Performance regulation of porous vegetation ecological concrete using carbonated recycled coarse aggregate and fly ash-silica fume blends

Pores, microcracks, and alkaline calcium-bearing phases in the adhered old mortar of recycled coarse aggregate (RCA) limit its application in porous vegetation ecological concrete (PVEC). To simultaneously satisfy PVEC requirements for load-bearing performance, connected porosity, and reduced alkalinity, this study combined gaseous accelerated carbonation of RCA with the addition of the fly ash-silica fume (FA-SF) blend. The effects of parent concrete strength, RCA carbonation duration, and the FA-SF replacement ratio on the compressive strength, connected porosity, and leachate pH of PVEC were investigated. The evolution of calcium-bearing phases and the aggregate-paste interfacial transition zone (ITZ) was analyzed using thermogravimetric and derivative thermogravimetric (TG/DTG) analysis, X-ray diffraction (XRD), and scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM-EDS). The results showed that, as the strength grade of the parent concrete increased from C30 to C50, the compressive strength of recycled-coarse-aggregate porous vegetation ecological concrete (R-PVEC) increased from 3.96 MPa to 5.15 MPa, whereas the leachate pH increased from 12.14 to 12.62. When RCA derived from C40 concrete was carbonated for 5 days at 20 °C, 70% relative humidity, and a CO 2 concentration of 20%, the compressive strength of carbonated recycled coarse aggregate porous vegetation ecological concrete (CR-PVEC) reached 5.63 MPa, equivalent to 98.4% of the compressive strength of CR-PVEC in the 7 days carbonation group. In addition, after accelerated carbonation for 5 days, the connected porosity of CR-PVEC remained at 27.99%, while its leachate pH decreased to 11.68. The addition of the FA-SF blend further improved the overall performance of CR-PVEC. At a FA-SF replacement ratio of 25%, CR-PVEC exhibited the highest compressive strength, at 6.56 MPa. When the replacement ratio was increased to 30%, the compressive strength, connected porosity, and leachate pH of the specimens were 6.32 MPa, 28.55%, and 10.51, respectively. Microstructural analyses showed that RCA carbonation and the addition of the FA-SF blend reduced the estimated Ca(OH) 2 content from 1.21% to 0.70%, increased the CaCO 3 content from 21.89% to 25.66%, and reduced the ITZ thickness from 36 μm to 15 μm. RCA carbonation primarily regulated the calcium-bearing phases and local defects in the adhered old mortar, while the addition of the FA-SF blend further optimized the new paste and interfacial structure. Together, these two measures enhanced the load-bearing capacity of the CR-PVEC skeleton and reduced the leachate pH.

Recycling waste concrete can reduce the consumption of natural aggregates and the stockpiling of solid waste and provides an important pathway for construction-material recycling (JTS 202–2011 Kabirifar et al., 2020; Wang et al., 2021). Waste concrete can be crushed and screened to produce recycled coarse aggregate (RCA). Its use in porous vegetation ecological concrete (PVEC) can provide water permeability, aeration, and drainage capacity through the connected pores formed by the packing of coarse aggregate, thereby supporting applications such as ecological slope protection, riverbank restoration, and permeable ecological structures (Tijani et al., 2022; Liu et al., 2023; Niyomukiza et al., 2023). Recent studies on no-fines concrete and pervious concrete have shown that permeability generally increases with connected porosity. However, permeability is also affected by pore-size distribution, pore connectivity, tortuosity, and local blockage by cement paste. Therefore, connected porosity alone cannot fully characterize the actual water-transport capacity of porous concrete (Li et al., 2021a; Li et al., 2021b). However, the engineering performance of PVEC involves a trade-off among skeletal load-bearing capacity, connected porosity, and ecological suitability. A higher connected porosity facilitates drainage, ventilation, and plant-root growth, but generally weakens the load-bearing capacity of the aggregate skeleton. When RCA replaces natural coarse aggregate, the adhered old mortar, internal pores, and microcracks introduced by crushing may further reduce skeleton stability and interfacial load-transfer efficiency. Therefore, improving the mechanical properties and reducing the leachate alkalinity of recycled coarse aggregate porous vegetation ecological concrete (R-PVEC) while maintaining open connected pores is a key challenge for the engineering application of this material (Chen et al., 2023; Mitrosz et al., 2023; Tamimi et al., 2023; Brasileiro et al., 2024).

The performance deterioration of RCA mainly originates from the adhered old mortar and the pores and microcracks introduced during crushing of the parent concrete. Compared with natural coarse aggregate (NCA), RCA generally has a lower apparent density and higher water absorption and porosity. Meanwhile, the presence of multiple interfacial transition zones (ITZs), including the NCA old mortar interface, adhered old mortar-new mortar interface, and NCA-new mortar interface, aggravates the structural heterogeneity within R-PVEC and weakens the continuity of load transfer (Forero et al., 2022; Soleimanifar et al., 2023). For R-PVEC, the adhered old mortar is not only a potentially weak mechanical region but may also be an important source of leachable alkaline components. Residual soluble components such as Ca(OH) 2 in the old mortar may be released into the liquid phase under the action of water, maintaining a high pH in the R-PVEC leachate and adversely affecting the ecological application of the material (Chen et al., 2013). Ryegrass and bermudagrass are herbaceous species commonly used for ecological slope protection. Previous studies have shown that perennial ryegrass can maintain a relatively high seed-germination rate at pH 5–10, although its root growth is more stable within the pH range of 5–8. Some bermudagrass germplasms can tolerate saline–alkaline conditions at pH 10.0, but their tolerance varies considerably among genotypes (Javaid et al., 2022; Tang et al., 2024). Therefore, plant adaptation to alkaline environments differs among species and cultivars and should be further evaluated in relation to the actual rhizosphere conditions and the selected target plants. Although increasing the paste content or degree of compaction can improve interaggregate bonding and load-bearing capacity, it may also cause local pore-channel contraction or blockage, thereby weakening the drainage and ventilation functions of R-PVEC. Accordingly, R-PVEC material design should not focus solely on increasing strength, but should also consider skeleton stability, preservation of connected pores, and regulation of leachate pH.

Accelerated carbonation is an effective method for improving RCA quality. CO 2 can diffuse into the particles through pores and microcracks in the adhered old mortar and react with Ca(OH) 2 and some reactive calcium-bearing hydration products to form CaCO 3. The resulting carbonate phases may be deposited in pores, microcracks, and local interfacial defects in the old mortar, thereby helping to improve its compactness. Recent studies have shown that carbonation treatment can reduce the water absorption of RCA and improve the resistance of the resulting concrete to chloride penetration. However, the durability improvement is affected by factors such as carbonation duration, carbonation temperature, aggregate replacement ratio, and external loading conditions (Russo and Lollini, 2022; Jiang et al., 2024; Wang et al., 2024). At the same time, some leachable alkaline calcium-bearing phases are consumed, reducing the reserve of alkaline substances carried by RCA when it is introduced into a new system (Pu et al., 2021). It should be noted that the carbonation effect of RCA is influenced by multiple factors, including CO 2 concentration, temperature, relative humidity, carbonation duration, aggregate particle size, initial moisture condition, and the characteristics of the adhered old mortar (Leemann et al., 2023; Qin et al., 2024). Among these factors, parent concrete strength affects both the particle integrity and compactness of the adhered mortar in RCA and may also affect the diffusion paths and reaction depth of CO 2 within the particles. Therefore, under clearly defined accelerated carbonation conditions, it is necessary to investigate the effects of parent concrete strength and RCA carbonation duration on the properties of R-PVEC.

