Low-heat high-anti-crack cementitious material for mass concrete of pumped storage power station

Abstract

The application provides a low-heat high-anti-crack cementitious material for pumped storage power station mass concrete, relates to the technical field of building materials, and comprises low-heat Portland cement, active metakaolin, sulfate adjusting components, crystal nucleus filling components, grinding aids and temperature peak memory reverse activation self-limiting compensation function particles. The function particles are composed of a thermal memory core, a liquid storage and slow-release ring, a low-modulus decoupling film and a reverse reaction shell, can release water in a directional manner and trigger a shell compensation reaction after the temperature rise of the concrete exceeds the melting threshold and falls back to the crystallization threshold, realize peak cutting and heat reduction, post-peak internal curing and staged compensation. The cementitious material can significantly reduce the adiabatic temperature rise peak value, reduce its own shrinkage and prolong the restrained cracking time, and is suitable for pumped storage power station dams, underground powerhouses, volute foundations and pressure pipelines and other mass concrete structures. The preparation process is simple, and the cementitious material is convenient for factory production and engineering application.

Description

Technical Field

[0001] This invention relates to the field of building materials technology, specifically to a low-heat, high-crack-resistant cementitious material for large-volume concrete in pumped storage power stations. Background Technology

[0002] Pumped storage power stations are crucial facilities for building new power systems. Their core structures, such as dams, pressure pipelines, spiral casing foundations, and underground powerhouses, are all constructed using large-volume concrete. These structures are enormous, and the concentrated heat release from cement hydration after concrete pouring causes a rapid rise in internal temperature, followed by slow cooling. During this process, the temperature stress generated by the temperature difference between the inside and outside of the concrete, and the shrinkage stress caused by subsequent cooling, can easily lead to concrete cracking.

[0003] The presence of cracks can seriously compromise the integrity, durability, and impermeability of a structure. This is especially true for energy storage power station structures with high pressurized water head and repeated hydraulic loads, where cracks can lead to leakage, steel corrosion, and even structural safety issues, with high costs associated with subsequent treatment.

[0004] Currently, intermediate-heat silicate cement is commonly used in engineering projects, supplemented by cooling water pipes and other measures to control temperature cracks. However, the heat of hydration of intermediate-heat cement remains high, and its crack resistance is still insufficient to fully meet the requirements of ultra-high and ultra-thick structures. Although existing technologies have introduced methods to reduce the heat of hydration by adding admixtures such as fly ash and mineral powder, these materials have low early-stage activity, which may affect the development of early strength. In addition, there are technologies that use magnesium oxide expansive agents to compensate for shrinkage, but its expansion process is often difficult to precisely match with the temperature drop and shrinkage process of concrete, and excessive addition may lead to excessive expansion in the later stages.

[0005] Therefore, developing a special cementitious material that can significantly reduce the heat of hydration, actively compensate for shrinkage, and ensure good mechanical properties is a key technical requirement for solving the cracking problem of large-volume concrete in pumped storage power stations. Summary of the Invention

[0006] Technical problems to be solved

[0007] To address the shortcomings of existing technologies, this invention provides a low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations, thus solving the problems of existing technologies.

[0008] Technical solution

[0009] To achieve the above objectives, the present invention provides the following technical solution: a low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations, comprising low-heat silicate cement, active metakaolin, sulfate adjusting component, crystal nucleus filling component, and grinding aid. Based on the total mass of the cementitious material, the cementitious material further comprises 5.0%–10.5% of temperature peak memory reverse-flow activated self-limiting compensation functional particles; the active metakaolin comprises 6.0%–10.0%, the sulfate adjusting component comprises 1.5%–3.5%, the crystal nucleus filling component comprises 0.5%–2.5%, the grinding aid comprises 0.1%–0.3%, and the balance is low-heat silicate cement.

[0010] The temperature peak memory reverse activation self-limiting compensation functional particle includes, from the inside out, a thermal memory core, a liquid storage slow-release ring, a low modulus decoupling membrane, and a reverse reaction shell.

[0011] The thermal memory core is a modified fly ash cenosphere loaded with phase change core material. The liquid storage slow-release ring includes alkali-resistant water-absorbing resin, silica-alumina filler and stabilizing components. The low-modulus decoupling membrane is coated on the outer surface of the liquid storage slow-release ring. The reverse reaction shell includes lightly calcined magnesium oxide, sulfoaluminate clinker, activated metakaolin, anhydrous gypsum, lithium silicate or silica sol and nano calcium carbonate.

[0012] Preferably, the sulfate adjusting component is anhydrous gypsum and / or hemihydrate gypsum, and the crystal nucleus filling component is limestone powder and / or nano-calcium carbonate.

[0013] Preferably, the particle size of the temperature peak memory reverse activation self-limiting compensation functional particles is 80-250 μm; based on the total mass of the functional particles, the thermal memory core accounts for 30%-45%, the liquid storage slow-release ring accounts for 8%-18%, the low modulus decoupling membrane accounts for 0.5%-3.0%, and the remainder is the reverse reaction shell.

[0014] Preferably, the modified fly ash cenospheres have a particle size of 50–150 μm and a cavity ratio of 60%–75%, and are obtained by washing fly ash cenospheres with dilute acid until neutral and then activating them at 350–450°C.

[0015] The thermal memory core has a melting initiation temperature of 34–38°C, a crystallization initiation temperature of 29–33°C, a thermal hysteresis window of 3–8°C, and an apparent latent heat of phase transition of 35–65 J / g.

[0016] Preferably, the liquid storage slow-release ring comprises alkali-resistant water-absorbing resin, active metakaolin and / or nano-silica, and calcium hydroxide and / or lithium salt stabilizing components; in simulated cement pore liquid, the liquid absorption ratio of the liquid storage slow-release ring is 4 to 10 g / g.

[0017] Preferably, before the temperature peak memory reverse activation self-limiting compensation function particles experience a maximum temperature of not less than 34°C and subsequently drop back to a temperature of not more than 33°C, the cumulative water release rate in 24 hours is not higher than 20% of the total liquid volume; after experiencing the thermal return line, the cumulative water release rate in 12 to 72 hours is 50% to 85% of the total liquid volume.

[0018] Preferably, the low-modulus decoupling membrane is an organosilicon-modified membrane or a siloxane-inorganic hybrid membrane with a thickness of 0.5–2.0 μm.

[0019] Preferably, by mass, the reverse reaction shell comprises 45-65 parts of lightly calcined magnesium oxide, 10-25 parts of sulfoaluminate clinker, 5-15 parts of activated metakaolin, 4-10 parts of anhydrous gypsum, 3-8 parts of lithium silicate or silica sol, and 1-5 parts of nano-calcium carbonate, wherein the activity index of the lightly calcined magnesium oxide is 140 s-220 s.

[0020] Preferably, the cementitious material is prepared by the following method:

[0021] First, premix and grind low-heat silicate cement, active metakaolin, sulfate conditioning component, crystal nucleation filler component and grinding aid to a specific surface area of ​​340-420 m². 2 / kg,

[0022] The temperature peak memory reverse-flow activated self-limiting compensation functional particles are then mixed with the premixed powder using a plow-type mixer or a twin-ribbon mixer under low shear for 3 to 10 minutes, and the temperature peak memory reverse-flow activated self-limiting compensation functional particles are not involved in high-energy ball milling.

[0023] Preferably, based on the total mass of the cementitious materials, the active metakaolin is 7.5% to 9.5%, the sulfate adjusting component is 2.0% to 3.0%, the crystal nucleus filling component is 1.0% to 2.0%, the grinding aid is 0.15% to 0.25%, the temperature peak memory reverse-process activated self-limiting compensation functional particles are 7.0% to 9.0%, and the balance is low-heat silicate cement.

[0024] Beneficial effects

[0025] This invention provides a low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations. It possesses the following beneficial effects:

[0026] 1. This invention introduces temperature peak memory and reverse-flow activation self-limiting compensation functional particles, composed of a thermal memory core, a liquid storage slow-release ring, a low-modulus decoupling membrane, and a reverse-flow reaction shell, into a low-heat silicate cement matrix. This enables the cementitious material to not only reduce the temperature rise rate and adiabatic temperature rise peak during the heating stage by relying on the latent heat of phase change absorption of the thermal memory core, but also to activate the liquid storage slow-release and shell compensation reactions only after the internal temperature of the concrete crosses the melting threshold and falls back to the crystallization threshold. This allows for the identification and response to the true thermal history of large-volume concrete. Compared with solutions that rely solely on low-heat cement or ordinary phase change particles, this invention not only reduces the peak temperature but also transforms the critical stage from a "high-temperature rise" to a "low-peak, slow-fall" temperature evolution path, significantly reducing the internal and external temperature difference and temperature stress.

