Synergistic Freeze-Resistant Strategy of Multi-Stage PCM Concrete Incorporated with Rice Husk Ash and Fly Ash

Abstract

Damage to buildings and infrastructure caused by freeze–thaw cycles is a common problem in cold regions. To counteract this, multi-stage phase change material (PCM) aggregate concrete has gained attention for its potential in structural protection. PCM concrete is a type of intelligent concrete that regulates and controls the temperature by incorporating PCM. PCM aggregate can efficiently absorb and release significant amounts of heat within a defined temperature range. This study explored the feasibility of using agricultural waste rice husk ash (RHA) and industrial waste fly ash (FA) to produce PCM concrete. The combined use of RHA and FA with multi-stage PCM aggregate concrete allowed the two materials, pozzolanic materials and PCM, which have different approaches to improving the freeze resistance of concrete, to be effectively utilized together and synergistically enhance the durability and energy efficiency of buildings in cold regions. An experimental program was conducted to prepare PCM concrete by replacing cement with 5%, 10% and 15% RHA and 10%, 20% and 30% FA in different mixtures. The results show that when the replacement amounts of RHA and FA are both 10%, the compressive strength of the concrete can be effectively improved, while the thermal conductivity and thermal diffusivity are reduced. The incorporation of RHA and FA improved the thermal regulation of PCM concrete. Strength loss, relative dynamic elastic modulus (RDEM) loss and mass loss were all minimal with RHA at 15% and FA at 10% replacement.

1. Introduction

In frosty zones, constructed facilities are frequently subjected to freeze–thaw cycles, which can result in material deterioration and structural performance degradation. The durability of concrete, as a key material in construction, is closely associated with the safety and longevity of the project [1]. Physical damage resulting from freeze–thaw cycles, especially the creation and development of cracks, is a significant factor leading to the degradation of concrete infrastructure’s long-term performance. Therefore, how to develop a type of concrete that can effectively resist freeze–thaw damage while taking into account economic and environmental sustainability has always been a hot topic in the scientific research of concrete.

At present, the strategies for enhancing the freeze resistance of concrete primarily focus on two aspects: Firstly, improving the internal pore characteristics of concrete, such as lowering the water–cement ratio or adding air-entraining agents [2,3], mineral admixtures [4,5], nanomaterials [6] or water-absorbent resins [7,8]. Secondly, enhancing the tensile strength and crack resistance properties of concrete, such as incorporating various fibers [9,10] and modified particles [11]. All of these methods can effectively enhance the freeze resistance of concrete, but they are all “passive” methods; that is, they reduce the freeze–thaw expansion stress or increase the critical tensile stress at the time of damage after the concrete has undergone freezing and thawing. However, they do not truly eliminate the freeze–thaw cycle damage, so there will still be some degree of freeze–thaw deterioration of the concrete. Using PCM aggregate to mitigate temperature fluctuations during freeze–thaw cycles can fundamentally reduce the damage caused by these cycles, thereby enhancing the freeze resistance of concrete as an ‘active’ method. This method can effectively extend the lifespan of concrete and has attracted extensive attention from the scientific community.

In recent times, research on the use of low-temperature PCM to enhance the freeze resistance of cementitious materials has gradually emerged [12]. PCM has good thermophysical properties and is an efficient energy storage material. When the external temperature reaches its phase change point, it will undergo a morphological transformation and absorb or release significant amounts of heat energy, effectively regulating the surrounding temperature [13]. As a heat storage material, PCM can be added to building materials to address the problem of a mismatch between the timing of energy supply and the timing of building demand. The excess energy is stored when the building temperature is too high, and it is released when the temperature is too low. PCM is used for freeze-resistance enhancement of concrete, which utilizes a considerable amount of latent heat during the solid–liquid conversion process to slow down the temperature drop inside the concrete, thus reducing material damage subjected to the freeze–thaw cycles of concrete and enhancing the freeze-resistance performance of concrete [14].