Existing studies have mainly focused on the water absorption, apparent density, crushing index, and carbonation degree of carbonated recycled coarse aggregate (CRCA), as well as its performance enhancing effects in conventional recycled aggregate concrete. By comparison, systematic research remains limited on whether RCA pre-carbonation can enhance the load-bearing capacity of the skeleton while maintaining connected pores and reducing leachate pH in no-fines porous systems. In particular, the strength of the parent concrete determines the initial structural characteristics and potential reserve of alkaline phases in RCA, whereas carbonation duration affects the CO 2 diffusion depth and the degree of conversion of calcium-bearing phases. The evolution of the macroscopic properties of PVEC under the combined action of these two factors requires further clarification.

RCA pre-carbonation acts mainly on the adhered old mortar and can reduce the alkaline calcium-bearing phases carried by the RCA itself, but it is difficult for this treatment to directly inhibit the formation of alkaline hydration products in the new paste. Therefore, reliance on accelerated carbonation pretreatment of RCA alone cannot adequately regulate the leachate pH of PVEC. Fly ash and silica fume, as commonly used mineral admixtures, can replace part of the cement and reduce the potential for Ca(OH) 2 generation through clinker hydration; they can also improve the cement paste and aggregate-paste interfacial structure through microfilling and pozzolanic reactions (Nazeer et al., 2023; Bai et al., 2024; Cai et al., 2024; Yao et al., 2024). On this basis, RCA pre-carbonation and the addition of the fly ash-silica fume (FA-SF) blend can act on PVEC at two levels, namely old-mortar modification and new-paste regulation, and thus have the potential to increase strength, preserve pores, and reduce leachate pH. However, previous studies have generally investigated RCA carbonation and mineral admixtures as separate modification strategies, mainly in conventional dense concrete. Their combined effects in PVEC, particularly on the balance among load-bearing capacity, connected porosity, and leachate alkalinity, remain unclear.

In summary, this study conducted a staged optimization of R-PVEC using parent concrete strength, RCA carbonation duration, and the FA-SF replacement ratio as the principal variables. First, the compressive strength, connected porosity, and leachate pH of PVEC prepared with RCA derived from parent concretes of different strengths were compared to identify an RCA source that balanced initial performance with reduced-alkalinity requirements. Second, the effects of accelerated RCA carbonation duration on the properties of carbonated recycled coarse aggregate porous vegetation ecological concrete (CR-PVEC) were investigated to determine a reasonable RCA carbonation duration. Finally, the FA-SF replacement ratio was optimized, and thermogravimetric and derivative thermogravimetric (TG/DTG) analysis, X-ray diffraction (XRD), and scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM-EDS) were used to characterize the composition of calcium-bearing phases and the evolution of the aggregate paste ITZ. This study aimed to reveal the hierarchical regulation of the old mortar by accelerated RCA carbonation and of the new PVEC paste by the addition of the FA-SF blend, thereby providing a scientific basis for designing R-PVEC that simultaneously achieves load-bearing performance, connected porosity, and a relatively low leachate pH.

Raw materials and properties

The cementitious material used to prepare the parent concrete was composite Portland cement (PC 42.5R) manufactured by Chongqing Huaxin Yanjing Cement Co., Ltd. The coarse aggregate was continuously graded natural crushed stone with a particle size of 5–20 mm and an apparent density of 2,715 kg/m 3. The fine aggregate was natural river sand with a fineness modulus of 2.58 and was classified as medium sand; its particle size distribution met the relevant requirements of JTS 202–2011 after screening (JTS 202-2011, 2011). Ordinary tap water was used as the mixing water.

In addition, the NCA used to prepare PVEC was single-sized natural crushed stone with a particle size of 10–20 mm. Its apparent density was 2,710 kg/m 3 and its water absorption was 0.33%. Grade II fly ash was used, with the following physical properties and chemical composition: density of 2,342 kg/m 3, specific surface area of 406 m 2 /kg, water demand ratio of 84.0%, and major chemical components including SiO 2 (55.9%), Al 2 O 3 (17.3%), CaO (6.6%), and Fe 2 O 3 (5.9%). Silica fume consisted mainly of amorphous SiO 2, with an SiO 2 content of approximately 94.26%, a specific surface area of 22.04 m 2 /g, and a relative density of approximately 2.2.

Mix proportions of parent concrete

To obtain RCA of different initial qualities, parent concretes with strength grades of C30, C40, and C50 were designed. The mix proportions of each group were designed in accordance with JGJ 55–2011 (JGJ 55-2011, 2011), as detailed in Table 1. Except for the water cement ratio and cementitious material content, the same sources of natural coarse aggregate, river sand, and mixing water were used for all groups.

Three 100 mm × 100 mm × 100 mm cubic specimens were prepared for each of the C30, C40, and C50 parent concretes, for a total of nine specimens, and their 28 days compressive strengths were measured. The results showed that the average 28 days compressive strengths of the three concrete strength grades were 31.80, 40.99, and 52.18 MPa, respectively, meeting the preset strength grades for the C30, C40, and C50 parent concretes.

RCA preparation and carbonation treatment

After casting and compaction, the parent concrete specimens were covered with plastic film and left undisturbed for 24 h. After demolding, they were cured in a saturated Ca(OH) 2 solution to an age of 28 days. The 28 days curing age was selected based on the RCA preparation protocols reported in previous studies (Gholizadeh-Vayghan et al., 2020; Li et al., 2020; Li et al., 2024), ensuring a consistent maturity of the parent concrete before crushing. Upon completion of curing, the specimens were crushed using a jaw crusher, and RCA with a particle size of 10–20 mm was obtained by screening, during which visible impurities and free loose particles were removed. To reduce the effect of changes in environmental humidity on the initial moisture condition of RCA, the screened RCA was stored in sealed containers and conditioned to the specified state before carbonation treatment or PVEC preparation. Before the accelerated carbonation test, the RCA was dried at (105 ± 5) °C to constant mass to establish a consistent dry initial condition. In addition, before PVEC mixing, the RCA or CRCA was pre-soaked for 24 h. During mixing, the RCA or CRCA was first premixed with 20% of the total water for no less than 30 s, after which the cementitious materials were added and mixed for 60 s; the remaining water was then added, and mixing continued for no less than 120 s.