[0027] 2. The synergistic effect of the liquid-storage slow-release ring and the low-modulus decoupling membrane in this invention enables the functional particles to maintain structural stability during the heating phase without premature water release. After the post-peak thermal loop is completed, the liquid in the liquid-storage slow-release ring is released directionally. The released liquid preferentially acts on the matrix surrounding the particles and the reverse reaction shell, effectively alleviating the problems of local self-drying and rapid decrease in internal relative humidity after the peak. On this basis, it triggers the rapid micro-expansion reaction of sulfoaluminate clinker and anhydrous gypsum in the reverse reaction shell, as well as the hysteretic compensation reaction of lightly calcined magnesium oxide. Thus, this invention unifies internal curing and shrinkage compensation into the same thermal process triggering system, enabling the establishment of a local micro-pre-compression field during the most sensitive stages of cooling and shrinkage stress, significantly reducing the risk of autogenous shrinkage and constrained cracking.

[0028] 3. The reverse reaction shell of this invention employs a composite design of lightly calcined magnesia, sulfoaluminate clinker, activated metakaolin, anhydrous gypsum, lithium silicate or silica sol, and nano-calcium carbonate. This allows the shell to first form a rapid micro-pre-compression dominated by AFt after the peak, followed by mid-to-late stage hysteresis compensation dominated by Mg(OH)2. Simultaneously, activated metakaolin, lithium silicate or silica sol, and nano-calcium carbonate promote the densification of the shell-matrix interface and gradually backfill the microchannels during the reaction, achieving self-limiting compensation. Compared with the free addition of MgO or ordinary expansion agents, the compensation reaction of this invention is more concentrated in the stress-sensitive region of the particle-matrix interface, with a more matched compensation timing and a lower risk of excessive expansion in the later stage. It can significantly extend the time to ring-constrained cracking and improve structural durability while maintaining high 28-day and 90-day strength. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the microstructure of the functional particles of the present invention;

[0030] Figure 2 This is a schematic diagram of the components and action pathways of the present invention;

[0031] Figure 3 This is a flowchart illustrating the preparation process of the present invention;

[0032] Figure 4 This is a comparison diagram of the adiabatic temperature rise curves of the present invention;

[0033] Figure 5 This is the strain evolution diagram after the thermal loop peak and after compensation according to the present invention;

[0034] Figure 6 This is a comparison chart of the cracking time of the annulus according to the present invention;

[0035] Figure 7 This is a comparison chart of 90-day drying shrinkage according to the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Specific Implementation Example 1:

[0038] like Figures 1 to 7 As shown, a low-heat, high-crack-resistant cementitious material for large-volume concrete in pumped storage power stations includes...

[0039] A low-heat, high-crack-resistant cementitious material for large-volume concrete in pumped storage power stations includes low-heat silicate cement, active metakaolin, sulfate adjusting components, crystal nucleation filler components, and grinding aids. Based on the total mass of the cementitious material, it further includes 5.0%–10.5% of temperature peak memory reverse-flow activated self-limiting compensation functional particles; active metakaolin comprises 6.0%–10.0%, sulfate adjusting components 1.5%–3.5%, crystal nucleation filler components 0.5%–2.5%, grinding aids 0.1%–0.3%, and the balance being low-heat silicate cement.

[0040] The temperature peak memory reverse-flow activation self-limiting compensation functional particle consists of, from the inside out, a thermal memory core, a liquid storage slow-release ring, a low-modulus decoupling membrane, and a reverse-flow reaction shell.

[0041] The thermal memory core is a modified fly ash cenosphere loaded with phase change core material. The liquid storage slow-release ring includes alkali-resistant water-absorbing resin, silica-alumina filler and stabilizing components. A low-modulus decoupling membrane is coated on the outer surface of the liquid storage slow-release ring. The reverse reaction shell includes lightly calcined magnesium oxide, sulfoaluminate clinker, activated metakaolin, anhydrous gypsum, lithium silicate or silica sol and nano calcium carbonate.

[0042] This invention addresses four key issues in the development of pumped-storage power station large-volume concrete: high temperature peak during the heating phase, superposition of self-drying and temperature drop shrinkage after the peak, sudden increase in tensile stress under strong constraints, and mismatch in the timing of traditional expansion compensation activation. The development process first determined that the main cementitious system must be based on low-heat silicate cement. This is because when the low-heat silicate cement content is below 78%, although the cementitious system can still continue to suppress heat release through high mineral admixtures, the formation rate of 3-day strength and post-peak elastic modulus decreases significantly, making it difficult to provide a sufficiently continuous matrix framework for subsequent micro-pre-compression after particle triggering. When the low-heat silicate cement content is above 90%, the total heat release and 24-hour heat release rate of the matrix increase significantly, weakening the core temperature rise control advantage in large-volume components of pumped-storage power stations. Based on the screening results of 36 cementitious systems, the overall ratio of functional particles to the matrix was ultimately controlled within the aforementioned range.

[0043] During the research and development process, based on a total cementitious material content of 1000g, five gradients were set for the active metakaolin: 4%, 6%, 8%, 10%, and 12%; four gradients were set for the sulfate adjusting component: 1%, 2%, 3%, and 4%; five gradients were set for the crystal nucleation filling component: 0%, 0.5%, 1.5%, 2.5%, and 4%; and six gradients were set for the functional particles: 0%, 5%, 7%, 8.5%, 10.5%, and 12%. A combination of orthogonal and quadratic response surface methodology was used for screening. Under standard slurry conditions, the water-cement ratio was fixed at 0.32, and the water-reducing agent dosage was fixed at 0.18% of the total cementitious material content. The results showed that when the reactive metakaolin content was 8%, the 3-day compressive strength was 27.4 MPa, the 7-day heat of hydration was 204 kJ / kg, and the cumulative exothermic peak was delayed by 4.6 hours over 72 hours. When the content was increased to 10%, the 3-day compressive strength further increased to 28.3 MPa, but the standard consistency water requirement increased from 25.8% to 27.1%, and the slurry viscosity increased significantly. When the content reached 12%, although the 7-day heat of hydration decreased slightly to 198 kJ / kg, the fluidity loss was significant, and the 1-hour spread retention rate decreased from 92% to 78%. Therefore, controlling the reactive metakaolin content within the range of 6.0% to 10.0% can balance low heat, early strength, and construction adaptability. When the sulfate adjusting component is in the range of 2.0%–3.0%, the initial setting time is controlled at 248–286 min, and the final setting time is controlled at 332–389 min. This ensures sufficient sulfate ions are available after activation following the reverse reaction shell peak, without any abnormal accelerated setting. When the sulfate content is below 1.5%, the first-stage expansion of the shell layer after the peak is insufficient, with a compensating strain of only 62 × 10⁻⁶ within 48 hours. When the sulfate content is above 3.5%, some samples show excessive micro-expansion after 28 days, indicating unsatisfactory volume stability. When the nucleus filling component is in the range of 1.0%–2.0%, the isothermal calorimetric peak is advanced by 0.8–1.6 hours, but the peak height does not increase significantly, indicating that it mainly plays a role in nucleation and structural optimization rather than as an additional heat source. Below 0.5%, its effect is not obvious; above 2.5%, the strength improvement at 28 days tends to plateau, while the unit cost and fine powder content increase significantly. The functional particles showed the best overall effect when the content was between 7.0% and 9.0%, with the 8.0% group having the lowest peak adiabatic temperature rise and the most balanced post-peak compensation.

[0044] To demonstrate that the overall technical solution is not a simple parallel addition but has a phased and sequential effect, this invention sets up four comparative examples: Comparative example A is a system containing only low-heat cement + metakaolin + gypsum + limestone powder, without functional particles; Comparative example B is a low-heat matrix + ordinary single-layer cenosphere phase change particles, without liquid storage slow-release ring, without decoupling membrane, and without counter-processing reaction shell; Comparative example C is a low-heat matrix + freely added MgO and sulfoaluminate clinker powder, but without constructing multi-layer particles; Comparative example D is a four-layer particle system but without low-modulus decoupling membrane. Adiabatic temperature rise, ring cracking, autogenous shrinkage, programmed temperature-controlled water release, and post-peak compensation tests were conducted using the same concrete mix proportions. The results showed that the peak adiabatic temperature rise of the system implemented in this invention was 38.9℃, a decrease of 8.7℃ compared to 47.6℃ in Comparative Example A, and a decrease of 3.9℃ compared to 42.8℃ in Comparative Example B. The cumulative compensation strain of the system implemented in this invention after 72 hours following the programmed temperature-controlled peak was 218 × 10⁻⁶, while Comparative Example A was only 14 × 10⁻⁶, Comparative Example B was 21 × 10⁻⁶, Comparative Example C was 126 × 10⁻⁶, and Comparative Example D was 79 × 10⁻⁶. The ring cracking time was 63 days without cracking in the system implemented in this invention, compared to 18 days for Comparative Example A, 29 days for Comparative Example B, 41 days for Comparative Example C, and 34 days for Comparative Example D. This indicates that the four-layer structure of the particles and their proportion in the system jointly determine the post-peak crack control capability.