Furthermore, to enhance the freeze resistance of PCM concrete while prioritizing environmental conservation and resource reutilization, research has increasingly focused on integrating industrial by-products, converted into mineral admixtures, into concrete. This approach represents a significant and promising avenue of investigation. Industrial by-products such as FA and RHA, due to their unique chemical composition and physical properties, have been shown to enhance the performance of concrete effectively. In particular, RHA, as a resource of amorphous silica derived from agricultural by-products, has an important role in forming compounds with cement-like properties due to the active components produced under suitable combustion conditions [15]. The study by Khan et al. [16] found that FA and silica fume mixtures can undergo a secondary hydration reaction with cement hydration products and reduce the alkali-aggregate reactivity of the concrete to increase the compressive strength, impermeability and crack resistance properties of the concrete. The study by Macquarie et al. [17] showed that the addition of nano-silica and FA can significantly enhance the early and later compressive strength of concrete. The study by Meddah et al. [18] found that the air content of concrete decreased after incorporating alumina nanoparticles and RHA, and the durability of concrete improved.

The existing literature extensively documents the use of RHA and FA in various modified concretes. However, there is a notable paucity of research on the synergistic incorporation of these two materials with PCM in concrete. In light of this, the focus of this study is to systematically investigate and evaluate the feasibility of incorporating RHA and FA together in PCM concrete, as well as the potential impact of this composite admixture on the concrete’s freeze–thaw resistance.

2. Materials and Experimental Details

2.1. Materials

In this research, 42.5 R grade Type I ordinary Portland cement from Conch Group in Anhui Province, China, was used as the cementitious material. Its physical properties and chemical compositions are shown in

Table 1. Physical properties of cement.

Density (g·cm−3)	Specific Surface Area (m2·g−1)	Loss on Ignition (%)	Initial Setting Time (min)	Final Setting Time (min)
3.12	0.35	3.02	135	250

and

Table 2. Chemical composition of material (wt.%).

	SiO2	CaO	Al2O3	Fe2O3	MgO	SO3	Na2O	K2O	TiO2	MnO	P2O5
Cement	18.68	64.27	4.01	3.97	2.87	3.26	0.12	0.43	0.57	0.10	1.05
RHA	86.82	1.70	0.47	1.71	0.61	0.55	0.20	5.12	0.05	0.45	1.89
FA	43.46	10.05	30.57	9.12	2.08	0.53	1.02	0.67	0.55	0.05	1.30

. Natural limestone gravel with a continuous gradation of 5–20 mm from Xi’an, Shaanxi Province, China, served as the coarse aggregate. To remove the influence of mud content, the coarse aggregate was pre-washed and dried in the laboratory. Natural river sand, obtained from Xi’an City, China, served as the fine aggregate, with a fineness modulus of 2.69 and a maximum particle size of 5 mm, categorized as ordinary medium sand. Experimental-grade ultrafine FA produced by Henan Yuanheng Environmental Protection Engineering Co. Ltd in Zhengzhou, Henan Province, China was utilized. The mineral composition is shown in Table 1. RHA, due to its porous and honeycomb structure, typically requires grinding treatment to improve its particle morphology and reactivity. According to the literature, mixing time and revolution count significantly affects the characteristics of cement containing RHA. Appropriate grinding processes can substantially reduce the porosity of RHA particles, increasing their specific surface area, thereby enhancing their activity and performance in concrete [19]. In this study, we employed a similar method to grind RHA, ensuring its effectiveness in phase change concrete. Specifically, the rice husk ash was ground using a laboratory Los Angeles abrasion machine and sieved to remove larger particles and impurities, with the resulting particle diameter not exceeding 20 mm. The macroscopic and microscopic morphologies of the processed RHA are shown in Figure 1.