Under dry conditions, RCA was subjected to gaseous accelerated carbonation in a TH-W concrete carbonation test chamber. For each batch, a 500 g RCA sample was evenly spread in a single layer in a stainless-steel mesh basket with 2 mm openings to avoid a significant effect of particle accumulation on CO 2 mass transfer. The temperature, relative humidity, and CO 2 concentration in the carbonation chamber were controlled at 20 °C, 70%, and 20%, respectively. Timing began after the chamber parameters had stabilized. Carbonation durations of 0, 1, 2, 3, 4, 5, 6, and 7 days were used, where 0 days represented uncarbonated RCA, and three parallel samples were prepared for each carbonation duration. Sample mass was recorded at regular intervals during the test, and carbonation was terminated for the corresponding group when the mass tended to stabilize. At the end of treatment, the corresponding CRCA specimens were immediately removed and stored in sealed containers for subsequent PVEC preparation.

Experimental design and specimen preparation

To reveal the effects of multiple factors on PVEC properties, a staged controlled-variable experimental design was adopted following the sequence of RCA initial quality, RCA carbonation pretreatment, and PVEC binder composition. The condition selected at each preceding stage was fixed in the subsequent stage to avoid simultaneous changes in multiple variables and distinguish the effect of each factor. The experimental design comprised the following stages: Parent concrete strength group: the properties of PVEC prepared with RCA derived from C30, C40, and C50 parent concretes were compared;

Carbonation duration group: the RCA source was fixed as C40 parent concrete, and the effects of carbonation durations of 0–7 days were investigated;

Mineral admixture group: NCA, uncarbonated RCA, and CRCA were used as raw materials to investigate changes in PVEC performance at different FA-SF replacement ratios (0%, 20%, 25%, 30%, and 35%).

Based on the results of the first two stages, representative groups comprising uncarbonated RCA, 5 days CRCA, and 5 days CRCA with 30% FA-SF were selected for microscopic testing and comparison to explain the differences in their macroscopic properties and leachate pH.

Unless otherwise specified, the coarse aggregate content, mixing water content, and total cementitious material content of PVEC in each group were controlled at 1,517, 85, and 303 kg/m 3, respectively. The FA-SF blend replaced part of the cement at a predetermined mass ratio of 2:1. The FA-SF replacement ratio was defined as the percentage of the total mass of fly ash and silica fume in the total mass of cementitious materials. As the FA-SF replacement ratio increased, fly ash and silica fume replaced cement on an equal-mass basis at the predetermined ratio, while the total cementitious material content remained unchanged. Except for the coarse aggregate type and cementitious material composition, all other mix parameters were held constant, as detailed in Table 2.

PVEC was prepared using a two-stage feeding method. First, all coarse aggregate was premixed with 20% of the total mixing water for 30 s to wet the aggregate surface uniformly. Cement, fly ash, and silica fume were then added and mixed for 60 s. Finally, the remaining mixing water was added, and mixing continued for 120 s. The mixture was placed into the mold in two layers and manually rodded in accordance with JC/T 2557–2020 (JC/T 2557-2020, 2020). For all PVEC mixtures, 100 mm × 100 mm × 100 mm cubic specimens were used to determine the 28 days compressive strength, connected porosity, and leachate pH. Each reported macroscopic measurement was based on three parallel specimens for each experimental group. After molding, the specimens were covered with plastic film and left undisturbed for 24 h. After demolding, they were cured for 28 days under standard conditions at (20 ± 2) °C and a relative humidity of at least 95%. Both RCA and CRCA were derived from C40 parent concrete, and the CRCA carbonation duration was fixed at 5 days. The experimental groups and variable settings are presented in Table 3.

Performance testing

The macroscopic properties evaluated were 28 days compressive strength, connected porosity, and leachate pH. Compressive strength was determined in accordance with GB/T 50081–2019 (GB/T50081-2019, 2019), connected porosity in accordance with Appendix A of CJJ/T 253–2016 (CJJ/T 253-2016, 2016), and leachate pH with reference to the procedure in Appendix A of DB21/T 4164–2025 (DB21/T 4164-2025, 2025). Specifically, the specimens were immersed in distilled water at a solid to liquid mass ratio of 1:3 for 24 h, after which the pH of the supernatant was measured using a pH meter calibrated with standard buffer solutions. The leachate pH was measured using a PH-8008 digital pH meter (XIMA, China). Before testing, the instrument was calibrated at 25 °C using standard buffer solutions of pH 6.86 and 9.18. The electrode was rinsed with distilled water before and after each measurement, and the pH value was recorded after the reading had stabilized. It should be noted that the 24 h static leaching test was primarily used to compare the leachate pH of different groups under a specified solid-to-liquid mass ratio and immersion duration. Therefore, the results reflect the short-term relative alkalinity reduction of the materials and may not fully represent their dynamic leaching behavior under continuous flow, intermittent rainfall, or repeated wetting–drying conditions.

To analyze the effects of CRCA and mineral admixtures on the composition of calcium-bearing phases in PVEC, R-PVEC-0, CR-PVEC-0, and CR-PVEC-30 were selected as three representative specimen groups for TG/DTG, XRD, and SEM-EDS analyses. R-PVEC-0 was PVEC containing uncarbonated RCA without mineral admixtures; CR-PVEC-0 was PVEC containing RCA carbonated for 5 days without mineral admixtures; and CR-PVEC-30 was PVEC containing RCA carbonated for 5 days and 30% FA-SF.

Before TG/DTG testing, the samples were immersed in absolute ethanol for 24 h to terminate hydration and then dried at 60 °C for 72 h. Paste samples were drilled from the specimen surface to a depth of approximately 3 mm, ground, and passed through a 45 μm sieve. A STA449F5 simultaneous thermal analyzer was used to heat the samples from 25 °C to 1,100 °C at a heating rate of 10 °C/min under a flowing nitrogen atmosphere. Based on the mass losses in the temperature ranges of 390–480 °C and 480–800 °C, the Ca(OH) 2 and CaCO 3 contents were semi-quantitatively estimated (Shen et al., 2016). Semi-quantitative estimates were made from the mass loss in the characteristic temperature ranges using Equations 1, 2: where w CH and w CC are the estimated mass fractions of Ca(OH) 2 and CaCO 3, respectively; Δ m 390-480 and Δ m 480-800 are the sample mass losses in the corresponding temperature ranges; m 0 is the initial sample mass for testing; and M CH, M H2O, M CC and M CO2 are the molar masses of Ca(OH) 2, H 2 O, CaCO 3, and CO 2, respectively, with values of 74.09, 18.02, 100.09, and 44.01 g/mol.

Representative paste samples for XRD testing were collected from the central interior region of the specimens. After hydration was terminated with absolute ethanol and the samples were dried at 60 °C for 24 h, they were ground into powder and passed through a 75 μm sieve to reduce the influence of particle orientation on the diffraction results. Phase testing was performed using an X-ray diffractometer, and MDI Jade 6 software was used to identify crystalline phases such as Ca(OH) 2, calcite (CaCO 3), and quartz (SiO 2). During testing, Cu Kα radiation was used with a 2θ scanning range of 10°–70°, a step size of 0.02°, and a scanning rate of 5°/min.