[0045] The sulfate conditioning component is anhydrous gypsum and / or hemihydrate gypsum, and the crystal nucleus filling component is limestone powder and / or nano-calcium carbonate.

[0046] During the research and development phase, anhydrous gypsum, hemihydrate gypsum, dihydrate gypsum, and anhydrous / hemihydrate gypsum composite systems were compared for the sulfate-adjusting component. The test conditions were: total cementitious material 1000g, 8% activated metakaolin, 8% functional particles, water-cement ratio 0.32, isothermal calorimetry test temperature 20℃, and compressive strength testing according to GB / T17671 mortar method. The experimental results show that: when using only dihydrate gypsum, the initial setting time is 302 min, the 7-day heat of hydration is 211 kJ / kg, and the micro-expansion strain after peak activation (48 h) is only 71 × 10^-6; when using only hemihydrate gypsum, the initial setting time is shortened to 219 min, and the 3-day compressive strength is 29.1 MPa, but the increase in post-peak micro-expansion strain after 7 days is insufficient, indicating that sulfate release is premature; when using only anhydrous gypsum, the initial setting time is 289 min, the 3-day compressive strength is 25.8 MPa, and the 48-h micro-expansion strain is 138 × 10^-6, indicating sufficient compensation in the first stage after peak; when using anhydrous gypsum / hemihydrate gypsum at a mass ratio of 6:4, the initial setting time is 256 min, the final setting time is 348 min, the 3-day compressive strength is 28.0 MPa, and the 48-h post-peak micro-expansion strain is 152 × 10^-6, showing the best overall effect. Therefore, anhydrous gypsum and / or hemihydrate gypsum should be preferred.

[0047] Regarding the nucleus-filling components, limestone powder, nano-calcium carbonate, and a mixture of both were compared. Limestone powder had a D50 of 7.2 μm and a specific surface area of ​​568 m². 2 / kg; the average particle size of nano-calcium carbonate is 65nm. Results showed that when only 1.5% limestone powder was added, the 24h exothermic peak of the slurry was advanced by 0.7h, and the 3d strength increased by 1.8MPa; when only 1.5% nano-calcium carbonate was added, the 24h exothermic peak was advanced by 1.4h, and the 3d strength increased by 2.9MPa, but the slurry fluidity decreased rapidly at 30min; when using 1.2% limestone powder + 0.4% nano-calcium carbonate, the 24h exothermic peak was advanced by 1.1h, the 3d strength increased by 2.6MPa, and the fluidity retention was better than the pure nano-calcium carbonate group. SEM observation showed that the CSH gel distribution at the particle-matrix interface was more uniform. Therefore, the preferred nucleation filling component is limestone powder and / or nano-calcium carbonate.

[0048] The particle size of the thermal memory reverse-flow activation self-limiting compensation functional particles is 80–250 μm. By total mass of functional particles, the thermal memory core accounts for 30%–45%, the liquid storage slow-release ring accounts for 8%–18%, the low-modulus decoupling membrane accounts for 0.5%–3.0%, and the remainder is the reverse-flow reaction shell.

[0049] In this invention, the particle size and layer ratio of functional particles directly determine the thermal memory capability, water release timing, and post-peak compensation capability. During development, 18 groups of functional particles with different particle size ranges and layer ratios were prepared. The particle size ranges were set to 40–80 μm, 80–120 μm, 120–180 μm, 180–250 μm, and 250–350 μm. The results showed that although the 40–80 μm group had good particle dispersion, its apparent latent heat of phase change (LST) was only 21.6 J / g, and the water release rate after 72 h thermal regression was only 34.2%, indicating insufficient core volume. The 80–120 μm group showed an increased apparent LST of 37.8 J / g, with a water release rate of 57.3% after thermal regression. The 120–180 μm group had an apparent LST of 48.1 J / g, a water release rate of 69.4%, and a post-peak compensation strain of 214 × 10⁻⁶, exhibiting the best overall performance. The 180–250 μm group had an apparent LST of 52.6 J / g and a water release rate of 74.8%, but the 28-day compressive strength of the concrete decreased slightly. The 250–350 μm group showed a particle breakage rate of 19.7% after simulated pumping cycles, which is unfavorable for engineering applications. Therefore, an overall particle size of 80–250 μm is preferred.

[0050] The mass ratios of the thermal memory core, the liquid storage slow-release ring, the low-modulus decoupling membrane, and the reverse reaction shell were also systematically screened. The thermal memory core ratios were set to five levels: 20%, 30%, 38%, 45%, and 55%. The results showed that: the apparent latent heat of phase change in the 20% group was only 18.9 J / g, with a limited decrease in the peak adiabatic temperature rise; the latent heat in the 30% group reached 34.7 J / g; the latent heat in the 38% group reached 47.9 J / g, with a peak temperature reduction of 7.6℃; the latent heat in the 45% group reached 58.3 J / g, but the post-peak compensation decreased slightly when the shell thickness was insufficient; and in the 55% group, due to the excessively thin outer structure, the post-peak compensation strain decreased significantly to 133 × 10^-6. The proportions of the reservoir-release ring were set at five levels: 5%, 8%, 12%, 18%, and 25%. The 12% group showed the best post-thermal return water release rate and post-peak compensation effect, reaching 67.1% and 221×10^-6, respectively. Although the 25% group had a high water release rate, its 28-day compressive strength decreased significantly. Comparisons were made under conditions where the proportion of the low-modulus decoupling membrane was 0.3%, 0.8%, 1.6%, 3.0%, and 4.5%. The 0.8%–1.6% range showed the highest particle integrity rate during the heating stage and the most sensitive post-peak water release triggering. Therefore, the aforementioned proportion range was determined.

[0051] Statistical analysis of the microstructure showed that the particle cross-section was most clearly defined and the shell continuity was best when the thermal memory core accounted for 38%, the liquid storage and slow-release ring for 12%, the low-modulus decoupling membrane for 1.2%, and the reverse reaction shell for 48.8%. FIB-SEM measurements revealed an average equivalent thickness of 46 μm for the thermal memory core, 11 μm for the liquid storage and slow-release ring, 1.1 μm for the decoupling membrane, and 17 μm for the reverse reaction shell. This structure exhibited a typical "low water release before the peak and high water release after the peak" characteristic in the temperature-controlled programmed experiment.

[0052] The modified fly ash cenospheres have a particle size of 50–150 μm and a cavity rate of 60%–75%. They are obtained by washing fly ash cenospheres with dilute acid until neutral and then activating them at 350–450 °C.

[0053] The melting initiation temperature of the thermal memory core is 34–38℃, the crystallization initiation temperature is 29–33℃, the thermal hysteresis window is 3–8℃, and the apparent latent heat of phase transition is 35–65 J / g.

[0054] Modified fly ash cenospheres were obtained from power plant cenospheres, sieved to obtain particles of 50–150 μm. These particles were washed with a 5% hydrochloric acid solution at 60°C for 35 min (liquid-to-solid ratio 6:1), then washed with deionized water until the filtrate pH reached 6.8–7.2. The cenospheres were then dried at 105°C for 4 h and subsequently activated at 400°C for 1.5 h to obtain the modified cenospheres. A comparison before and after activation showed that the BET specific surface area of ​​the cenospheres increased from 3.9 m² / s². 2 / g increased to 6.7m 2 / g, the average pore size increased from 78nm to 112nm, and EDS analysis showed a significant reduction in surface free calcium and iron impurities. The crushing strength decreased slightly from 2.4MPa to 2.2MPa, but still met the particle preparation requirements. If the activation temperature was increased to 500℃, some cenospheres experienced shell embrittlement, and the crushing strength dropped to 1.5MPa, making them no longer preferred.