We used shale ceramisite from Gongyi, Henan Province, China, as the PCM vessel. This shale ceramisite granule had a particle size range of 5–20 mm, which met the requirement for continuous grading. The shale ceramisite particles were incinerated, cooled and crushed to form a rich pore structure, which provided sufficient space for the impregnation of PCM [20]. The porosity measured by mercury intrusion porosimetry (MIP) was 33.30% and the pore volume was 0.3321 cm³/g. These properties made shale ceramisite an ideal carrier for PCM. Dodecane polymer, tridecane polymer and tetradecane polymer were prepared by encapsulating three alkanes in shale ceramisite by the vacuum impregnation method [21]. In this work, a dual-coating system was used to increase the PCM aggregate’s thermal conductivity and stop leaks during phase transitions. The system consisted of a first layer of epoxy resin composite and a second layer of cement powder. The coating material, E51 epoxy resin, was applied, and to speed up the curing process, T31 phenolic glycidyl ether epoxy resin curing agent was used. To further enhance the thermal conductivity of the epoxy composite, diluent-based glycidyl ether (CH1122) and 10,000 mesh ultrafine graphite powder were incorporated. The first layer consisted of ultrafine graphite powder, diluent, curing agent and epoxy resin in a ratio of 0.1:0.15:0.2:1. The second layer, made of cement powder, was applied to the epoxy resin to improve the interface properties between the PCM aggregate and the matrix. The completed PCM aggregate was prepared as shown in Figure 2 [22].

The fixed phase transition temperature of a single organic PCM makes it difficult to flexibly meet the energy storage requirements of the actual project. Organic PCM also has the disadvantage of a large degree of subcooling, which further limits its application in the field of construction. To broaden the phase change temperature range, three organic PCMs—dodecane, tridecane and tetradecane—were selected as PCM aggregate preparations. Their phase change temperatures, ranging from 5 °C to 15 °C, are suitable for concrete in cold regions, enabling multi-stage thermal potential release during temperature drops and mitigating freezing–thawing damage.

2.2. Preparation of Specimens

Previous studies have demonstrated that concrete with dodecane, tridecane and tetradecane aggregates mixed in a 4:3:3 ratio exhibits optimal low temperature conditioning properties [22]. Therefore, this study maintained this ratio and replaced the coarse aggregate entirely by volume. In designing the experiment, this research prepared 10 groups of samples with different ratios. The replacement ratios of RHA ranged from 5% to 15%, while the replacement ratios of FA ranged from 10% to 30%. Additionally, a control group of PCM concrete without these two materials was set up. The specific mixing ratios are detailed in

Table 3. The mixtures of concrete (kg/m³).

Mix	RHA Content	FA Content	Cement	RHA	FA	Sand	PCM Aggregate	Water	SP
R-0	0%	0%	550	0	0	652	560	176	5.5
R-1	5%	10%	467.5	27.5	55	652	560	176	5.5
R-2	5%	20%	412.5	27.5	110	652	560	176	5.5
R-3	5%	30%	357.5	27.5	165	652	560	176	5.5
R-4	10%	10%	440	55	55	652	560	176	5.5
R-5	10%	20%	385	55	110	652	560	176	5.5
R-6	10%	30%	330	55	165	652	560	176	5.5
R-7	15%	10%	412.5	82.5	55	652	560	176	5.5
R-8	15%	20%	357.5	82.5	110	652	560	176	5.5
R-9	15%	30%	302.5	82.5	165	652	560	176	5.5

.

To ensure the concrete meets the C50 design strength requirement, a water–cement ratio of 0.32 was set [23,24], and the dosage of each component was calculated accurately according to the specification-recommended concrete density range (2400 kg/m³ to 2500 kg/m³). To maintain workability, a water content of 176 kg per cubic meter of concrete was determined, considering the specification-recommended water content of 220 kg/m³ and the 20% efficiency of the superplasticizer. After experimental adjustments, the final sand/gravel ratio was set at 0.376 [25,26], and superplasticizers were added at 1% of the cement content.

In consideration of the potential fragmentation of PCM aggregates during mixing, the mixing sequence was tailored in this study to safeguard the integrity of the aggregates. The specific steps are outlined below:

(1)

Cement, sand, gravel, RHA and FA were first combined in a concrete mixer for 60 s.

(2)

Superplasticizer and water were then introduced, and mixing continued for another 120 s.

(3)

To avoid damaging the PCM aggregate, it was added in the third step and mixed for an additional 120 s.

(4)

The mixed material was finally poured into molds and compacted using vibration.