SEM-EDS testing was performed using a ZEISS Sigma 300 field emission scanning electron microscope. The aggregate-new paste interfacial region was selected for morphological observation and Ca and Si elemental line scanning. The samples were dried under vacuum for no less than 3 days and sputter-coated with gold before testing; the SEM accelerating voltage was set to 15 kV. Based on the variation in the Ca/Si atomic ratio with distance along each scanning path, the position at which the Ca/Si atomic ratio transitioned from a region of pronounced fluctuation to a relatively stable paste region was taken as the outer boundary of the ITZ, and the perpendicular distance from the aggregate surface to this position was considered the ITZ thickness (Pang et al., 2016; Bao et al., 2022). Accordingly, the reported ITZ thickness represents a path-specific, semi-quantitative estimate of the chemical transition zone and should not be interpreted as a nanoindentation-defined mechanical boundary.

Effect of parent concrete strength on the initial properties of R-PVEC

To clarify the effect of the initial quality of RCA on PVEC properties, the compressive strength, connected porosity, and leachate pH of PVEC prepared with RCA derived from C30, C40, and C50 parent concretes were compared. The results in this section were used to identify the gain in load-bearing performance associated with higher parent concrete strength and the accompanying potential increase in leachate pH, and to determine the RCA source for the subsequent experiments on RCA carbonation duration.

Compressive strength

As shown in Figure 1, the 28 days compressive strengths of C30-R-PVEC, C40-R-PVEC, and C50-R-PVEC were 3.96, 4.51, and 5.15 MPa, respectively. Compared with C30-R-PVEC, the compressive strengths of C40-R-PVEC and C50-R-PVEC increased by 13.9% and 30.0%, respectively. As the parent concrete strength grade increased from C30 to C50, the compressive strength of R-PVEC increased continuously, indicating that the source of RCA had a marked effect on the load-bearing capacity of the no-fines porous skeleton.

This trend may be related to the particle integrity and adhered old mortar characteristics of RCA prepared from different parent concretes. Compared with low-strength parent concrete, higher-strength parent concrete generally has a lower w/c and a denser hardened paste structure. After crushing, its RCA may therefore contain relatively fewer defects such as loosely adhered mortar, pores, and microcracks. Consequently, the RCA particles may form a more stable packing skeleton and improve the interfacial load-transfer conditions between the RCA and the new paste, which may contribute to the increased compressive strength of PVEC. Therefore, increasing the strength of the parent concrete can improve the load-bearing capacity of the R-PVEC skeleton, but this advantage must still be comprehensively evaluated in conjunction with the effects of parent concrete strength on connected pores and leachate pH.

Connected porosity

As shown in Figure 2, the connected porosities of C30-R-PVEC, C40-R-PVEC, and C50-R-PVEC were 27.14%, 27.33%, and 28.36%, respectively. As the parent concrete strength grade increased, the connected porosity increased only slightly, and all three groups remained above 27%, indicating that, under fixed aggregate particle size, cementitious material content, and molding method, parent concrete strength had a limited effect on the principal open connected pore structure of PVEC.

It is noteworthy that compressive strength and connected porosity did not exhibit the typical negative correlation in this experiment. This may be because the RCA particle size, cementitious material content, and molding method were held constant among the groups, and the increase in PVEC strength mainly resulted from improvements in RCA particle quality and aggregate-paste interfacial load transfer rather than from a reduction in porosity through structural densification. RCA prepared from higher-strength parent concrete may have greater particle integrity and fewer crushing defects, enabling it to improve skeleton stability while maintaining the principal connected pore channels formed by particle packing. Changes in parent concrete strength did not significantly alter the principal open connected pore structure of PVEC. Thus, leachate pH became a key criterion for selecting the RCA source.

Leachate pH

As shown in Figure 3, the leachate pH of R-PVEC increased with increasing parent concrete strength grade. The leachate pH values of C30-R-PVEC, C40-R-PVEC, and C50-R-PVEC were 12.14, 12.49, and 12.62, respectively; relative to C30-R-PVEC, the pH differences of the latter two groups were 0.35 and 0.48, respectively.

The increase in leachate pH may be related to the increased cement content of the parent concrete and the greater reserve of alkaline components in the adhered old mortar of RCA. The water-cement ratios of the C30, C40, and C50 parent concretes were 0.60, 0.50, and 0.40, respectively, corresponding to an increase in cement content from 325.0 kg/m 3 to 487.5 kg/m 3. The higher cement content may have resulted in more Ca(OH) 2 and other soluble alkaline components in the adhered old mortar. Under standard leaching conditions, these components may enter the liquid phase, thereby increasing the leachate pH.

Determination of the RCA source

Based on the combined results of Sections 3.1.1–3.1.3, as the parent concrete strength increased from C30 to C50, the compressive strength of R-PVEC increased continuously, but the leachate pH also increased accordingly. Meanwhile, the connected porosity of all three specimen groups remained stable within the range of 27%–28%, with relatively limited differences. C30-R-PVEC had the lowest leachate pH, but its compressive strength was only 3.96 MPa C50-R-PVEC had the highest compressive strength, but its leachate pH reached 12.62. By comparison, C40-R-PVEC had a compressive strength of 4.51 MPa, 13.9% higher than that of C30-R-PVEC, and its leachate pH was lower than that of C50-R-PVEC. Therefore, RCA prepared from C40 parent concrete was selected as the aggregate source for the subsequent optimization experiments on carbonation duration and mineral admixtures.

Effect of RCA carbonation duration on PVEC performance and determination of a reasonable duration

Based on the comprehensive comparison in Section 3.1, RCA derived from C40 parent concrete was selected for gaseous accelerated carbonation. To determine a carbonation duration that balanced treatment efficiency and the overall performance of PVEC, the changes in compressive strength, connected porosity, and leachate pH of PVEC prepared with CRCA obtained at carbonation durations of 0–7 days were compared.

Compressive strength

As the RCA carbonation duration was extended from 0 days to 7 days, the compressive strength of PVEC increased from 4.42 MPa to 5.72 MPa, representing a cumulative increase of 29.47% (Figure 4). The increase in the compressive strength of CR-PVEC exhibited distinct stages: the increase was small at 0–2 days, accelerated at 3–5 days, and slowed markedly after 5 days. This result suggests that, under the carbonation conditions used in this experiment, the carbonation modification of RCA may have gradually transitioned from rapid development to stabilization.

At the initial stage of carbonation, CO 2 preferentially reacted with readily reactive calcium-bearing phases such as Ca(OH) 2 in the surface and shallow layers of the adhered old mortar of RCA, but the reaction range was limited; therefore, the improvement in overall load-bearing performance was relatively small. When the carbonation duration was extended to 3–5 days, CO 2 could diffuse further along the pores and microcracks in the old mortar, promoting the conversion of more calcium-bearing phases and improving the old mortar and local interfacial defects through CaCO 3 deposition, thereby enhancing RCA skeleton stability and load-transfer continuity (Pu et al., 2021; Qin et al., 2024). When the carbonation duration exceeded 5 days, the reactive calcium-bearing phases in RCA gradually decreased, and deposition of carbonate products may have increased resistance to further CO 2 diffusion; therefore, strength growth tended to level off. After 5 days of carbonation, the compressive strength of CR-PVEC reached 5.63 MPa, equivalent to 98.4% of that of the 7 days group. Compared with the uncarbonated group, its strength increase accounted for 93.1% of the total increase from 0 to 7 days. Thereafter, extending the duration from 5 days to 7 days increased the compressive strength by only 0.09 MPa. Considering the time cost of treatment, 5 days can be regarded as a reasonable carbonation duration that balances treatment efficiency and strength gains under the experimental conditions used.