[0055] The thermal memory core was supported by a composite paraffin core material. Industrial-grade paraffin with a melting point of 35℃ was used as the main core material, with 8wt% high-melting-point paraffin and 1.5wt% fatty acid amide crystallization regulators added, along with 1.2wt% carboxylated multi-walled carbon nanotubes as a thermal conductivity enhancement component. Modified cenospheres were immersed in a molten composite phase change liquid at 72℃, vacuumed at -0.093MPa for 30 min, then pressurized to 0.45MPa and held for 45 min. After depressurization, the surface liquid was removed through a 100-mesh filter, and the core was allowed to stand at 42℃ for 20 min to obtain the thermal memory core. DSC testing conditions were: 8mg sample, N2 atmosphere, heating / cooling rate 5℃ / min, test range 15–55℃. The measured melting initiation temperature was 35.4℃, crystallization initiation temperature was 31.0℃, thermal hysteresis window was 4.4℃, and apparent latent heat of phase change was 48.6 J / g. The control group without crystallization regulator had a melting initiation temperature of 34.8℃ and a crystallization initiation temperature of 33.6℃, ​​with a thermal hysteresis window of only 1.2℃. The water release behavior after the thermal loop was not obvious, indicating that the establishment of a thermal hysteresis window is necessary.

[0056] To verify the necessity of melting initiation temperatures of 34–38℃ and crystallization initiation temperatures of 29–33℃, three sets of thermal memory cores were further established: a low-temperature group (melting initiation temperature 29.7℃, crystallization initiation temperature 27.8℃), a target group (35.4℃, 31.0℃), and a high-temperature group (40.6℃, 35.8℃). Under simulated programmed temperature control conditions for large-volume concrete, the low-temperature group began to absorb a large amount of heat and release water prematurely in the early stage of temperature rise, resulting in insufficient residual liquid volume after the peak; the high-temperature group was not fully triggered in some samples, and the post-peak compensation was significantly insufficient; the target group was able to complete melting near the temperature peak and initiate water release and compensation when the temperature dropped to around 31℃ after the peak, showing the best effect.

[0057] The liquid storage slow-release ring includes alkali-resistant water-absorbing resin, active metakaolin and / or nano silica, and calcium hydroxide and / or lithium salt stabilizing components; in simulated cement pore liquid, the liquid absorption ratio of the liquid storage slow-release ring is 4 to 10 g / g.

[0058] The liquid-storing slow-release ring is constructed from an alkali-resistant acrylate / acrylamide copolymer superabsorbent polymer, activated metakaolin, nano-silica, and calcium hydroxide. The alkali-resistant superabsorbent polymer accounts for 42% of the ring's solid mass, activated metakaolin 34%, nano-silica 14%, calcium hydroxide 8%, and lithium salt stabilizer 2%. In preparation, the resin precursor is first mixed with nano-silica and the lithium salt stabilizer, then metakaolin and calcium hydroxide are introduced to form a rheology-stabilized slurry. The slurry's solid content is controlled at 31%, and its viscosity is 420 mPa·s. This slurry is then fluidized bed sprayed onto the surface of a thermally shaped core and cured at 55°C for 3 hours to obtain the liquid-storing slow-release ring.

[0059] The liquid absorption ratio test was conducted using two liquid environments: deionized water and simulated cement pore liquid. The simulated pore liquid consisted of saturated Ca(OH)₂ solution + 0.2 mol / L NaOH + 0.15 mol / L KOH. The test results showed that the liquid absorption ratio of the reservoir-release ring was 18.6 g / g in deionized water after 24 hours, while it was 6.8 g / g in the simulated pore liquid, both within the target range. Control group 1, using ordinary superabsorbent resin without metakaolin or nano-silica, achieved a high liquid absorption ratio of 31.2 g / g in deionized water, but rapidly collapsed to only 2.1 g / g in the simulated pore liquid. Control group 2, without calcium hydroxide or lithium salt stabilizers, achieved a liquid absorption ratio of 4.0 g / g in the simulated pore liquid, but significant liquid instability and precipitation occurred after 12 hours. Control group 3, without nano-silica but with only metakaolin, achieved a liquid absorption ratio of 5.4 g / g, but after peak release, a large residual cavity remained. The above results demonstrate that the composition of the reservoir slow-release ring is not arbitrary.

[0060] Furthermore, LF-NMR was used to test the water content of the reservoir before and after the thermal loop was triggered. The results showed that the free water peak area of ​​the reservoir changed very little before the thermal loop was triggered. After the thermal loop was triggered at 20℃→40℃→30℃, the free water peak area of ​​the reservoir decreased by 58% within 24 hours, which was basically consistent with the 61% water release rate calculated by the mass method. SEM observation showed that local contraction channels formed in the reservoir 48 hours after the peak, and hydration products gradually filled them, which is consistent with the synchronous activation of the reverse reaction shell.

[0061] Before the temperature reaches a maximum of 34°C and then drops back to 33°C, the cumulative water release rate of the particles with the peak temperature memory reverse activation self-limiting compensation function is no more than 20% of the total liquid volume in 24 hours; after reaching the thermal return line, the cumulative water release rate is 50% to 85% of the total liquid volume in 12 to 72 hours.

[0062] This invention designs a specialized programmed temperature-controlled water release experiment to address this behavioral characteristic. The experiment was conducted in four groups: Group A was kept at a constant temperature of 20℃; Group B was heated to 32℃ and then cooled back to 30℃; Group C was heated to 40℃ but did not cool back; Group D was heated to 40℃ and held for 8 hours before cooling back to 30℃. In each group, the functional particles were pre-treated to the same liquid storage state, and then placed in a constant humidity environment to test the water release rate over 72 hours. The results are as follows: Group A had a cumulative water release rate of 8.7% over 24 hours and 14.2% over 72 hours; Group B had a rate of 12.4% over 24 hours and 18.3% over 72 hours, indicating that water release was still limited before the melting threshold was crossed; Group C had a rate of 17.9% over 24 hours and 24.8% over 72 hours, indicating that water release was not significantly initiated when only heating and melting were performed but the return cycle was not completed; Group D had a rate of 18.6% over 24 hours, but the cumulative water release rate reached 64.7% within 12–72 hours after the thermal return cycle was completed, showing a clear post-peak water release behavior. The control particles without the low-modulus decoupling membrane had a water release rate of only 31.5% after 72 hours under Group D conditions, indicating that the reverse-flow gate structure is a necessary condition for this behavior.

[0063] To demonstrate that this behavior is consistent with the actual temperature history of large-volume concrete, a 500mm adiabatic temperature rise chamber was used to simulate the actual temperature rise process of large-volume concrete, with temperature and resistivity sensors embedded in the center of the chamber. The sample containing the particles of this invention reached a maximum temperature of 39.2℃ at 58 hours, then dropped to 32.8℃ at 84 hours, with a significant decrease in resistivity during this corresponding stage, indicating an increase in locally migrateable water. In contrast, the resistivity changes of the control group without functional particles and containing only monolayer phase change particles were gradual, without exhibiting a similar post-peak drop characteristic. Simultaneous sampling and moisture content analysis showed that the local moisture content in the 0–100μm region surrounding the particles in the experimental group was 2.7 percentage points higher than that in the control group at 96 hours, indicating that post-peak water release did indeed occur.

[0064] The low-modulus decoupling membrane is an organosilicon-modified membrane or a siloxane-inorganic hybrid membrane with a thickness of 0.5–2.0 μm.

[0065] A low-modulus decoupling membrane was constructed using vinyltriethoxysilane, hydroxyl-terminated polydimethylsiloxane, and colloidal silica as a siloxane-inorganic hybrid membrane. The total solid content in the preparation solution was 3.6%, with a solid-to-silica ratio of 7:3. Particles with the formed reservoir-release rings were dipped into the membrane for 12 min, filtered, dried at 60 °C for 2 h, and then cured at room temperature for 12 h. SEM cross-sectional analysis showed an average membrane thickness of 1.08 μm and a standard deviation of 0.21 μm. DMA measurements revealed an equivalent tensile modulus of approximately 86 MPa and an elongation at break of approximately 24%.

[0066] In the control experiment, without the decoupling membrane, 11.3% of the particles developed localized shell microcracks during the temperature rise from 20℃ to 40℃. While the shell integrity rate was high with the high-modulus inorganic pure silica sol membrane, the water release rate after 72 hours of thermal runaway was only 28.6%, and the post-peak compensation strain was only 94×10^-6. With the low-modulus decoupling membrane of this invention, the particle integrity rate during the heating stage was 96.8%, the water release rate after 72 hours of thermal runaway was 64.7%, and the post-peak compensation strain reached 218×10^-6. This indicates that the decoupling membrane simultaneously performs the dual functions of "buffer protection during the heating stage" and "reverse activation during the cooling stage."