(5)

After a day, the specimens were demolded, and to guarantee adequate hardening, they were cured for 28 days at 20 °C in a controlled atmosphere with 96% humidity. The specimens, measuring 100 mm × 100 mm × 100 mm, were suitable for subsequent testing of mechanical, thermal and pore properties.

2.3. Testing Methods

2.3.1. Mechanical Properties

To evaluate the mechanical properties of the concrete, the cubic compressive strength was measured as a key index. Ten concrete groups were tested for compressive strength at 7, 14 and 28 days utilizing a 2000 kN microcomputer-controlled electro-hydraulic servo universal testing machine in this research. To ensure data accuracy, each group was tested using a minimum of three specimens.

2.3.2. Thermal Properties

In this research, thermal conductivity and thermal diffusivity were measured, and volumetric specific heat capacity was computed employing the transient planar source (TPS) method [27] with a Hot Disk 2500S device. A double-helix structure etched on a thin nickel sheet served as a dynamic temperature sensor and the heat source. A mica-insulating layer was placed between two specimens to form a sandwich structure, which was then electrically heated by the Joule effect. A least squares fitting technique was used to measure the thermal diffusivity and conductivity, and the volumetric specific heat capacity was computed according to Equation (1) [28]:

$$\lambda =\rho c_p\alpha$$

(1)

where ρc_p represents the volumetric specific heat capacity, α stands for the thermal diffusivity and the thermal conductivity is denoted by λ.

2.3.3. Time–Temperature Curve

Time–temperature histories curves were utilized to assess the thermal regulation and thermal response of PCM concrete incorporated with RHA and FA. Concrete specimens with thermocouples inserted were placed in a freeze–thaw tester with air as the surrounding medium to simulate heat exchange conditions in a natural environment. The specimens were first cooled to −20 °C for a duration of 300 min to simulate low-temperature conditions. Subsequently, the specimens were heated to 15 °C for about 200 min to simulate the thermal recovery process after sunlight or an increase in ambient temperature. The type K thermocouple was positioned at the geometric center of the concrete cube to ensure representative temperature measurement, and the temperature data were collected in real-time by a temperature recorder to ensure the accuracy of the experimental results.

2.3.4. Freeze Resistance

The evaluation indices for freeze–thaw cycles include mass loss, RDEM and compressive strength. The 28-day compressive strength of concrete and the compressive strength after cycles of freeze–thaw were tested by a universal concrete testing machine. This test followed the rapid freezing–thawing method, and the test steps were conducted in strict alignment with ordinary concrete long-term performance and durability test methods (GB/T50082-2009) [29]. The freeze–thaw cycles testing apparatus was a TDR-28 rapid freezing and thawing tester produced by Tianjin Gangyuan Instrument Experimental Factory. In this test, an electronic scale with a range of 50 kg was used to measure the concrete mass change under different cycle times of prismatic specimens, and the RDEM was tested by an ultrasonic detector with a size of 100 × 100 × 400. The compressive strength of the cubic specimen was tested by the concrete universal testing machine after the freeze–thaw cycles, with specimens sized at 100 × 100 × 100. In each cycle, the temperature at the center of the specimen is set to a minimum of −18 °C and a maximum of 5 °C. Rapid freezing–thawing tests were conducted for 50, 100, 150 and 200 cycles to evaluate freeze resistance. Three specimens from each mix proportion were tested, and the average value was recorded to ensure data accuracy.

Concrete specimens were weighed using rags to remove surface moisture from the specimen blocks before weighing, and the rate of mass loss was measured using Equation (2) after the specified cycles of freeze–thaw. The rate of RDEM loss was calculated using Equation (3) after testing the cubes’ RDEM using an ultrasonic tester. The rate of loss of compressive strength was calculated using Equation (4) after testing the cubes for compressive strength using a compression tester.

$$\Delta {\mathrm{m}}_{\mathrm{n}}={\displaystyle \frac{{\mathrm{m}}_0-{\mathrm{m}}_{\mathrm{n}}}{{\mathrm{m}}_0}}$$

