Calcium sulphoaluminate cement has a potential application in sub-zero temperature environments due to its rapid hydration, rapid setting, fast strength development, and high early strength. In cold construction sites, such as polar regions, the materials and the concrete both are exposed to sub-zero temperatures for extended periods. It is thus necessary to investigate the formation, condensation, hardening and microstructures of calcium sulphoaluminate cement at sub-zero temperatures. Cold concreting method widely recognized involves directly blending raw materials on-site at the same temperature as the ambient environment, even at sub-zero temperatures, follows via pouring and curing fresh concrete without supplemental heating. Choosing a suitable mixing solution is crucial for the effective cold concreting technology. The mixing solution must prevent freezing of both the solution and the cement pastes at sub-zero temperatures, initiating the cement hydration reaction and enabling continuous progression. Also, the properties of the mixing solution (i.e., pH value and ion concentration) affect the dissolution and hydration of clinker minerals, leading to consequential changes in cement characteristics. In this paper, the performance evolution laws and mechanisms of calcium sulphoaluminate cement mixed with different mixing solutions at sub-zero temperatures were investigated.

Deionized water was used as a solvent. Six inorganic analytical pure chemical reagents and four organic analytical pure chemical reagents were used as solutes. Each of these reagents was dissolved individually in deionized water at room temperature. The solution was stirred until in a homogeneous state and left standing for 24 h. To ensure the temperature of the materials as the environmental temperature, the calcium sulphoaluminate cement and the mixing solution both were placed in a sub-zero temperature experimental system for a storage period of at least 24 h. All processes, i.e., sample shaping and curing, were carried out in the sub-zero temperature experimental system.

At -10 ℃, the mixing solution and cement were poured sequentially into a mixing bowl. They were stirred at a low speed for 120 s, paused for 15 s, and then stirred at a high speed for 120 s. The resultant paste was poured into the mold. After pouring, the vibrating table was used for 60 compactions. The samples were then placed on a rack for cured without being covered in the sub-zero temperature environment. Once the samples were cured to the specified age, a portion of them was taken out from the sub-zero temperature experimental system for the measurement of compressive strength. Another portion was crushed and sampled on-site. These samples were placed in anhydrous ethanol to terminate hydration at room temperature for at least 7 d. Afterwards, they were dried in vacuum at 40 ℃ and -0.08 MPa for no less than 24 h. A portion of the samples dried was used for the microscopic observation, while another portion was ground and further analyzed for the phase composition.

CaCl₂/Ca(NO₃)₂ solution is capable of accelerating hydration, resulting in an increased hydration temperature and a greater early strength. The peak temperature reaches up to 23.6 ℃ and 20.4 ℃, and the compressive strength at 12 h reaches 21.5 MPa and 16.5 MPa, respectively. Furthermore, the compressive strengths at 28 d are 101.8 MPa and 19.6 MPa, respectively. K₂CO₃ solution generates an increased hydration temperature and results in a high early strength, with a peak temperature of 7.7 ℃ and a compressive strength of 13.9 MPa at 12 h. However, the compressive strength only increases to 19.3 MPa after a 28-d period. MgCl₂, NaCl and NaNO₂ solutions exhibit a limited effect on enhancing hydration, presenting a low heat evolution and an early strength at their peak temperature of <5 ℃ and the compressive strength of <5 MPa after 12 h. Nonetheless, after a 28-d period, their compressive strength increases to 74.0, 42.7 MPa and 39.2 MPa, respectively. After 12-h hydration, the main products formed from CaCl₂, Ca(NO₃)₂, MgCl₂, NaCl and NaNO₂ in calcium sulphoaluminate cement are AFt and AFm phases, having a needle-bar and flake morphology. Also, the main products formed from K₂CO₃ solution-mixed samples are CaCO₃ and K₂SO₄ phases, having a cube and ellipsoid morphology.

CH₃OH, CH₃CH₂OH, (CH₂OH)₂ and C₃H₈O₃ solutions have a slight promotion effect on the hydration, resulting in a negligible heat evolution (peak temperature of < -5 ℃), minimal strength development at 12 h, and extremely low compressive strength for the samples after 28 d. However, the compressive strength of C₃H₈O₃ solution-mixed samples reaches 23.5 MPa after 28 d. For the samples mixed with four organic solutions, no hydration products such as AFt appear after 12-h hydration.

CaCl₂ and Ca(NO₃)₂ solutions with the same ions (Ca²⁺) as the cement minerals promoted cement hydration, resulting in an increased temperature and a greater early strength. MgCl₂, NaCl and NaNO₂ solutions had negligible effects on the hydration, leading to a decreased temperature and a weaker early strength. CO₃^2- in K₂CO₃ solution reacted with Ca²⁺ produced due to the dissolution of anhydrite to create CaCO₃ precipitates. This reaction increased the freezing point of the solution and reduced its frost resistance, causing the sample to freeze and preventing the effective cement hydration reactions.

The hydroxyl group of alcohol molecule in CH₃OH, CH₃CH₂OH, (CH₂OH)₂ and C₃H₈O₃ solutions was adsorbed by Ca²⁺ on the surface of cement particles to form an adsorption film. This film delayed hydration. In addition, the interaction between water and cement particles also impeded when the alcohol molecule contained a methyl group, exacerbating a retarding effect.