[0067] Further in-situ hot-stage microscopy was used to observe the interfacial behavior of the particles during the heating-cooling process. The results showed that the decoupling membrane of this invention did not exhibit significant instability above 35°C, and only a small amount of controllable fine interfacial debonding appeared on the outer layer of the particles when the temperature dropped back to around 31°C. In contrast, the control group without the membrane showed random shell cracking at around 36°C. This result is highly consistent with the programmed temperature-controlled water release experiment.

[0068] By mass, the reverse reaction shell comprises 45-65 parts of lightly calcined magnesium oxide, 10-25 parts of sulfoaluminate clinker, 5-15 parts of activated metakaolin, 4-10 parts of anhydrous gypsum, 3-8 parts of lithium silicate or silica sol solids, and 1-5 parts of nano-calcium carbonate. The activity index of the lightly calcined magnesium oxide is 140s-220s.

[0069] The preferred formulation for the reverse-flow reaction shell is: 56 parts light-calcined magnesium oxide, 18 parts sulfoaluminate clinker, 10 parts activated metakaolin, 7 parts anhydrous gypsum, 5 parts lithium silicate solids, and 4 parts nano-calcium carbonate. The activity index of light-calcined magnesium oxide, measured by the citric acid method, is 168s. The shell slurry has a solid content of 46%, and the shell is formed by roller coating. After drying, the shell layer is approximately 16.5 μm thick.

[0070] To investigate the necessity of each component, the following comparisons were set up: When MgO was removed and only sulfoaluminate clinker and other components were retained, the post-peak compensation strain was 121 × 10^-6 for 1–3 days, but tended to stop after 7 days, with a total compensation of only 139 × 10^-6 after 28 days; when sulfoaluminate clinker was removed and only MgO was retained, the post-peak compensation was only 37 × 10^-6 for the first 48 hours, but the total compensation after 28 days could reach 162 × 10^-6; the two-stage shell of the present invention could reach 148 × 10^-6 for the first 48 hours after the peak, with a total compensation of 246 × 10^-6 after 28 days, and the total compensation remained at 271 × 10^-6 after 90 days, without excessive expansion. XRD analysis showed that the shell of the present invention had a significant AFt characteristic peak detected 24 hours after the peak, and the Mg(OH)2 characteristic peak gradually increased from 7 days; while the single MgO group had almost no AFt in the early stage, and the single sulfoaluminate group had no Mg(OH)2 in the later stage. This indicates that the reverse-flow reaction shell does indeed achieve a two-stage mechanism of "rapid micro-pre-compression followed by hysteresis compensation".

[0071] To verify the necessity of the MgO activity index window, three types of lightly calcined magnesia with activity indices of 92s, 168s, and 258s were selected. The 92s group showed excessively rapid expansion within 7 days, and exhibited a tendency for local overcompensation 3 days after the peak, with fine bulges visible on the surface of the constrained sample. The 168s group showed the most balanced performance; the 258s group still showed insufficient compensation up to 28 days. Therefore, the optimal MgO activity index was determined to be between 140s and 220s.

[0072] Cementitious materials are prepared by the following method:

[0073] First, premix and grind low-heat silicate cement, active metakaolin, sulfate conditioning component, crystal nucleation filler component and grinding aid to a specific surface area of ​​340-420 m². 2 / kg,

[0074] Then, the temperature peak memory reverse-flow activated self-limiting compensation functional particles and the premixed powder are mixed with low shear for 3-10 minutes using a plow-type mixer or a double ribbon mixer, and the temperature peak memory reverse-flow activated self-limiting compensation functional particles are not involved in high-energy ball milling.

[0075] The matrix premixing and grinding was carried out using a 5L planetary ball mill at 180 rpm for 22 minutes, resulting in a premixed powder with a specific surface area of ​​378 m². 2 / kg. If the same formula is only ground to 325mg... 2 / kg, then the 3-day mortar strength is only 24.1MPa, and the kaolin is unevenly dispersed; if ground to 436m 2 If the water content is increased by 1.3 hours, the peak heat release over 72 hours shifts forward from 25.9% to 27.4%. Therefore, the optimal specific surface area is determined to be between 340 and 420 m². 2 / kg.

[0076] The final mixing was performed using a twin-ribbon mixer at 46 rpm for 6 minutes. Before mixing, the functional particles were dried at 45°C to a moisture content of 0.4%. The particle integrity rate after final mixing was 95.6% as determined by sieving and microscopic statistical analysis. As a control, the same functional particles were ball-milled with the matrix powder for 20 minutes. The results showed that the particle integrity rate decreased to 61.8%, the apparent latent heat of phase change (DSC) decreased from 48.6 J / g to 19.7 J / g, and the water release rate after 72 hours of thermal reflow decreased from 64.7% to 22.5%. This indicates that the absence of functional particles in high-energy ball milling is a necessary condition for achieving the desired invention effect.

[0077] C35 grade mass concrete was prepared using this cementitious material, with a cementitious material dosage of 320 kg / m³, a water-cement ratio of 0.38, a sand ratio of 38%, a continuous gradation of crushed stone from 5 to 25 mm, and a polycarboxylate superplasticizer dosage of 0.16%. The process of "aggregate + 70% mixing water premixed for 20 seconds — adding matrix powder and mixing for 45 seconds — adding functional particles and remaining water and mixing for 30 seconds" resulted in an initial slump of 205 mm, which increased to 188 mm after 1 hour, meeting the engineering construction requirements.

[0078] Based on the total mass of cementitious materials, the active metakaolin is 7.5%–9.5%, the sulfate adjusting component is 2.0%–3.0%, the crystal nucleation filler component is 1.0%–2.0%, the grinding aid is 0.15%–0.25%, the temperature peak memory reverse process activated self-limiting compensation functional particles are 7.0%–9.0%, and the balance is low-heat silicate cement.

[0079] Within the aforementioned preferred window, three preferred embodiments were determined during the research and development phase. In Embodiment 1, the composition was 8.0% active metakaolin, 2.4% sulfate adjusting component, 1.4% nucleation filler component, 0.20% grinding aid, and 7.5% functional particles; in Embodiment 2, the composition was 8.5% active metakaolin, 2.6% sulfate adjusting component, 1.6% nucleation filler component, 0.20% grinding aid, and 8.0% functional particles; in Embodiment 3, the composition was 9.0% active metakaolin, 2.8% sulfate adjusting component, 1.8% nucleation filler component, 0.22% grinding aid, and 8.5% functional particles. The remainder was low-heat silicate cement.

[0080] The three sets of examples were used with the same concrete mix proportion, and the test results are as follows: Example 1 had a 72-hour equivalent adiabatic temperature rise peak of 40.2℃, reached the peak in 61 hours, had a 3-day compressive strength of 28.1 MPa, a 28-day compressive strength of 51.6 MPa, an autogenous shrinkage of 112 × 10⁻⁶ over 7 days, and a ring-shaped first crack time > 56 days; Example 2 had a 7-day adiabatic temperature rise peak of 38.9℃, reached the peak in 66 hours, had a 3-day compressive strength of 28.7 MPa, a 28-day compressive strength of 53.2 MPa, an autogenous shrinkage of 96 × 10⁻⁶ over 7 days, and a ring-shaped first crack time > 63 days; Example 3 had a 7-day adiabatic temperature rise peak of 38.5℃, reached the peak in 69 hours, had a 3-day compressive strength of 29.0 MPa, a 28-day compressive strength of 53.8 MPa, an autogenous shrinkage of 91 × 10⁻⁶ over 7 days, but a slightly larger slump loss. Considering overall strength, temperature control, fluidity retention, and post-peak compensation balance, Example 2 is the best.

[0081] Compared to the comparative examples, the 7-day heat of hydration of Example 2 was 201 kJ / kg, while that of Comparative Example A was 239 kJ / kg; the 90-day drying shrinkage of Example 2 was 276 × 10⁻⁶, while that of Comparative Example A was 421 × 10⁻⁶, and that of Comparative Example C was 348 × 10⁻⁶; the cumulative compensated strain after the peak of the programmed temperature control in Example 2 reached 218 × 10⁻⁶, while that of Comparative Example A was only 14 × 10⁻⁶, and that of Comparative Example C was 126 × 10⁻⁶; the ring test in Example 2 did not crack after 63 days, while Comparative Example A cracked after 18 days, and Comparative Example C cracked after 41 days. These data indicate that the above-mentioned optimized ratio range not only has the advantage of low heat generation but also significantly alters the post-peak stress evolution path.