(2)

where m₀ represents the original mass, ${\mathrm{m}}_0$ stands for the mass following n time freeze–thaw cycles and the mass loss rate following $\mathrm{n}$ time freeze–thaw cycles is denoted by $\Delta {\mathrm{m}}_{\mathrm{n}}$.

$$\Delta {\mathrm{E}}_{\mathrm{d}}={\displaystyle \frac{{\mathrm{E}}_{\mathrm{d}0}-{\mathrm{E}}_{\mathrm{dn}}}{{\mathrm{E}}_{\mathrm{dn}}}}$$

(3)

where ${\mathrm{E}}_{\mathrm{d}0}$ represents the original RDEM, ${\mathrm{E}}_{\mathrm{dn}}$ stands for the RDEM following n time freeze–thaw cycles and the RDEM loss rate following n time freeze–thaw cycles is denoted by $\Delta {\mathrm{E}}_{\mathrm{d}}$.

$$\Delta f_{\mathrm{cu}}={\displaystyle \frac{f_{\mathrm{cu}0}-f_{\mathrm{cun}}}{f_{\mathrm{cun}}}}$$

(4)

where $f_{\mathrm{cu}0}$ represents the original compressive strength, $f_{\mathrm{cun}}$ stands for the compressive strength following n time freeze–thaw cycles and the cubic compressive strength loss rate following n time freeze–thaw cycles is denoted by $\Delta f_{\mathrm{cu}}$.

2.3.5. NMR

The nuclear magnetic resonance (NMR) technique has unique advantages in measuring and characterizing the pore distribution of materials. It can non-destructively penetrate the material and provide information about the internal structure [30]. The NMR technique relies on the magnetic resonance phenomenon of atomic nuclei in a magnetic field, where different pore sizes and shapes affect the relaxation behavior of the nuclei, thus providing information about the pore structure. NMR infers the size, shape and distribution of pores by analyzing the relaxation behavior of fluids in the pores. When pores in a material are filled with water, protons (hydrogen nuclei) in the pore water will exhibit specific relaxation behavior in an external magnetic field. The transverse relaxation time (T2) spectrum measures how long it takes for a proton to lose coherence and return to its ground state in an external magnetic field. Larger pores typically correspond to longer T2 relaxation times, while smaller pores correspond to shorter T2 relaxation times. The correlation between relaxation times and pore properties can be estimated using Equation (5).

$${\displaystyle \frac{1}{T_2}}={\rho }_{2,sur}({\displaystyle \frac{S}{V}})$$

(5)

where ρ_2,sur represents the T₂ surface relaxivity, S/V stands for the pore surface area-volume ratio and the relaxation time is denoted by T₂.

3. Results and Discussion

3.1. Mechanical Properties

Figure 3 illustrates the compressive strength results. Replacing cement with RHA and FA in appropriate proportions significantly enhanced the compressive strength of concrete. The 28-day compressive strength of PCM concrete increased when RHA was added at 5–10% and FA at 10–20%. This improvement is attributed to the fine particles of RHA filling the voids between cement particles and aggregates, reducing porosity and enhancing densification [31]. Additionally, RHA contributes to the formation of hydrated calcium silicate gel (C-S-H gel), which further reduces porosity and improves concrete compactness [32]. This results in better stress distribution, reducing local stress concentrations and allowing the concrete to withstand pressure more uniformly, thereby increasing compressive strength. However, the 28-day compressive strength values for all groups with the same FA admixture were significantly lower at a 15% RHA admixture. This may be due to the increased water demand of RHA particles when their content exceeds 10%, leading to reduced compressive strength [33]. The optimal combination of 10% RHA and 10% FA increased the 28-day compressive strength by 15.43%.

The control concrete’s 7-day compressive strength was 76% of its 28-day strength, whereas the concrete with 10% RHA and FA replacement had a 75% strength. The addition of RHA and FA did not significantly mitigate the adverse effect of PCM aggregates on early hydration.