The compressive strength results show that the additional strength gain obtained by extending treatment beyond 5 days of carbonation was limited. A suitable carbonation duration must also be evaluated in conjunction with connected porosity and leachate pH.

Connected porosity

Corresponding to the increasing trend in compressive strength, the connected porosity of CR-PVEC gradually decreased as the RCA carbonation duration increased, from 29.16% in the uncarbonated group to 27.86% in the 7 days carbonation group (Figure 5). The change in connected porosity was relatively pronounced during the 3–5 days stage.

This phenomenon may be related to the deposition of the carbonation product CaCO 3 in pores, microcracks, and local interfacial defects in the adhered old mortar of RCA. The carbonation reaction may cause some small connected pore channels to contract or become locally disconnected, thereby reducing connected porosity. However, the main channels formed by coarse aggregate packing are relatively large and were not significantly closed. Therefore, the connected-porosity results suggest that carbonation treatment largely preserved the principal open pore structure of CR-PVEC, while the improved compactness of the aggregate and interfaces provides a possible explanation for the observed changes.

After 5 days of carbonation, the connected porosity of PVEC was 27.99%; continued carbonation to 7 days reduced it by only a further 0.13 percentage points. Together with the compressive-strength results, these findings indicate that 5 days of carbonation can significantly improve load-bearing performance while maintaining connected porosity at approximately 28%; it can therefore be used as the carbonation duration in the subsequent mineral-admixture optimization experiments. Although 5 days of carbonation reduced connected porosity relative to the uncarbonated group, a connected porosity of approximately 28% was maintained, providing an acceptable pore basis for subsequent mix-proportion optimization.

Leachate pH

RCA carbonation treatment reduced the leachate pH of CR-PVEC (Figure 6). As the carbonation duration increased from 0 days to 7 days, the leachate pH decreased from 12.56 to 11.63. After only 1 day of carbonation, the pH had already decreased to 11.90, indicating that the leachate alkalinity of the RCA system could be significantly reduced at the initial stage of carbonation.

During the initial carbonation stage (0–1 day), the leachate pH decreased by 0.66, indicating that readily reactive calcium-bearing phases in the surface and shallow layers of the adhered old mortar of RCA may have carbonated preferentially. As the carbonation duration increased, the rate of pH decrease gradually slowed: the pH decreased by only a further 0.22 during 1–5 days and by only 0.05 during 5–7 days. This may be because the readily reactive calcium-bearing phases gradually decreased, while deposition of carbonation products created some resistance to further CO 2 diffusion. The tendency of the pH to stabilize after 5 days was consistent with the slowing of compressive strength growth and the reduced change in connected porosity.

Comprehensive determination of carbonation duration

The combined changes in compressive strength, connected porosity, and leachate pH show that RCA carbonation treatment had a relatively pronounced regulatory effect on CR-PVEC performance within 0–5 days. After 5 days of carbonation, the compressive strength of PVEC was 5.63 MPa, reaching 98.4% of that of the 7 days carbonation group; the connected porosity was 27.99%, indicating that the principal open connected pore characteristics were retained; and the leachate pH decreased to 11.68. After continued carbonation to 7 days, the compressive strength increased by only 0.09 MPa, the connected porosity decreased by only 0.13 percentage points, and the leachate pH decreased by only a further 0.05. Therefore, under the carbonation conditions used in this study (a temperature of 20 °C, a relative humidity of 70%, and a CO 2 volume fraction of 20%), 5 dwas determined as the RCA carbonation duration for the subsequent mineral-admixture mix-proportion optimization experiments. Accordingly, CRCA derived from C40 concrete and carbonated for 5 days was used to prepare CR-PVEC in the subsequent mineral-admixture optimization experiments. The CO 2 concentration of 20% and carbonation duration of 5 days should be regarded as laboratory-scale screening conditions rather than industrial operating parameters. At a larger scale, carbonation efficiency would depend on reactor sealing and gas recirculation, aggregate-bed thickness, and moisture control. Therefore, pilot-scale studies quantifying CO 2 uptake, energy demand, and processing capacity are required before practical implementation.

Effect of FA-SF on PVEC performance and mix-proportion optimization

After selecting RCA derived from C40 concrete and a carbonation duration of 5 days, the effect of the FA-SF replacement ratio on PVEC performance in different aggregate systems was further investigated. Three PVEC systems using NCA, RCA, and CRCA, namely PVEC, R-PVEC, and CR-PVEC, were establishedto compare the regulatory effects of mineral admixtures on the compressive strength, connected porosity, and leachate pH of PVEC under different aggregate conditions and thereby determine the recommended FA-SF replacement ratio for CR-PVEC.

Compressive strength

As shown in Figure 7a, with increasing FA-SF replacement ratio, the compressive strengths of PVEC, R-PVEC, and CR-PVEC all first increased and then decreased. The maximum compressive strengths of PVEC, R-PVEC, and CR-PVEC occurred at replacement ratios of 30%, 35%, and 25%, respectively, with corresponding values of 9.37, 5.36, and 6.56 MPa. The optimal replacement ratio varied among aggregate systems, indicating that the optimum depended on the cementitious material composition and aggregate water absorption characteristics, particle surface condition, and aggregate paste interfacial conditions. Therefore, the mineral-admixture optimization results for the NCA system should not be applied directly to R-PVEC or CR-PVEC.

At low and moderate replacement ratios, the microfilling effect of silica fume may improve particle packing in the paste and interfacial regions, while the pozzolanic reactions of fly ash and silica fume may consume part of the Ca(OH) 2 and promote the formation of secondary calcium silicate hydrate (C-S-H) gel, thereby improving interaggregate bonding and interfacial load-transfer conditions (Nazeer et al., 2023). As the replacement ratio continued to increase, the dilution effect caused by the reduced cement content gradually intensified. At the same time, the reaction rate of fly ash was relatively low and may not fully compensate, at an age of 28 days, for the decrease in hydration products caused by the reduced clinker content. Consequently, the compressive strength of PVEC decreased after reaching its peak.

At the same replacement ratio, the compressive strength of R-PVEC was 32.02%–37.17% lower than that of the corresponding PVEC, reflecting the adverse effects of defects in the adhered old mortar and multiple interfaces in uncarbonated RCA on skeleton load-bearing performance. After carbonation treatment, the compressive strength of CR-PVEC was 23.40%–31.38% higher than that of R-PVEC and was consistently higher at all replacement ratios, indicating that RCA carbonation treatment enhanced the load-bearing performance of the RCA system (Figure 7b).