[0082] The process of the cementitious material of the present invention in the large-volume concrete of pumped storage power stations can be divided into four continuous processes: the temperature rise identification stage, the thermal return line triggering stage, the post-peak rapid compensation stage, and the mid-to-late stage lag compensation stage.

[0083] During the heating identification stage, low-heat silicate cement initially constitutes the low-heat exothermic main body. Preferably, the mineral composition of the low-heat silicate cement clinker used satisfies the following: C2S 38%–52%, C3S 18%–32%, C3A 1.5%–5.0%, and C4AF 8%–18%. Under this mineral composition, the main exothermic source of the cementitious system in the first 72 hours is mainly composed of C3S hydration and a small amount of early aluminum phase reaction, while the contribution of C2S is more reflected in the mid-to-late stage strength formation. After the reactive metakaolin enters the system, it relies on its highly reactive lamellar surface to adsorb Ca... 2+ With OH - This process slows down localized supersaturation crystallization and, on the other hand, reacts with CH to form C-(A)-SH gel and part of the CAH phase, allowing a more continuous and finer microstructure to be established at the particle-matrix interface before the temperature peak arrives. This process means that the invention does not simply "reduce exothermics," but rather reduces exothermics without significantly sacrificing early structural integrity. During the research and development process, TG analysis of 3-day hydration samples showed that the CH peak area in the preferred system of this invention decreased by approximately 18.4% compared to the control group without metakaolin, while the weight loss of gel-bound water in the 50–200℃ range increased by approximately 11.7%, indicating that metakaolin did indeed participate in the early secondary reaction.

[0084] When the internal temperature of the concrete rises with hydration and approaches 34–38°C, the composite phase change core material in the thermal memory core begins to transform from a solid to a molten state, absorbing a large amount of latent heat. This endothermic process is not for permanent heat storage, but rather to reduce the core temperature rise rate and establish a thermal history of "peak experience" within the actual temperature rise window of large-volume concrete. Because the thermal memory core of this invention has a controlled thermal hysteresis window of 3–8°C, the particles do not immediately return to their original state when the temperature drops slightly; instead, they only begin to show significant crystallization and reversion when the temperature continues to drop to 29–33°C. Therefore, the thermal memory core inside the particles does not undergo a typical temperature-sensitive process, but rather a complete thermal loop process of "peak crossing followed by return." It is this thermal loop behavior that creates the necessary conditions for the subsequent activation of the liquid storage slow-release ring and the reverse reaction shell.

[0085] During the thermal loop triggering stage, the thermal memory core undergoes a localized volume shrinkage and interfacial adhesion change as it transitions from a molten state to a crystalline state. Because the thermal memory core is surrounded by a liquid-releasing ring, and this ring is further surrounded by a low-modulus decoupling membrane, this shrinkage does not directly cause the entire shell to rupture. Instead, it forms a preferential debinding region near the decoupling membrane. During development, a hot-stage microscope combined with in-situ digital image analysis was used to observe a transparent simulated matrix sample containing single particles. The results showed that the particles of this invention maintained their integrity when heated to 38°C, but when the temperature dropped to around 31°C, a small number of continuous micro-dark bands appeared along the particle radius. Subsequent SEM cross-sections confirmed that this region corresponds to a microchannel opened at the decoupling membrane-reverse reaction shell interface. In contrast, the control group without a decoupling membrane exhibited random shell cracks during the heating stage, indicating that the triggering mechanism of this invention has a clear reverse gating characteristic.

[0086] During the rapid post-peak compensation phase, the liquid in the reservoir release ring migrates directionally along the microchannels formed after the thermal recirculation towards the reverse reaction shell and the matrix surrounding the particles. This liquid release at this stage reduces the local self-drying rate of the slurry around the particles and immediately provides a reaction medium for the sulfoaluminate clinker and anhydrous gypsum in the reverse reaction shell, promoting the preferential formation of AFt and creating the first stage of local micro-pre-compression around the particles. This reaction can be simplified as follows: the active aluminum phase in the sulfoaluminate clinker generates needle-like or bundle-like AFt in the presence of sulfate and water, and the corresponding volume effect is initially manifested as microscale compressive strain around the particles. Temperature-controlled deformation tests show that the preferred system of this invention can form a local compensation strain of 86 × 10⁻⁶ to 121 × 10⁻⁶ within 24 hours after the thermal peak recedes, while the control group of free MgO + CSA parallel-doped systems only achieves 41 × 10⁻⁶ to 67 × 10⁻⁶, indicating that the shell localization reaction indeed enhances early post-peak compensation.

[0087] In the mid-to-late stage of hysteresis compensation, the lightly calcined magnesium oxide in the reverse reaction shell begins to slowly hydrate to form Mg(OH)2 under localized high humidity, forming a second stage of compensation that is sequentially connected to the formation of the AFt micro-precompression in the earlier stage. This stage is not a large-scale expansion, but rather a relatively slow, long-term, and stable compensation process. Since the water released in the early stage of the reservoir slow-release ring is not completely depleted at once, the hydration of the shell layer has a certain degree of continuity. At the same time, the active metakaolinite, lithium silicate, or silica sol in the shell layer continue to interact with Ca(OH)2 in the matrix and the local ionic environment, generating additional C-(A)-SH or silicate gel, which gradually backfills the previously opened microchannels and the capillaries of the shell layer itself. This makes the particles exhibit a self-limiting "reaction-backfilling-flow restriction" behavior in the later stages. SEM / EDS analysis of particle cross-sections at 3, 7, and 28 days after the thermal peak revealed numerous needle-like AFt clusters within the shell at 3 days; an increase in plate-like or flocculent Mg(OH)2 products at 7 days; and the filling of relatively dense gel regions around the particles at 28 days. Correspondingly, the compensating strain growth rate slowed significantly after 7 days, indicating that a self-limiting mechanism had indeed been established.

[0088] The low-heat matrix reduces the total heat release and establishes an early continuous framework; the thermal memory core is responsible for identifying temperature peaks and establishing thermal history during the heating phase; the liquid storage slow-release ring releases water directionally after the complete thermal loop; the low-modulus decoupling membrane ensures "no premature opening before the peak and controllable opening after the peak"; and the reverse-flow reaction shell is responsible for the two-stage compensation of rapid initial reaction followed by slow reaction, ultimately achieving self-limitation through the self-closure of reaction products. This mechanism chain is complete and has a clear temporal sequence, which is the fundamental difference between this invention and ordinary phase change materials, ordinary SAP materials, ordinary MgO expansion systems, and ordinary composite cementitious materials.

[0089] Low-heat silicate cement must be used as the main cementitious component because the subsequent thermal backlash triggering and post-peak compensation in this invention are based on a matrix premise of "sufficient but not out-of-control temperature rise, and continuous early skeleton but not too rapid hardening". Three matrix groups were compared during the research and development: Group 1 consisted of 86% low-heat silicate cement, Group 2 consisted of 74% low-heat silicate cement, and Group 3 consisted of 92% low-heat silicate cement, with the proportions of the remaining components adjusted proportionally according to the preferred system. The results showed that Group 1 had a 72-hour adiabatic temperature rise peak of 38.9℃ and a 3-day mortar strength of 28.7 MPa; Group 2 had a lower adiabatic temperature rise peak of 37.4℃, but a lower 3-day mortar strength of 22.8 MPa, and significantly worse strain transfer after the post-peak compensation in the programmed temperature control system; Group 3 had an increased 3-day mortar strength of 30.1 MPa, but a higher adiabatic temperature rise peak of 42.5℃. This indicates that if the proportion of low-heat cement in the main body is too low, the matrix will be too soft; if it is too high, the temperature peak will be too high, both of which are not conducive to the full functioning of this invention.

[0090] Reactive metakaolin cannot be arbitrarily replaced with ordinary fly ash or ordinary mineral powder. In the research and development, 8.5% reactive metakaolin was replaced with 8.5% Class II fly ash and 8.5% S95 mineral powder for comparison. The results showed that the 3-day compressive strength of the metakaolin group was 28.7 MPa, the average width of the interfacial transition zone around the particles was 12–18 μm, and the post-peak 72-hour compensation strain was 218 × 10⁻⁶. The 3-day compressive strength of the fly ash group was only 24.3 MPa, the interfacial zone width expanded to 21–29 μm, and the post-peak compensation strain was 151 × 10⁻⁶. The 3-day compressive strength of the mineral powder group was 25.1 MPa, and the post-peak compensation strain was 163 × 10⁻⁶. This indicates that metakaolin is significantly superior to conventional admixtures in early interfacial construction and secondary densification around the shell. Its preferred D90 is no greater than 15 μm, the kaolinite conversion activity should be high, the loss on ignition should be less than 2.0%, and the specific surface area is preferably 800–1500 m². 2 / kg. When the material is too coarse or lacks sufficient activity, the synergistic effect between the interface and the shell decreases significantly.

