Frozen Soil Mechanical Properties Testing: Standards, Methods, and Applications

Frozen soil mechanical properties testing determines the strength, deformation, and creep characteristics of soils under sub-zero temperature conditions, providing essential design parameters for construction in permafrost and seasonally frozen ground. Accredited third-party laboratories perform these tests using temperature-controlled cryogenic equipment in accordance with standards such as GB/T 50123, GB 50324, and applicable ASTM methods, supporting safe and durable engineering in cold regions.

Technical Overview

Frozen soil is a complex multi-phase geomaterial composed of mineral particles, ice, unfrozen water, and gases. The ice phase acts as a cementing and load-bearing agent, giving frozen soil mechanical behavior that differs fundamentally from that of unfrozen soil. As temperature decreases below the freezing point, compressive and shear strength generally increase, but the material simultaneously exhibits pronounced time-dependent deformation (creep), strong sensitivity to strain rate, and a gradual reduction of strength under sustained loading.

Because these characteristics cannot be extrapolated from conventional geotechnical tests, dedicated frozen soil mechanical testing is required. Testing is performed in environmental chambers and cryogenic triaxial cells that maintain specimen temperatures with an accuracy of approximately ±0.1 °C to ±0.5 °C, while applied stress, strain, and time are continuously recorded. The resulting parameters are used for bearing capacity evaluation, deformation prediction, frozen wall design, slope stability analysis, and frost heave and thaw settlement assessment in cold region engineering and artificial ground freezing projects.

Testing Samples

Undisturbed Frozen Core Samples

Undisturbed samples are recovered by core drilling from permafrost strata or artificially frozen ground. They must be transported and stored in a frozen state at or below the in-situ temperature, sealed to prevent moisture loss and sublimation. Specimens are typically trimmed into cylindrical shapes, such as 61.8 mm × 125 mm or 50 mm × 100 mm, in accordance with GB/T 50123 requirements, with minimal disturbance to the natural ice structure.

Remolded Frozen Specimens

Remolded specimens are prepared in the laboratory at controlled dry density, water content, and grain size distribution, then frozen under controlled temperature gradients. These specimens are widely used for parametric studies that isolate the influence of temperature, density, ice content, and salinity on mechanical behavior, and for projects where undisturbed sampling is impractical.

Artificially Frozen Soil Samples

Samples taken from artificial ground freezing zones, such as mine shafts, metro tunnels, and cross passages, are tested to verify the design strength and thickness of frozen walls. Testing often simulates the actual freezing temperature and confining stress conditions expected during construction.

Saline and Special Frozen Soils

Saline frozen soils, ice-rich soils, and frozen fine-grained soils containing unfrozen water require specialized test programs, because dissolved salts depress the freezing point and significantly reduce strength and stiffness compared with non-saline frozen soils at the same temperature.

Testing Items

Typical mechanical testing items for frozen soil include the following parameters:

• Uniaxial (unconfined) compressive strength at specified negative temperatures
• Triaxial compressive strength and complete stress-strain behavior under confining pressure
• Shear strength parameters of frozen soil, including cohesion and internal friction angle
• Tensile strength and splitting (Brazilian) tensile strength
• Elastic modulus, deformation modulus, and Poisson's ratio
• Creep characteristics, including creep curves, creep failure time, and long-term (ultimate) strength
• Strain-rate sensitivity of peak strength and failure strain
• Frost heave ratio, frost heave force, and frost susceptibility classification
• Thaw settlement coefficient and strength reduction after thawing
• Interface shear strength between frozen soil and structural materials such as concrete or steel

Testing Methods

Temperature-Controlled Uniaxial Compression Test

Specimens are first equilibrated at the target negative temperature in a cryostat or environmental chamber, commonly at levels such as −1 °C, −2 °C, −5 °C, −10 °C, −15 °C, and −20 °C. Loading is applied at a constant strain rate, typically ranging from 0.1% to 1.0% per minute, while the axial stress-strain curve is recorded. Peak stress or the stress at a defined failure strain is taken as the compressive strength, following the frozen soil test procedures in GB/T 50123.

