3.4.1. Autogenous Shrinkage
The non-contact mortar shrinkage tester was used to detect the autogenous shrinkage development of low-temperature-rise concrete from the initial setting stage to the early hydration stage within 7 days. The test results are shown in
Figure 5. From the early autogenous shrinkage variation trend of the four groups of proportions, it can be seen that the first three groups all experienced significant shrinkage in the early stage, while the CAMC-0.6% group experienced a slight expansion phenomenon. Among them, the blank group and the CAMC-0.2% group both experienced a relatively rapid shrinkage at the earliest stage, followed by an expansion behavior. After the expansion reached its peak, the sample experienced a second shrinkage. Based on its early sample shrinkage behavior, the entire change stage of the sample can be divided into three stages: shrinkage, expansion, and then shrinkage [
25]. The four characteristic values of ΔP, ΔE, ΔH, and ΔS represent the changes in shrinkage behavior in the three stages, where ΔP is the first-stage shrinkage change value, ΔE is the absolute value of the early expansion phenomenon in the second stage, ΔH is the change value after the expansion reaches its maximum value and then shrinks up to day 7 in the third stage, and ΔS is the final shrinkage change value of the system during the entire early autogenous shrinkage process from the initial setting time to 7 days.
As shown in
Table 5, the blank group and CAMC-0.2% group had a shrinkage value of 12.91 and 12.59 με for ΔP, respectively, while the CAMC-0.4% and CAMC-0.6% groups did not exhibit the initial shrinkage phenomenon. When developing into the second stage of early expansion, the blank group and CAMC-0.2% sample group had an expansion value of 15.8 and 18.9 με for ΔE, respectively, with the CAMC-0.2% sample group having a slightly increased expansion value compared to the blank group in the second stage. Similarly, the CAMC-0.4% sample did not exhibit any expansion phenomenon, while the CAMC-0.6% sample had an expansion value of 9.29 με for ΔE. When developing into the third stage of re-shrinkage, it was found that the first three groups of samples had shrinkage values of 87.44, 88.07, and 14.85 με for ΔH, respectively, while the CAMC-0.6% sample continued to expand with an expansion value of 69.2 με. When analyzing the final shrinkage value of the entire system, it was found that the shrinkage of the sample decreased with an increasing CAMC content, with the blank group, CAMC-0.2%, and CAMC-0.4% samples having shrinkage values of 86.82 με, 70.89 με, and 14.52 με, respectively. However, the CAMC-0.6% sample ultimately expanded, with an expansion value of 78.49 με.
There are currently three main theories that can explain the mechanism of autogenous shrinkage: surface tension theory, disjoining pressure theory, and capillary tension theory. The surface tension theory suggests that a decrease in humidity leads to a reduction in the adsorbed water layer between particles within the gelling material, resulting in an increase in surface tension and macroscopic shrinkage of the material [
26]. The disjoining pressure theory suggests that the separation pressure between solid particles is a combination of complex forces such as van der Waals forces and layer repulsion. When the humidity decreases, the separation pressure between solid particles decreases and shrinkage occurs [
27]. The capillary tension theory suggests that a decrease in humidity leads to the formation of a meniscus at the interface between the gas phase and the liquid phase, resulting in macroscopic shrinkage [
28,
29,
30]. The hydration of cement particles proceeds with the consumption of water and the formation of a porous structure, as shown in
Figure 6a. In capillary theory, due to the existence of surface tension, a curved liquid surface forms in unsaturated pores. This meniscus causes capillary stress in the pores, resulting in a decrease in volume [
31]. The lower the water–cement ratio, the higher the autogenous shrinkage rate, the finer the pore structure, and the lower the porosity. The consumption of free water gradually causes it to enter small pores from large pores to achieve thermodynamic equilibrium. Due to the significant reduction in pore volume, the consumption of water in low-water–cement-ratio mixtures results in faster changes in pore saturation. This helps to reduce the relative humidity faster in low-water–cement-ratio slurries. In capillary pores with a low relative humidity, the radius of curvature of the meniscus is smaller, so the capillary stress is larger and the autogenous shrinkage Is also larger [
32,
33,
34].
Through the analysis of the above data, it is found that, during the early autogenous shrinkage process, some samples undergo a process of micro-expansion. The expansion source in this stage mainly includes three reasons: first, the expansion pressure caused by the formation of hydration products (CH, AFt), followed by the thermal expansion phenomenon caused by the heat release of hydration [
35,
36], and then the influence of slurry bleeding and its reabsorption [
29,
30]. When the sample contains a high content of CAMC, the hydration process is significantly delayed, and there is a significant amount of excess free water in the cement paste, resulting in a “bleeding-like” phenomenon (as shown in
Figure 6b,c). Previous
1H-NMR research results have also confirmed that CAMC can cause a “bleeding-like” phenomenon [
18]. In most cases, the reabsorption of secreted water is the most important reason for this. The free water content and relative humidity inside the high-water–cement-ratio sample are still high, and the capillary stress is relatively small, so the autogenous shrinkage is very small, and the expansion phenomenon occurs [
37,
38].
3.4.2. Shrinkage Analysis at Different Stages of the Entire Process
The drying shrinkage is shown in
Figure 7. The long-term drying shrinkage rate of low-temperature-rise concrete decreases with time, and the long-term drying shrinkage gradually decreases with the increase in CAMC content. The drying shrinkage value of the blank group at 90 days is about 551.46 με. After adding 0.2% CAMC, the long-term drying shrinkage value decreases by 42 με. As the CAMC content increases, the long-term drying shrinkage decreases to 500.97 and 401.94 με. The long-term drying shrinkage of the four groups of samples is basically consistent with the change law of autogenous shrinkage described in the previous section.
The overall shrinkage change in low-temperature-rise concrete during the whole process is mainly the sum of the early autogenous shrinkage before demolding and the long-term drying shrinkage after demolding, and was determined in a standard environment. The early autogenous shrinkage and long-term drying shrinkage of low-temperature-rise concrete have been tested and measured in our previous article.
Table 6 analyzes the early autogenous shrinkage and long-term drying shrinkage data of the four groups of samples. The specific division method can be found in the literature [
25]. A
3 in
Table 6 is the early autogenous shrinkage values of the four groups of samples in
Figure 5, A
14 is the stable autogenous shrinkage values at 14 days, T
14 is the long-term total shrinkage values of the four groups of low-temperature-rise concrete after demolding at 14 days, and E
14 is the total shrinkage values of the low-temperature-rise concrete after initial setting. E
14 can be calculated by the following formula:
D
14 is the drying shrinkage value of concrete at the age of 14 days, and D
14 can be calculated by the following formula:
The analysis of
Table 6 reveals that no expansion was observed in any of the low-temperature-rise concrete samples, indicating that the shrinkage caused by drying can balance (or mask) the expansion observed in the early autogenous shrinkage phase. The overall trend is that the total shrinkage value decreases with the increase in CAMC content, where the total shrinkage value of the blank group is 473.37 με, and after the addition of 0.6% CAMC, the total shrinkage value decreases to 370.91 με. The decrease in the total shrinkage of the CAMC-0.6% sample is due to the expansion phenomenon of autogenous shrinkage and the reduction in long-term drying shrinkage. The total shrinkage rate mainly depends on the drying shrinkage rate, as the observed trends are the same in both cases. This phenomenon is due to the larger magnitude of drying shrinkage compared to autogenous shrinkage.