[0091] The reason why anhydrous gypsum and / or hemihydrate gypsum are preferred as sulfate modifiers is that these components not only regulate matrix setting but also participate in the first post-peak stage reaction of the reverse reaction shell. If all gypsum is replaced with dihydrate, the dissolution and release characteristics are slower and it is easily consumed by the earlier system, making it impossible to guarantee the amount of available sulfate ions after the peak. In the research and development, after completely replacing gypsum with dihydrate, the rapid compensation strain 48 h after the thermal peak decreased from 148×10^-6 to 87×10^-6. Conversely, if all hemihydrate gypsum is used and the total amount is too high, it will cause the early reaction to be too fast, the initial setting to be significantly earlier, and the post-peak compensation to be unstable.

[0092] The nucleation filler component, consisting of limestone powder and / or nano-calcium carbonate, aims to improve the early structural continuity of the matrix and optimize the nucleation environment at the particle periphery without significantly increasing exothermic activity. Limestone powder primarily provides gradation and general nucleation, while nano-calcium carbonate mainly provides high surface energy nucleation sites. If the nucleation filler component is completely removed, the 3-day compressive strength of the preferred system decreases from 28.7 MPa to 25.9 MPa, and the stress transfer efficiency of the matrix periphery during the post-peak compensation stage decreases. If the nano-calcium carbonate content is increased to above 3.0%, significant agglomeration and loss of fluidity occur. Therefore, this component is maintained at 0.5%–2.5%, more preferably 1.0%–2.0%.

[0093] The carrier for the thermal memory core must be processed fly ash cenospheres, rather than ordinary microspheres, glass microspheres, or conventional porous silica particles. During the research and development, four carriers were compared: modified cenospheres, unmodified cenospheres, glass microspheres, and porous diatomaceous earth particles. Modified cenospheres had a phase change core loading rate of 41.8%, an apparent latent heat (DSC) of 48.6 J / g, and a latent heat retention rate of 87.3% after mixing. Unmodified cenospheres had a loading rate of 31.2% and a latent heat retention rate of 74.5%. Glass microspheres had a loading rate of less than 15%, making it almost impossible to achieve the target thermal hysteresis behavior. Although diatomaceous earth particles had a high loading rate, their poor sphericity led to severe breakage after mixing. Therefore, cenospheres with a hollow spherical shell structure and activated modification must be used.

[0094] The reservoir-release ring must employ a composite system of alkali-resistant water-absorbing resin, aluminosilicate filler, and stabilizing components, and cannot directly use ordinary SAP. Ordinary SAP exhibits a sharp decrease in absorbance and uncontrollable release in high-alkali, high-calcium pore solutions. In control experiments, the ordinary SAP system showed an absorbance of only 2.1 g / g in simulated pore solutions, with large fluctuations in the release rate after 72 hours, resulting in poor repeatability. The reservoir-release ring of this invention, under the same conditions, showed an absorbance of 6.8 g / g, and the release rate remained stable between 61% and 68% after 72 hours of thermal regression, demonstrating significantly better repeatability.

[0095] Low-modulus decoupling membranes cannot be replaced by ordinary dense hard shells. While high-modulus inorganic shells can improve the initial integrity of particles, they can lead to ineffective opening after thermal reversion; completely film-free membranes will crack prematurely during the heating phase. Only low-modulus decoupling membranes can achieve both pre-peak protection and post-peak opening. The selection of materials must balance alkali resistance, flexibility, and adhesion to the inner and outer layers.

[0096] The reverse reaction shell cannot be simplified to the simultaneous addition of "free MgO + free CSA" to the system. When free powders are added in parallel, the reaction site is diffuse, making it difficult to form a local pre-compression field at the particle-matrix interface. However, the shell positioning design of this invention ensures that the compensating stress first acts on the areas around the particles most prone to stress concentration and water loss. Comparative experiments show that while the addition of free MgO and CSA can also reduce the risk of cracking, the time to initial cracking of the ring is only extended to 41 days, while this invention achieves 63 days without cracking.

[0097] Preparation of modified fly ash cenospheres. Fly ash cenospheres obtained from dry separation in power plants were selected and subjected to double-layer sieving at 50 μm and 150 μm, with the intermediate particle size retained for later use. The sieved cenospheres were added to a 5% hydrochloric acid solution with a liquid-to-solid ratio controlled at 6:1 and magnetically stirred at 60℃ for 35 min. After washing, the cenospheres were filtered and repeatedly washed with deionized water until the pH of the filtrate was 6.8–7.2. The cenospheres were then dried in an oven at 105℃ for 4 h, and then transferred to a muffle furnace for activation at 400℃ for 1.5 h. After natural cooling to room temperature, the cenospheres were sealed and stored. After activation, the cenospheres were required to have a void ratio of 60%–75% and a crushing strength of not less than 2 MPa.

[0098] Preparation of thermal memory cores. 100 parts of industrial-grade paraffin A (melting point 35℃), 8 parts of paraffin B (melting point 42℃), 1.5 parts of fatty acid amide crystallization regulator, and 1.2 parts of carboxylated multi-walled carbon nanotubes were added to a thermostatically controlled container with a stirrer. The mixture was stirred at 72℃ for 40 min, followed by 20 min of high-shear dispersion to form a homogeneous composite phase change liquid. Pretreated cenospheres were completely immersed in this phase change liquid. A vacuum was applied at -0.093 MPa for 30 min, followed by pressurization to 0.45 MPa and holding for 45 min, then depressurization. The cenospheres were removed and passed through a 100-mesh sieve to remove surface liquid. The mixture was then kept at 42℃ for 20 min to allow the free phase change liquid to reflux. The resulting thermal memory core requires a phase change core material loading rate of 35%–48%, a melting initiation temperature of 34–38℃, a crystallization initiation temperature of 29–33℃, a thermal hysteresis window of 3–8℃, and an apparent latent heat of 35–65 J / g as measured by DSC.

[0099] Preparation of the reservoir-release ring. A ring-layer slurry was prepared using 42 parts of alkali-resistant acrylate / acrylamide copolymer water-absorbing resin solids, 34 parts of activated metakaolin, 14 parts of nano-silica, 8 parts of calcium hydroxide, and 2 parts of lithium salt stabilizer. A mixture of deionized water and a small amount of ethanol was used as the dispersion medium, and the solid content was adjusted to 31%. The slurry viscosity was controlled at 350–500 mPa·s. The thermal memory core was placed in a fluidized bed coating device, and the slurry was continuously sprayed and cured with hot air for 2–3 h at 45–55 °C until the reservoir-release ring was formed. After formation, the particles were further cured at 50 °C for 2 h to improve the ring layer stability. The prepared reservoir-release ring should have an absorption rate of 4–10 g / g in simulated pore liquid after 24 h.

[0100] Preparation of low-modulus decoupling membranes. A membrane solution was prepared by mixing vinyltriethoxysilane, hydroxyl-terminated polydimethylsiloxane, and colloidal silica at a solid content mass ratio of 4:3:3, with a total solid content controlled at 3%–5%. Particles with a reservoir-release ring were dipped into the membrane for 10–15 min, then dried at 60°C for 2 h, and allowed to stand at room temperature for 12 h to complete polycondensation and curing. The thickness of the resulting decoupling membrane was controlled at 0.5–2.0 μm, preferably 0.8–1.5 μm. Membrane continuity was checked by SEM cross-section sampling, requiring a continuity rate of no less than 90% among 30 randomly selected particles.

[0101] Preparation of the reverse-flow reaction shell: 56 parts of lightly calcined magnesia, 18 parts of sulfoaluminate clinker, 10 parts of activated metakaolin, 7 parts of anhydrous gypsum, and 4 parts of nano-calcium carbonate were dry-mixed evenly. Lithium silicate solution was used as a liquid-phase binder to achieve a total solid content of 46%, and the slurry viscosity was controlled at 300–900 mPa·s. Particles with a decoupling membrane were fed into a roller-coating granulation device, and the shell slurry was sprayed in stages and simultaneously dried with hot air at 50–65℃ until a continuous shell was formed. After the shell was formed, it was further cured at 45℃ for 4 hours to obtain complete functional particles. The finished functional particles were required to have a particle size of 80–250 μm, a shell thickness preferably of 10–25 μm, and an MgO activity index of 140–220 s.