For CR-PVEC, the highest compressive strength, 6.56 MPa, was obtained at a replacement ratio of 25%. When the replacement ratio was increased to 30%, the compressive strength was 6.32 MPa, only 3.7% lower than the peak value, while the leachate pH was further reduced. Therefore, if maximizing compressive strength is the objective, a replacement ratio of 25% is more suitable; if compressive strength, connected porosity, and leachate pH are considered comprehensively, a replacement ratio of 30% provides a better overall balance.

Connected porosity

As the FA-SF replacement ratio increased from 0% to 35%, the connected porosities of PVEC, R-PVEC, and CR-PVEC all decreased slowly (Figure 8a). Specifically, the connected porosities of PVEC, R-PVEC, and CR-PVEC decreased from 27.84%, 29.71%, and 29.07%–27.00%, 28.40%, and 28.37%, respectively, corresponding to absolute decreases of 0.84, 1.31, and 0.70 percentage points. Overall, increasing the FA-SF replacement ratio had only a small effect on the connected porosity of PVEC, and all groups retained open connected porosities of approximately 27%–30%.

The slight decrease in connected porosity may be related to filling of the paste and interfacial regions and deposition of reaction products resulting from the addition of the FA-SF blend. Fine silica-fume particles may fill local micropores, while the pozzolanic reaction products of fly ash and silica fume may further reduce the connectivity of some small pore channels. It should be noted that connected porosity was measured in this study; therefore, these results mainly reflect changes in the volume of connected pore channels in PVEC and cannot simply be equated with a proportional reduction in total pore volume. Moreover, connected porosity mainly characterizes the volume of interconnected pores in PVEC and does not directly represent hydraulic conductivity or permeability. Actual water flow is also affected by pore-size distribution, pore tortuosity, channel dimensions, and local blockage; therefore, connected porosity does not necessarily exhibit a simple one-to-one relationship with hydraulic performance.

The difference in connected porosity between CR-PVEC and R-PVEC decreased as the FA-SF replacement ratio increased: the absolute difference decreased from 0.64 percentage points in the group without mineral admixtures to 0.03 percentage points in the group with a FA-SF replacement ratio of 35% (Figure 8b). This result indicates that, as the FA-SF replacement ratio increased, the influence of the new paste and its interfacial structure on connected pore characteristics gradually strengthened, partially attenuating the initial difference in connected porosity between RCA before and after carbonation. Nevertheless, the connected porosity of all groups remained at approximately 27%–30%, indicating that mineral-admixture addition did not alter the overall open pore characteristics of PVEC.

Leachate pH

The effect of mineral admixtures on leachate pH differed from their effect on compressive strength. Within the FA-SF replacement range of 0%–35%, the leachate pH of all three types of PVEC decreased continuously as the replacement ratio increased (Figure 9a). Specifically, the pH of R-PVEC decreased from 12.56 to 11.63, that of CR-PVEC from 11.62 to 10.32, and that of PVEC from 11.47 to 10.21. These results indicate that the replacement ratio producing the maximum compressive strength does not necessarily correspond to the lowest leachate pH; mix design must balance material load-bearing performance, connected pores, and reduced-alkalinity requirements.

The decrease in leachate pH may have been caused jointly by the following factors. First, the replacement of part of the cement by fly ash and silica fume reduced the potential production of cement-clinker hydration products in the system. Second, the pozzolanic reactions of fly ash and silica fume may have consumed part of the Ca(OH) 2. Third, the filling effect of the mineral admixtures and the formation of secondary reaction products may have modified the local pore and interfacial structure of PVEC, thereby affecting the migration of alkaline components into the leachate. However, this possible mechanism requires further verification through direct ion-transport measurements. It should be emphasized that ion leaching was not measured directly in this study; therefore, “restricted migration” is only a possible explanation for the pH changes and cannot be regarded as a directly verified mechanism. At all replacement ratios, the leachate pH of R-PVEC was higher than those of PVEC and CR-PVEC, indicating that the adhered old mortar of uncarbonated RCA may remain an important source of alkaline components in the system.

Compared with the corresponding R-PVEC, CR-PVEC exhibited a lower leachate pH at all mineral-admixture replacement ratios, indicating that accelerated RCA carbonation could further reduce leachate alkalinity in addition to the alkalinity-reducing effect of mineral admixtures (Figure 9b). For CR-PVEC-30, the compressive strength, connected porosity, and leachate pH were 6.32 MPa, 28.55%, and 10.51, respectively, representing favorable overall performance in terms of the indicators tested in this study. Nevertheless, a leachate pH of 10.51 remains alkaline and should not be interpreted as being universally suitable for vegetation establishment. Its ecological suitability depends on plant-specific alkali tolerance and site conditions; therefore, plant growth tests and, where necessary, appropriate buffering measures are required before field application.

Comprehensive performance evaluation and recommended mix proportion

Table 4 summarizes the comprehensive performance of CR-PVEC at key FA-SF replacement ratios. For CR-PVEC, the highest compressive strength, 6.56 MPa, was obtained at a FA-SF replacement ratio of 25%. When the FA-SF replacement ratio was increased to 30%, the compressive strength was 6.32 MPa, only 3.7% lower than the peak value, while the connected porosity was 28.55% and the leachate pH decreased to 10.51. Increasing the FA-SF replacement ratio beyond 30% could further reduce the leachate pH but would cause a further decrease in compressive strength. Therefore, if maximizing compressive strength is the objective, a replacement ratio of 25% is recommended. If the objective is an overall balance among load-bearing performance, preservation of connected pores, and reduced leachate pH, a FA-SF replacement ratio of 30% is recommended. To explain the effects of CRCA and the addition of the FA-SF blend on the macroscopic properties and leachate pH of PVEC, R-PVEC-0, CR-PVEC-0, and CR-PVEC-30 were selected for microscopic characterization.

Microstructural evolution and mechanism of leachate pH reduction

Based on the macroscopic performance results presented in Sections 3.2, 3.3, R-PVEC-0, CR-PVEC-0, and CR-PVEC-30 were selected as three representative specimen groups for microstructural characterization. These groups corresponded to the uncarbonated RCA system, the CRCA system, and the CRCA system with a 30% FA-SF replacement ratio, respectively, and could be used to distinguish the modification of the adhered old mortar and old ITZ by RCA carbonation from the further regulation of the new paste and interfacial region by the mineral admixtures.

Evolution of calcium-bearing phases: TG/DTG and XRD

As shown in Figure 10, the samples exhibited mass-loss characteristics associated with Ca(OH) 2 dehydroxylation and CaCO 3 decarbonation in the temperature ranges of 390–480 °C and 480–800 °C, respectively. Compared with R-PVEC-0, the DTG peak corresponding to Ca(OH) 2 in CR-PVEC-0 weakened, while the mass-loss peak in the CaCO 3 decomposition range intensified. At a 30% FA-SF replacement ratio, the Ca(OH) 2 related peak in CR-PVEC-30 weakened further. These results suggest that both accelerated RCA carbonation and the addition of the FA-SF blend may reduce the relative Ca(OH) 2 content in the system. The former may be mainly associated with the carbonation conversion of calcium-bearing phases in the adhered old mortar of RCA, whereas the latter may be jointly associated with the cement replacement effect and pozzolanic reactions.