3.2. Workability of Concrete

Concrete’s workability is an important prerequisite for its excellent mechanical properties and durability. One can assess the concrete’s workability by observing its slump. The slumps of PCM concrete with different mixtures are shown in Figure 4. This study showed that the workability of PCM concrete incorporated with RHA and FA was significantly improved. The effect of the slump could be improved well by adding FA and RHA. This phenomenon occurred because the incorporation of RHA and FA reduced the amount of cement, and its micro-bead effect released the filling water between the cement particles so that the paste and the PCM aggregate was better encapsulated, which enhanced the fluidity of the concrete [34]. This phenomenon may also be traced to the fact that the incorporation of RHA and FA improve the water retention capacity of the concrete, further contributing to the slump.

3.3. Thermal Properties

Figure 5 presents the thermal diffusivity and thermal conductivity of each concrete group. In comparison to the control group, PCM concrete included with RHA and FA had reduced heat conductivity. Increasing FA content further reduced thermal conductivity with a constant RHA content. This decrease is ascribed to the development of thermal resistance at the interfaces with cement hydration products, specifically at the interfaces with RHA, FA and the cement matrix. These interactions reduce heat transfer, which is consistent with results in the literature and lowers PCM concrete’s thermal conductivity [35].

The capacity of heat transport in relation to heat storage is known as thermal diffusivity. Furthermore, the combination of RHA and FA decreased the thermal diffusivity of the PCM concrete. Lower diffusivity coefficient materials provide better insulating properties and better thermal inertia. The thermal diffusivity of the PCM concrete containing 15% RHA and 30% FA was determined to be 0.448, suggesting a 34.59% reduction, compared to the control concrete’s 0.685.

Equation (1) was used to calculate the volumetric heat capacity, which is displayed in Figure 6. Instead of specific heat capacity, volumetric heat capacity was used in this study. The findings demonstrated that the incorporation of FA and RHA enhanced the volumetric heat capacity of concrete. PCM concrete with 15% RHA and 30% FA substitution exhibited a volumetric heat capacity of 2.259 MJ/m³K, which was 1.08 times greater than that of the control concrete without FA and RHA. This suggests that in order to alter the temperature of PCM concrete with RHA and FA, more thermal energy is needed, thereby increasing its thermal inertia and aiding in temperature regulation.

3.4. Time–Temperature Curve

The time–temperature histories curves of PCM concrete included with RHA and FA are shown in Figure 7a. The performance of PCM concrete in terms of heat regulation during cooling and heating processes is shown by this curve. The temperature differential between the control concrete and the PCM concrete with RHA and FA is shown in Figure 7b. After being cooled for almost 300 min, each specimen was heated for nearly 200 min.

Due to heat loss, all specimens showed the same internal temperature trend during the cooling process, which decreased with the outside temperature. The incorporation of RHA and FA increased the volumetric heat capacity of the concrete, resulting in a slower thermal response compared to the control group. Specifically, PCM concrete with 15% RHA and 30% FA exhibited the slowest thermal response. The PCM concrete with 15% RHA and 30% FA only had an interior temperature of −12.7 °C when the outside temperature reached −18 °C. In contrast, the control group’s internal temperature was nearly −15.4 °C. Figure 7b shows that the latent heat release phase for R-8 and R-9 occurred between 0 and 350 min, which was primarily influenced by the RHA admixture. For the remaining groups, the latent heat release phase occurred between 0 and 150 min, which was mainly influenced by the PCM aggregate. Higher FA content improved the thermal effect during this phase, with temperature differences reaching up to 4 °C compared to the control group. After the latent heat release stage, the temperature difference between the control concrete and the PCM concrete with RHA and FA steadily approached zero due to the decreasing ambient temperature.

The specimens were heated following the cooling phase. According to Figure 7a, groups R-8 and R-9 showed a slower rate of temperature increase, with the maximum temperature difference between R-9 and the control group reaching up to 5 °C. The other groups were close to the control group, as displayed in Figure 7b. The absorption of thermal energy at this phase was greater than the latent heat release process, probably because of the medium’s rates of cooling and heating. The results indicated that a higher FA content in PCM concrete improved thermal performance, extending the duration for releasing and absorbing latent heat.