Cryogenic Triaxial Compression Test

A temperature-controlled triaxial cell applies confining pressures that represent realistic in-situ stress conditions. Multiple confining pressures at each temperature level allow construction of the strength envelope, from which cohesion and internal friction angle of the frozen soil are derived, together with elastic modulus and Poisson's ratio. The methodology adapts conventional triaxial testing practice, such as ASTM D4767 principles, to sub-zero temperature conditions.

Creep Testing

Creep tests apply a sustained deviatoric stress at constant temperature and measure axial strain as a function of time. The results reveal attenuative creep, steady-state (non-attenuative) creep, and progressive creep stages leading to failure. Isochronous stress-strain curves constructed from a series of creep tests are used to determine the long-term strength, which is substantially lower than short-term strength and governs permanent, load-bearing structures in frozen ground.

Direct Shear and Tensile Testing

Direct shear apparatus in a low-temperature enclosure measures the shear strength of frozen soil and frozen soil-structure interfaces relevant to pile foundations and frozen walls. Tensile strength is commonly obtained by axial tension or splitting tests, since tensile cracking governs the thermal contraction behavior of frozen ground during rapid temperature drops.

Frost Heave and Thaw Settlement Testing

Frost heave tests are conducted in open or closed systems with one-directional freezing, measuring heave amount, heave rate, and water migration, in accordance with methods such as ASTM D5918 or the frozen soil test procedures of GB/T 50123. Thaw settlement tests apply a constant load while the specimen is warmed, yielding the thaw settlement coefficient needed to predict ground deformation during permafrost degradation or seasonal thawing.

Quality Control Measures

Reliable results depend on strict control of specimen temperature uniformity, accurate calibration of load and displacement sensors at negative temperatures, prevention of moisture migration and surface sublimation during testing, and verification of specimen dimensions and ice distribution before and after each test. Accredited third-party laboratories maintain documented procedures and instrument traceability to national metrology standards.

Applications

Frozen soil mechanical test data directly support engineering design and safety assessment in the following areas:

• Design of railways, highways, bridges, and transmission towers on permafrost and seasonally frozen ground
• Artificial ground freezing for mine shafts, metro tunnels, cross passages, and deep excavation support
• Oil and gas pipelines in Arctic and high-altitude permafrost regions, including frost heave and thaw settlement mitigation
• Foundation design of buildings in frozen soil regions in accordance with codes such as GB 50324 and JGJ 118
• Slope stability and landslide assessment in frozen and thawing ground under climate warming
• Site evaluation for cold storage facilities, LNG terminals, and cryogenic containment structures
• Research on permafrost degradation, frozen wall performance, and long-term serviceability of frozen ground structures

FAQ

Q: What temperature range is typically covered in frozen soil mechanical testing?

A: Most engineering projects require testing between −1 °C and −30 °C, with test points selected around the in-situ or design freezing temperature. Because strength and creep behavior vary strongly with temperature, a minimum of three to five temperature levels is usually recommended to establish reliable design relationships.

Q: How should frozen soil samples be preserved before testing?

A: Samples must be kept in a continuously frozen state at or below their in-situ temperature, sealed in moisture-proof wrapping, and transported under an unbroken cold chain. Any thawing and refreezing cycle alters the ice fabric and can significantly change the measured mechanical properties.

Q: Which standards govern frozen soil mechanical testing?

A: Commonly applied references include GB/T 50123 for geotechnical testing methods including frozen soil tests, GB 50324 and JGJ 118 for frozen ground investigation and foundation design, and ASTM methods such as D5918 for frost heave susceptibility, with triaxial and unconfined compression principles adapted from ASTM D4767 and ASTM D2166 to cryogenic conditions.

Q: Why is the long-term strength of frozen soil lower than its short-term strength?

A: Under sustained loading, ice at grain contacts undergoes pressure melting, viscous flow, and progressive bond breakage, producing creep deformation and a gradual loss of load capacity. Design of permanent structures in frozen ground must therefore be based on long-term strength determined from creep testing rather than on short-term peak strength.

Q: How long does frozen soil mechanical testing take?

A: Strength tests at controlled strain rates generally require several hours per specimen, including temperature equilibration. Creep tests are more time-consuming, often running from several days to several weeks per stress level, so project schedules should allocate sufficient time for the full test program.