Table 5 presents a semi-quantitative analysis of the TG/DTG results. The estimated Ca(OH) 2 contents in R-PVEC-0, CR-PVEC-0, and CR-PVEC-30 were 1.21%, 0.83%, and 0.70%, respectively, and the corresponding CaCO 3 contents were 21.89%, 24.33%, and 25.66%. Compared with R-PVEC-0, the Ca(OH) 2 content in CR-PVEC-0 decreased by 0.38 percentage points, and the CaCO 3 content increased by 2.44 percentage points. At a 30% FA-SF replacement ratio, the Ca(OH) 2 content decreased by a further 0.13 percentage points, and the CaCO 3 content increased by a further 1.33 percentage points. These results are consistent with the changes in the DTG peak shapes and support the judgment that CRCA and mineral-admixture addition can jointly reduce the reserve of alkaline calcium-bearing phases.

The decrease in Ca(OH) 2 content and increase in CaCO 3 content indicate that accelerated RCA carbonation promoted the conversion of some calcium-bearing phases in the adhered old mortar into carbonate phases. On the one hand, this phase transformation may reduce the reserve of Ca(OH) 2 capable of producing highly alkaline leachate in the system; on the other hand, the carbonate phases formed may improve the compactness of the old mortar and local ITZs. This explanation is consistent with the increased compressive strength, reduced leachate pH, and slight decrease in connected porosity exhibited by CR-PVEC relative to R-PVEC. After mineral-admixture addition, the Ca(OH) 2 content of PVEC continued to decrease, indicating that these admixtures could further regulate the calcium-bearing phase composition of the new paste beyond the effects of CRCA alone.

As shown in Figure 11, crystalline phases including Ca(OH) 2, calcite, and quartz could be identified in the XRD patterns of R-PVEC-0, CR-PVEC-0, and CR-PVEC-30. Compared with R-PVEC-0, the characteristic peaks of Ca(OH) 2 weakened and those of calcite intensified in CR-PVEC-0. At a 30% FA-SF replacement ratio, the characteristic peaks of Ca(OH) 2 in CR-PVEC-30 weakened further. This trend is consistent with the semi-quantitative TG/DTG results, indicating that both CRCA and mineral-admixture addition were associated with a reduction in the relative Ca(OH) 2 content of the PVEC system.

From R-PVEC-0 to CR-PVEC-0, the characteristic calcite peaks intensified, indicating that accelerated carbonation promoted the carbonation of some Ca(OH) 2 and reactive calcium-bearing hydration products in the adhered old mortar of RCA. The calcite peaks in CR-PVEC-30 retained a relatively high intensity, indicating that the carbonate phases formed by carbonation were preserved in the subsequent system. Meanwhile, the quartz peaks in CR-PVEC-30 were relatively stronger, which was associated with the crystalline quartz introduced by fly ash and the relative reduction in crystalline cement-hydration phases. Silica fume and C-S-H gel mainly exhibit amorphous or poorly crystalline characteristics and therefore do not readily form clear, sharp characteristic peaks in XRD patterns.

The TG/DTG and XRD results corroborate each other: RCA carbonation treatment mainly promoted the carbonation of some calcium-bearing phases in the adhered old mortar, while the addition of the FA-SF blend further reduced the relative Ca(OH) 2 content of the PVEC system. This evolution of calcium-bearing phases provides a phase-composition basis for subsequently explaining the reduction in leachate pH.

Evolution of the aggregate-paste interface: SEM-EDS

As shown in Figure 12, from the aggregate edge toward the new paste, the Ca/Si ratios of all three sample groups gradually transitioned from a highly fluctuating interfacial region to a relatively stable plateau region. In this study, the position at which the Ca/Si ratio changed from marked variation to relative stability was used as the criterion for the ITZ boundary (Pang et al., 2016; Bao et al., 2022).

On this basis, the ITZ thicknesses of R-PVEC-0, CR-PVEC-0, and CR-PVEC-30 were estimated to be 36 μm, 23 μm, and 15 μm, respectively. Compared with R-PVEC-0, the ITZ thickness of CR-PVEC-0 decreased from 36 μm to 23 μm, a reduction of 36.1%. At a 30% FA-SF replacement ratio, the ITZ thickness of CR-PVEC-30 decreased to 15 μm, representing an overall reduction of 58.3% relative to R-PVEC-0. This suggests that both CRCA and mineral-admixture addition may help improve the homogeneity of the aggregate-paste interface. Specifically, accelerated RCA carbonation can reduce interfacial heterogeneity by improving the surface layer and local defects of the adhered old mortar, while the addition of the FA-SF blend may improve the new paste and its adjacent interfacial region through fine-particle filling and pozzolanic reactions. The reduced estimated ITZ thickness is consistent with the increased compressive strength of PVEC and provides a possible interface-scale explanation for this change.

The SEM images (Figure 13a) further show differences in the interfacial morphology of different PVEC specimens. The aggregate-paste interface of R-PVEC-0 was relatively loose, with observable pores, microcracks, and local debonding, reflecting pronounced structural heterogeneity between the adhered old mortar of uncarbonated RCA and the new paste. After accelerated RCA carbonation, the interfacial continuity of CR-PVEC-0 improved, and granular deposits were visible in some pore and microcrack regions (Figure 13b). In combination with the measured TG/DTG and XRD results, these deposits may be related to CaCO 3 formed by carbonation. In CR-PVEC-30, the interfacial region was denser overall, pores and microcracks were further reduced, and the transition between the aggregate and paste was smoother (Figure 13c). These morphological characteristics are consistent with the decreasing trend in ITZ thickness shown by EDS line scanning. Accordingly, gel-like or fibrous products cannot be conclusively identified as C-S-H without supporting EDS point analysis or other spectroscopic evidence.

Hierarchical regulatory mechanism of CRCA and mineral admixtures

At all replacement ratios, the leachate pH of CR-PVEC was lower than that of the corresponding R-PVEC (Figure 14a), indicating that accelerated RCA carbonation could consistently reduce the leachate alkalinity of R-PVEC. To directly compare the pH-regulating effect of carbonation treatment at different replacement ratios, the difference in leachate pH between the uncarbonated RCA group and the CRCA group at the same replacement ratio was defined as the “apparent pH reduction” calculated using Equation 3: where pH R-PVEC and pH CR-PVEC are the leachate pH values of the uncarbonated RCA group and the CRCA group, respectively, at the same replacement ratio; ΔpH carb represents only the apparent pH difference between the two treatment pathways.