3.5. Freeze Resistance

3.5.1. Mass Loss

Figure 8 displays the mass loss rate of PCM concrete with varying amounts of RHA and FA replacements after cycles of freeze–thaw. The inclusion of RHA and FA generally reduced mass loss. The mass loss rate dropped with increasing RHA and FA amounts through 50 freeze–thaw cycles, all of which showed lower mass loss than the control group. At the same RHA replacement level, higher FA content resulted in smaller mass loss. The mass loss of PCM concrete tended to rise with increased FA replacement through 100 freeze–thaw cycles. Through 150 and 200 freeze–thaw cycles, some specimen groups showed mass loss rates approaching or exceeding that of the control group. Among these, groups with 10% FA replacement exhibited smaller mass loss rates, which decreased further with higher RHA replacement. Increasing the content of RHA and FA was helpful in the beginning phases of the freeze–thaw cycles to reduce mass loss. But as the cycles went on and the damage intensified, the pores in the concrete developed further, and the impact of concrete strength on mass loss became more significant. The beneficial effects of RHA and FA on freeze resistance were insufficient to offset the lack of strength. Consequently, after 200 freeze–thaw cycles, the concrete with 15% RHA and 10% FA replacement demonstrated the best freeze resistance due to the combined influence of its strength and thermal effects. The macroscopic appearance of this group compared to the control group after freeze–thaw cycles is shown in Figure 9.

Through the first 50 freeze–thaw cycles, the concrete had few pores, and the damage was not severe. At this stage, the thermal effect of the concrete primarily controlled the mass loss, and the addition of RHA and FA enhanced the concrete’s freeze resistance. However, as freeze–thaw damage intensified, the pores in the concrete continued to develop, and the impact of the concrete’s strength on mass loss gradually increased. The beneficial effect of RHA and FA on freeze resistance was insufficient to offset the lack of strength. Concrete with high compressive strength can better resist internal stresses and crack propagation over cycles of freeze–thaw [36]. Consequently, because of the combined influence of its strength and thermal effects, the concrete with 15% RHA and 10% FA replacement demonstrated the strongest freeze resistance in over 200 freeze–thaw cycles.

3.5.2. RDEM Loss

Figure 10 illustrates the RDEM losses of PCM concrete with varying replacement amounts of RHA and FA. During the initial 50 cycles, the differences in RDEM loss among the groups were minimal. Through 100 freeze–thaw cycles, the variations became more pronounced, indicating differing rates of loss. Following 100 cycles, the control group’s RDEM dropped to 90%, and after 200 cycles, it was at 66%, indicating significant damage. In contrast, the concrete with RHA and FA additions exhibited varied RDEM losses compared to the control group, due to the different temperature control effects exerted by the varying replacement amounts of RHA and FA.

The results indicated that with 10% and 15% RHA replacement there was a more pronounced effect on reducing internal damage to PCM concrete. At the beginning of the freeze–thaw cycles, adding RHA and FA improved RDEM loss. However, through 100 freeze–thaw cycles, the concrete groups with RHA and FA showed varying degrees of accelerated RDEM loss. Through 150 cycles, the control group’s RDEM dropped to 80.7%, and after 200 cycles, it further decreased to 65.7%. The experimental results found that with 15% RHA and 10% FA replacement, the concrete exhibited the highest RDEM and best freeze resistance, with a 25% lower RDEM loss rate compared to the control group. This was also consistent with the mass loss rate results.

3.5.3. Strength Loss

Figure 11 illustrates the strength loss of PCM concrete with varying substitution amounts of RHA and FA. Some specimens managed to lessen the loss of mechanical properties compared to the control group. The control group exhibited crucial strength loss over cycles of freeze–thaw, with strength dropping to 91.8% over 100 cycles and further decreasing to 71.5% over 200 cycles, indicating a strength loss exceeding 25%. In contrast, the strength loss of PCM concrete with RHA and FA varied due to the different effects of the replacement amounts. The addition of RHA and FA improved the strength of PCM concrete during the initial phases of freeze–thaw cycles. Through 100 cycles, the differences between the groups with varying replacement amounts of RHA and FA became more apparent. When the replacement rate was 15% RHA and 10% FA, the strength loss was minimal, maintaining 76.4% of the strength after 200 cycles. However, an FA replacement rate higher than 10% negatively impacted freeze–thaw resistance, resulting in greater strength loss compared to the control group over 200 cycles.

