composite predictions, whereas IC specimens, in which ice bridges and cements particle contacts, show a much faster sigmoidal increase with saturation. These results demonstrate that thermal conductivity is not simply a proxy for ice abundance, but a joint indicator of ice content and pore‐ice morphology, with direct implications for interpreting future thermophysical measurements in the lunar south polar region.

Swelling potential (SP) has long been used as a terminology to describe a soil's expansibility. It is commonly defined in terms of pressure or deformation under certain constraints. However, fundamentally, SP originates from the soil‐water interactions in the interlayer space of expansive minerals and should not depend on displacement or force constraints. Here, the writers propose a SP based on the concepts of soil sorptive potential, unitary definition of matric potential, and water retention hysteresis. Water retention hysteresis in low matric potential is the result of interlayer hydration against the interlayer energy barrier. This energy barrier prevents water from entering the interlayer space. SSP synthesizes all the known sources of water adsorption, which provides the energy for soil swelling and can be determined under the unified definition of matric potential. The SP is defined as the energy hysteresis of interlayer hydration during wetting and drying. It is a function of relative humidity and can be calculated solely from the soil water isotherm (SWI). The SWI data of a wide variety of fine‐grained soils are used to determine and assess the proposed SP. For validation, the SP index (SPI), defined as the maximum energy consumed to overcome the energy barrier during wetting, is used. The SPI compares well with several expansive soil classification systems, confirming the validity of the SP. This study provides a scientific basis for linking soil water potential and energy used for swelling and understanding the volumetric behavior of expansive soil under varying humidity environments.

Deep excavation often fails after rainfall partly because of the complex hydro-mechanical coupling process. The clayey soils under high-stress unloading will undergo a quick volume expansion and become unsaturated with nonnegligible suction. The subsequent rainfall yields a strong infiltration condition and causes noticeable volume expansion or collapse, dictated by soil types and stress paths. This hydro-mechanical process remains elusive. Herein, a series of high-stress oedometer tests was conducted to investigate the evolutions of suction, volume, K ₀ coefficient, and small-strain shear modulus in clayey soils. Two types of clayey soils were employed: nonexpansive kaolinite and expansive clay. Samples were tested under three loading–1 ^st unloading–soaking–2 ^nd unloading stress paths, covering two maximum vertical stresses (1.6 and 3.2 MPa) and two unloading ratios at soaking, defined as the ratio of the pre-soaking vertical stress to the pre-unloading maximum vertical stress (0.25 and 0.50). Results reveal three key findings: (1) Unloading-induced matric suction increment reaches up to 408 and 610 kPa for kaolinite and expansive clay, respectively, when the vertical stress decreases from 3.2 MPa to ∼5 kPa under the K ₀ condition. Moreover, this matric suction change increases linearly with the logarithm of the vertical stress change. (2) In the soaking stage, the volume changes reach up to 47.2% and 55.8% of the prior rebound for kaolinite and expansive clay, respectively, while the K ₀ coefficient and the small-strain shear modulus vary less than 3% for both clayey soils. (3) An effective stress-based model is proposed for the volume evolution upon high-stress unloading and subsequent soaking, incorporating the effect of unloading-induced suction. This model effectively captures the volume change in the unloading and soaking stages with R ² ≥ 0.86.

The random inhomogeneity of rubble stone–concrete targets is the fundamental cause of the significant randomness in anti-penetration performance, thereby limiting the engineering application of rubble stone–concrete. Here, to study the penetration damage effects and the anti-penetration mechanisms of the rubble stone–concrete, a random inhomogeneity modeling method is proposed, which constructs high-fidelity hexahedral mesh models of rubble stone–concrete targets. The reliability of the numerical simulation methodology and parameters is validated by comparison with penetration test data. The effects of corundum rubble size, corundum rubble volumetric percentage, concrete strength, and projectile impact velocity on penetration damage and penetration depth are discussed. The results reveal that the optimal corundum rubble size should equal 1.0 times the projectile diameter to balance the performance and stability of anti-penetration at the 30% volumetric percentage. As the concrete strength increases, corundum rubble gains more support during the penetration process, which also makes significant contributions to improving anti-penetration performance. The percentages of the penetration resistance impulse contribution from concrete and corundum rubble are unrelated to the projectile impact velocity. Moreover, an empirical formula for predicting penetration depth in corundum rubble–concrete targets is proposed, which provides a reference for evaluating the anti-penetration performance of corundum rubble–concrete.