[0102] Preparation of the matrix powder. Low-heat silicate cement, active metakaolin, sulfate modifier, nucleation filler, and grinding aid were weighed according to the formula and pre-mixed and ground using a ball mill or vertical mill. After grinding, the specific surface area was controlled to be 340–420 m². 2 / kg. The powder should be uniformly dispersed, with no obvious agglomeration of kaolin or nanofiller.

[0103] Preparation of the final cementitious material: The matrix powder is placed in a twin-ribbon or plow-type mixer, and functional particles are added after low-speed operation. The speed is controlled at 30-80 rpm, and mixing is carried out for 4-8 minutes. Immediately after mixing, samples are taken to test the particle integrity rate, the apparent latent heat retention rate of functional particles, and the particle size distribution. The preferred control indicators are: particle integrity rate not less than 90%, latent heat retention rate not less than 80%, and the increase in fine powder not more than 8%. After passing the tests, the material is sealed and packaged.

[0104] For concrete mixing process requirements, it is recommended to adopt the following post-addition method: "pre-wetting aggregates and part of the mixing water—adding matrix powder—and then adding functional particles and water-reducing agent." Functional particles should be protected from prolonged impact from the pump blades during high-speed operation. In research and development, a 60 L forced mixer was used for verification. The best particle integrity was maintained under the following process: first adding aggregates and 70% water and mixing for 20 seconds, then adding matrix powder and mixing for 45 seconds, and finally adding functional particles, remaining water, and water-reducing agent and mixing for 30 seconds.

[0105] To ensure the stable implementation of this invention, the functional particles and the final cementitious material should be controlled according to the following indicators:

[0106] The modified cenospheres have a cavity ratio of 60%–75% and a crushing strength ≥2 MPa.

[0107] The thermal memory core loading rate is 35%–48%, the apparent latent heat of DSC is 35–65 J / g, and the thermal hysteresis window is 3–8℃.

[0108] The liquid absorption ratio of the reservoir slow-release ring in the simulated pore liquid is 4–10 g / g;

[0109] The water release rate is ≤20% in the first 24 hours after the heat return line, and 50%–85% in the 12–72 hours after the heat return line.

[0110] The decoupling membrane thickness is 0.5–2.0 μm, with a continuity rate ≥90%.

[0111] The thickness of the reverse reaction shell is 8–35 μm, and the shell components are uniformly distributed.

[0112] After final mixing, the particle integrity rate is ≥90%, and the latent heat retention rate is ≥80%.

[0113] The corresponding testing method can be:

[0114] DSC is used for testing thermal memory latent heat, melting / crystallization initiation temperature, and thermal hysteresis window.

[0115] LF-NMR, mass fractionation, and temperature-controlled programmed analysis were combined for testing water release behavior.

[0116] SEM / FIB-SEM is used for layer thickness and interface observation;

[0117] XRD and TG-DSC were used to verify the two-stage formation of AFt and Mg(OH)2 after the shell peak;

[0118] Temperature-controlled deformation tests are used to characterize post-peak compensation strain.

[0119] The circular ring constraint cracking test is used to evaluate crack resistance.

[0120] Insulated temperature rise tanks are used to evaluate temperature peak reduction capabilities.

[0121] Representative data obtained when using the optimized formula during research and development include:

[0122] The peak temperature rise in the adiabatic matrix over 72 hours was 38.9℃, compared to 47.6℃ in the control low-heat matrix group.

[0123] The cumulative compensated strain 72 h after the peak temperature control was 218×10^-6, compared with 126×10^-6 in the control group of free MgO+CSA.

[0124] 7-day autogenous shrinkage was 96×10^-6, compared to 228×10^-6 in the control low-heat matrix group;

[0125] The ring test showed no cracking after 63 days, while the control low-heat matrix group cracked after 18 days.

[0126] The 28-day compressive strength was 53.2 MPa, which showed no strength loss compared to the 50.4 MPa of the control group without functional particles; in fact, it was slightly higher.

[0127] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a reference structure" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0128] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations, comprising low-heat silicate cement, active metakaolin, sulfate adjusting components, crystal nucleation filling components, and grinding aids, characterized in that, Based on the total mass of the cementitious material, the cementitious material further includes 5.0% to 10.5% of temperature peak memory reverse-flow activated self-limiting compensation functional particles; the active metakaolin is 6.0% to 10.0%; the sulfate adjusting component is 1.5% to 3.5%; the crystal nucleus filling component is 0.5% to 2.5%; the grinding aid is 0.1% to 0.3%; and the balance is low-heat silicate cement. The temperature peak memory reverse activation self-limiting compensation functional particle includes, from the inside out, a thermal memory core, a liquid storage slow-release ring, a low modulus decoupling membrane, and a reverse reaction shell. The thermal memory core is a modified fly ash cenosphere loaded with phase change core material. The liquid storage slow-release ring includes alkali-resistant water-absorbing resin, silica-alumina filler and stabilizing components. The low-modulus decoupling membrane is coated on the outer surface of the liquid storage slow-release ring. The reverse reaction shell includes lightly calcined magnesium oxide, sulfoaluminate clinker, activated metakaolin, anhydrous gypsum, lithium silicate or silica sol and nano calcium carbonate.

2. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 1, characterized in that, The sulfate conditioning component is anhydrous gypsum and / or hemihydrate gypsum, and the crystal nucleus filling component is limestone powder and / or nano-calcium carbonate.

3. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 1, characterized in that, The particle size of the temperature peak memory reverse activation self-limiting compensation functional particles is 80-250 μm; based on the total mass of the functional particles, the thermal memory core accounts for 30%-45%, the liquid storage slow-release ring accounts for 8%-18%, the low modulus decoupling membrane accounts for 0.5%-3.0%, and the remainder is the reverse reaction shell.

4. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 3, characterized in that, The modified fly ash cenospheres have a particle size of 50–150 μm and a cavity rate of 60%–75%, and are obtained by washing fly ash cenospheres with dilute acid until neutral and then activating them at 350–450 °C. The thermal memory core has a melting initiation temperature of 34–38°C, a crystallization initiation temperature of 29–33°C, a thermal hysteresis window of 3–8°C, and an apparent latent heat of phase transition of 35–65 J / g.

5. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 3, characterized in that, The liquid storage slow-release ring comprises alkali-resistant water-absorbing resin, active metakaolin and / or nano-silica, and calcium hydroxide and / or lithium salt stabilizing components; in simulated cement pore liquid, the liquid absorption ratio of the liquid storage slow-release ring is 4 to 10 g / g.

6. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 5, characterized in that, Before the temperature peak memory reverse activation self-limiting compensation function particles experience a maximum temperature of not less than 34°C and subsequently drop back to a temperature of not more than 33°C, the cumulative water release rate in 24 hours is not higher than 20% of the total stored liquid volume; after experiencing the thermal return line, the cumulative water release rate in 12 to 72 hours is 50% to 85% of the total stored liquid volume.

7. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 3, characterized in that, The low-modulus decoupling membrane is an organosilicon-modified membrane or a siloxane-inorganic hybrid membrane with a thickness of 0.5–2.0 μm.

8. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 3, characterized in that, By mass, the reverse reaction shell comprises 45-65 parts of lightly calcined magnesium oxide, 10-25 parts of sulfoaluminate clinker, 5-15 parts of activated metakaolin, 4-10 parts of anhydrous gypsum, 3-8 parts of lithium silicate or silica sol, and 1-5 parts of nano-calcium carbonate. The activity index of the lightly calcined magnesium oxide is 140s-220s.

9. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 1, characterized in that, The cementitious material is prepared by the following method: First, premix and grind low-heat silicate cement, active metakaolin, sulfate conditioning component, crystal nucleation filler component and grinding aid to a specific surface area of ​​340-420 m². 2 / kg, The temperature peak memory reverse-flow activated self-limiting compensation functional particles are then mixed with the premixed powder using a plow-type mixer or a twin-ribbon mixer under low shear for 3-10 minutes, and the temperature peak memory reverse-flow activated self-limiting compensation functional particles are not involved in high-energy ball milling.

10. The low-heat, high-crack-resistance cementitious material for large-volume concrete in pumped storage power stations according to claim 1, characterized in that, Based on the total mass of cementitious materials, the active metakaolin is 7.5%–9.5%, the sulfate adjusting component is 2.0%–3.0%, the crystal nucleus filling component is 1.0%–2.0%, the grinding aid is 0.15%–0.25%, the temperature peak memory reverse process activated self-limiting compensation functional particles are 7.0%–9.0%, and the balance is low-heat silicate cement.

Images