3.6. Evaluation of NMR

In this study, we have depicted the T2 spectrum of concrete with various FA and RHA contents after 28 d of curing in Figure 12. These spectra, based on NMR technology, reveal detailed information about the concrete’s pore structure. The signal intensity is directly proportional to the porosity, and the T2 relaxation time is correlated with the size of the pores, making the precise quantification of pore sizes possible [37,38]. We identified four types of pores, large pores, capillary pores, transition pores and gel pores, each with distinct size ranges and T2 values. The T2 spectrum of all samples observed three peaks, corresponding to these types of pores. In particular, the high signal intensity of Peak 1 indicates that gel and transition pores are the primary types of pores in concrete.

As shown in Figure 12, with the same RHA content, an increase in the signal intensity of Peak 1 was observed as the FA content increased, indicating a higher proportion of gel and transition pores. As FA content rises, the amount of unreacted FA gradually increases because the volume occupied by the pozzolanic reaction products is smaller than the starting FA, leaving a larger proportion of large pores [39,40]. Further analysis of RHA’s effect on concrete porosity at the same FA content revealed that when both RHA and FA were at 10%, the signal intensity of Peak 1 was the lowest and the total spectral area was also the smallest. This is consistent with its mechanical performance [41,42].

The NMR T2 spectrum in Figure 13 further reveals the freeze–thaw cycles’ effects on the concrete’s pore structure. The reduction in the micropore peak area corresponds to an increase in the meso and macropore peak areas, indicating that freeze–thaw cycles promote the development of micropores into larger pores. The rightward shift of the highest intensity point and the analysis of concrete damage over cycles of freeze–thaw indicate that the control concrete experiences severe internal damage, particularly due to meso and macropores. In contrast, when the content of RHA and FA is 10%, the concrete shows better mechanical properties, with a relatively slower increase in meso and macropores, confirming the potential advantages of RHA and FA in improving the mechanical properties of PCM concrete.

4. Conclusions

This study systematically explores the application of RHA and FA as cement replacement materials in multi-stage PCM concrete for temperature control in building structures in cold regions. The experimental results indicate that selecting appropriate replacement ratios of RHA and FA can significantly enhance the thermal inertia and temperature regulation capability of concrete while maintaining sufficient mechanical performance, providing an innovative solution for concrete structures in cold regions.

(1)

Balancing Mechanical and Thermal Performance: This study finds that an appropriate proportion of RHA and FA as cement replacements can effectively enhance the compressive strength of concrete while reducing thermal diffusivity and thermal conductivity. The results show that the compressive strength of concrete is the highest when the content of RHA and FA is 10%. This suggests that by precisely adjusting the replacement amounts of RHA and FA, significant improvements in the thermal properties of concrete can be achieved without sacrificing mechanical performance.

(2)

Thermal Regulation Capability: Incorporating RHA and FA enhances multi-stage PCM concrete’s capacity to regulate temperature. When the RHA content is 15% and the FA content is 30%, the highest temperature regulation capability can reach up to 5 °C. This capability is particularly important during environmental temperature fluctuations, helping to reduce concrete damage caused by temperature changes.

(3)

Pore Structure Optimization: The pore structure of the concrete was analyzed in detail using nuclear magnetic resonance technology. The results showed that the addition of RHA and FA altered the porosity of the concrete, with the best mechanical properties observed when the RHA and FA content was 10%.

(4)

Freeze Resistance: By simulating freeze–thaw environments, this study investigated the impact of different RHA and FA contents on the freeze resistance of concrete. The results show that when the RHA content is 15% and the FA content is 10%, the concrete exhibits the least mass loss and strength loss, demonstrating the best freeze resistance.

(5)

Environmental and Economic Benefits: The utilization of RHA and FA not only improves the durability of concrete but also reduces production costs and decreases the amount of cement used, offering significant environmental benefits.