At replacement ratios of 0%, 20%, 25%, 30%, and 35%, ΔpHcarb values were 0.94, 0.84, 0.97, 1.37, and 1.31, respectively. The 30% FA-SF replacement group exhibited the largest apparent pH reduction, namely ΔpHcarb = 1.37, indicating that CRCA had a more pronounced effect on the final leachate pH with this cementitious material composition. It should be noted that pH is a logarithmic indicator, and a pH difference cannot be converted directly into a linear reduction in the concentration of alkaline substances. Therefore, ΔpH carb is used only to compare the apparent pH regulation effects among different groups and does not represent the absolute contribution of the carbonation reaction to the consumption of alkaline components.

At FA-SF replacement ratios of 20%, 25%, 30%, and 35%, the leachate pH decreased relative to CR-PVEC-0 by 0.25, 0.56, 1.11, and 1.30, respectively (Figure 14b). These results indicate that the addition of the FA-SF blend can further reduce leachate pH beyond the effect of accelerated RCA carbonation, possibly through the cement replacement effect, the consumption of part of the Ca(OH) 2 by pozzolanic reactions, and optimization of the paste-interfacial structure.

As the FA-SF replacement ratio increased, the pH reduction caused by the mineral admixtures gradually increased, indicating that the composition of the new PVEC paste had an increasingly strong influence on the alkalinity-reduction effect. It should be emphasized that the comparison in this study is an apparent intergroup comparison based on pH differences; the independent and interactive effects of CRCA and the mineral admixtures were not rigorously separated quantitatively. Therefore, this is not expressed as a “contribution rate” in the strict sense. Carbonation of the old mortar. Accelerated RCA carbonation promoted the conversion of some Ca(OH) 2 and reactive calcium-bearing phases in the adhered old mortar to CaCO 3, reducing the reserve of alkaline calcium-bearing phases carried into the system by the RCA. The decrease in Ca(OH) 2 content and increase in CaCO 3 content in the TG/DTG results, together with the weakening of the characteristic Ca(OH) 2 peaks and intensification of the calcite peaks in the XRD patterns, jointly support this judgment.

Regulation of the new paste by mineral admixtures. By replacing part of the cement, the addition of the FA-SF blend reduced the potential formation of alkaline hydration products such as Ca(OH) 2. Meanwhile, the pozzolanic reactions may have further consumed part of the Ca(OH) 2 and improved the new-paste structure, thereby reducing leachate pH.

Contribution of interfacial-structure optimization. The deposition of carbonation products, microfilling by silica fume, and formation of secondary reaction products jointly improved the aggregate-paste interface, reducing the ITZ thickness from 36 μm in R-PVEC-0–23 μm in CR-PVEC-0 and further to 15 μm in CR-PVEC-30. Interfacial densification can improve load-transfer continuity in PVEC and may affect the transport pathways of alkaline components; however, this interpretation requires further verification through direct ion-transport measurements.

In summary, CRCA and the addition of the FA-SF blend formed a hierarchical regulatory pathway involving old-mortar treatment, new-paste optimization, and interfacial structure improvement. CRCA mainly reduced the reserve of alkaline calcium-bearing phases carried by the RCA itself by promoting the conversion of some Ca(OH) 2 and reactive calcium-bearing phases in the adhered old mortar to CaCO 3. The addition of the FA-SF blend further reduced the relative Ca(OH) 2 content in the new paste through partial cement replacement, pozzolanic reactions, and optimization of the interfacial region. The two effects jointly improved the aggregate paste interface, increased the compressive strength of PVEC, and reduced leachate pH while preserving the principal connected pores. Because the concentrations of Ca 2+, Na +, K +, and OH − in the leachate and the ion-diffusion parameters were not measured directly in this study, the quantitative effect of interfacial densification on the migration of alkaline components remains to be verified in future research.

Through screening of parent concrete strength, accelerated RCA carbonation, and the addition of the FA-SF blend, this study investigated the effects of the initial quality of RCA, carbonation duration, and cementitious material composition on the compressive strength, connected porosity, and leachate pH of PVEC. The evolution of calcium-bearing phases and the aggregate-paste interfacial transition zone in PVEC was analyzed using TG/DTG, XRD, and SEM-EDS. The main conclusions are as follows: Parent concrete strength significantly affected the initial performance of R-PVEC. As the parent concrete strength grade increased from C30 to C50, the 28 days compressive strength of PVEC increased from 3.96 MPa to 5.15 MPa, the connected porosity remained within 27.14%–28.36%, and the leachate pH increased from 12.14 to 12.62. RCA prepared from higher-strength parent concrete enhanced the load-bearing capacity of the PVEC skeleton, but its leachate pH was also relatively high. Considering both initial strength and reduced-alkalinity requirements, RCA derived from C40 concrete was selected for the subsequent carbonation-duration experiments.

Under gaseous accelerated carbonation conditions of 20 °C, 70% relative humidity, and 20% CO 2 concentration, accelerated RCA carbonation significantly improved the compressive strength of PVEC and reduced its leachate pH. After 5 days of carbonation, the compressive strength of CR-PVEC was 5.63 MPa, reaching 98.4% of that of the 7 days carbonation group; the connected porosity was 27.99%, and the leachate pH was 11.68. Continued carbonation to 7 days produced only small changes in these indicators, indicating that 5 days was a reasonable carbonation duration that balanced treatment efficiency, load-bearing performance, connected pores, and leachate pH.

The addition of the FA-SF blend further optimized the overall performance of PVEC beyond the effects of CRCA alone. For CR-PVEC, the highest compressive strength, 6.56 MPa, was obtained at a replacement ratio of 25%. At a replacement ratio of 30%, the compressive strength, connected porosity, and leachate pH were 6.32 MPa, 28.55%, and 10.51, respectively. Therefore, the 30% replacement ratio provided a better balance among PVEC strength, connected pores, and reduced alkalinity.

The microscopic results showed that both accelerated RCA carbonation and the addition of the FA-SF blend could regulate the calcium-bearing phases and aggregate paste interfacial structure in the system. Semi-quantitative TG/DTG results showed that the estimated Ca(OH) 2 contents in R-PVEC-0, CR-PVEC-0, and CR-PVEC-30 were 1.21%, 0.83%, and 0.70%, respectively, while the corresponding CaCO 3 contents were 21.89%, 24.33%, and 25.66%. Correspondingly, the ITZ thickness decreased from 36 μm to 23 μm and then to 15 μm. Accelerated RCA carbonation acted mainly by converting some alkaline calcium-bearing phases in the adhered old mortar and improving its local defects, while the addition of the FA-SF blend further regulated the new paste and interfacial region. Together, the two measures improved load transfer through the PVEC skeleton and reduced leachate pH.

In this study, connected porosity was used to characterize the interconnected pore structure of PVEC, whereas its permeability was not directly measured. Moreover, the alkalinity-reduction effect was evaluated primarily based on leachate pH measured under a specified solid-to-liquid ratio and immersion duration, which may not fully reflect the long-term release of alkaline substances under dynamic water-flow conditions. Future research should therefore determine the permeability coefficient, leached ion concentrations, CO 2 uptake, and energy consumption, and conduct dynamic and long-term leaching tests as well as plant growth and adaptability experiments to comprehensively evaluate the hydraulic performance, long-term alkalinity reduction, and practical ecological applicability of CR-PVEC.