Synergistic Performance and Microscopic Mechanisms of Mortar Incorporating Recycled Brick Fine Aggregate and Brick Powder

Abstract

The recycling of waste clay bricks as raw materials for cement-based materials presents an effective solution to ecological pollution and resource shortages. Previous research has separately examined the effects of recycled brick fine aggregate and recycled brick powder on mortar or concrete, but few studies have investigated their combined use. This study aims to clarify the synergistic effect of recycled brick fine aggregate (RBA) and recycled brick powder (RBP) on mortar performance, quantify the influence of the RBP substitution rate on hydration characteristics and microstructural evolution, and determine the optimal mix proportion and curing system for fully recycled brick mortar. Mortar was prepared using 100% RBA and RBP at substitution rates of 0%, 10%, 20%, and 30%. The physical properties, mechanical performance, and durability of the mortar were evaluated, alongside an analysis of its microstructural morphology, mineral composition, and pore structure. The results indicate that adding an appropriate amount of RBP helped maintain the flowability of the mortar. As the RBP substitution rate increased, the mortar strength generally decreased in the early stages, but long-term curing (≥90 days) effectively mitigated this decline. The inclusion of RBP improved chloride ion permeability, with the 20% substitution rate achieving a favorable balance between compressive strength, fluidity, and durability without significantly affecting carbonation resistance. Microstructural analysis revealed that RBP regulated the morphology of hydration products and optimized the pore structure of the mortar, while the mineral composition of hydration products was similar to that of natural mortar. These findings provide a theoretical basis and technical support for the high-value utilization of construction and demolition waste in cement-based materials.

1. Introduction

As the economy grows rapidly and large-scale infrastructure projects expand, concrete has emerged as the most commonly utilized construction material [1,2]. However, the production and use of concrete have contributed to environmental degradation and ecological damage due to the overexploitation of natural resources [3,4,5]. This not only harms the Earth’s ecosystem but also poses a significant threat to the sustainable development of human society [6,7,8]. Additionally, the rapid progress of urbanization has led to the demolition of old houses and buildings, generating substantial amounts of construction and demolition waste [9,10]. Many of these demolished structures were made of masonry, resulting in a large number of discarded clay bricks in the construction waste [11]. These discarded clay bricks are often landfilled after minimal processing, which occupies valuable land resources and contributes to environmental pollution [12,13]. Therefore, recycling waste clay bricks for resource utilization can conserve land resources, protect the ecological environment, and reduce the consumption of non-renewable natural resources [14,15].

The recycling of waste clay bricks into raw materials, such as aggregates and powders, provides a viable solution to mitigate ecological pollution and reduce resource depletion [16]. Researchers have crushed waste clay bricks into recycled coarse aggregates to produce recycled brick coarse aggregate concrete [17,18,19]. However, the low strength, high porosity, and significant water absorption of recycled brick coarse aggregates negatively impact the physical and mechanical properties of the produced concrete, particularly its compressive strength [20,21,22]. Furthermore, Ahmed et al. [23] discovered that replacing up to 75% of river sand with RBA had little effect on the compressive strength of the concrete. Even with a 100% substitution rate, the compressive strength only decreased by approximately 15%. Ge et al. [24] highlighted that pre-wetting RBA can reduce drying shrinkage strain in mortar. These findings demonstrate the potential of RBA for practical applications.

To further reduce carbon emissions within the construction industry, numerous researchers have investigated the incorporation of waste clay brick powder (RBP) into cement-based materials to decrease the amount of cement required [25,26,27]. Waste clay bricks contain amorphous compounds that exhibit pozzolanic activity, making them suitable as supplementary cementitious materials after crushing [28,29,30]. Compared to the high porosity and low strength of recycled brick coarse and fine aggregates, RBP offers distinct advantages. When waste clay bricks are crushed into RBP, their low-strength defects become less significant, and the pozzolanic activity of the material is effectively utilized. Several studies [9,31,32] have comprehensively examined the mechanical properties and durability of concrete and mortar incorporating RBP. These studies suggest that incorporating less than 25% RBP enhances the resistance of cement-based materials to chloride and sulfate ion attacks. Due to the pozzolanic activity of RBP, cement-based materials with this addition exhibit superior mechanical performance over time. Dai et al. [30] and Luga et al. [33] reached similar conclusions, finding that when the substitution rate of RBP is less than 20%, the compressive strength of the resulting concrete is similar to that of conventional natural concrete, which also shows a denser microstructure. However, when the substitution rate of RBP exceeds a certain threshold, the strength of cement-based materials decreases, porosity increases, and carbonation resistance deteriorates [34,35]. Overall, based on existing research, RBP is considered to have significant practical application potential.

Both RBA and RBP exhibit significant potential for practical applications. However, most studies to date have focused on the effects of either RBA or RBP on mortar or concrete, with limited research exploring the simultaneous incorporation of both into cement-based materials. The synergistic effect between RBA and RBP remains unclear, and the influence of the RBP substitution rate on the hydration mechanism and microstructural evolution of fully recycled brick mortar lacks systematic investigation. In this study, mortar was prepared using 100% RBA and RBP at substitution rates of 0%, 10%, 20%, and 30%. The physical properties, mechanical performance, and durability of the mortar were evaluated, and the microstructural morphology, mineral composition, and pore structure were analyzed to reveal the mechanism by which RBP affects the performance of fully recycled brick fine aggregate mortar. The objectives of this study are to: (1) clarify the synergistic effect of RBA and RBP on mortar performance; (2) quantify the influence of RBP substitution rate on hydration characteristics and microstructural evolution; and (3) determine the optimal mix proportion and curing system for fully recycled brick mortar.

2. Materials and Methods

2.1. Materials

In this study, ordinary Portland cement (P.O. 42.5) was used as a cementitious material, with its chemical composition listed in

Table 1. Chemical components of cement and RBP (wt%).

Chemical Composition	SiO2	Al2O3	Fe2O3	CaO	K2O	MgO	Na2O	TiO2
Cement	24.99	8.26	4.03	51.42	0.80	3.71	0.26	0.37
RBP	64.95	18.01	7.12	2.93	2.49	1.41	1.19	1.08

. Natural river sand (NA) with a fineness modulus of 2.46 served as the natural fine aggregate. RBA fully replaced NA, while RBP partially substituted cement. The particle size distribution curves of NA and RBA are presented in Figure 1. Both RBA and RBP were derived from waste clay bricks collected during the demolition of brick–concrete structures. The waste clay bricks were first dried to a constant weight in an oven, then crushed using a jaw crusher and a hammer crusher (Jaw Crusher/Hammer Crusher, Shanghai Jianshe Heavy Machinery Co., Ltd., Shanghai, China). Subsequently, RBA (with particle sizes ranging from 0.075 mm to 2.36 mm) and RBP (with particles smaller than 0.075 mm) were separated by a vibrating sieve (Xinxiang Hengyu Machinery Co., Ltd., Xinxiang, Henan, China). A schematic diagram of the preparation process for RBA and RBP is shown in Figure 2. The chemical composition of RBP was analyzed by X-ray fluorescence (XRF, Axios MAX, PANalytical B.V., Almelo, the Netherlands), as presented in Table 1. The pozzolanic activity of RBP was determined via the lime absorption method, with a 28-day activity index of 78%. The particle size distribution of RBP was measured using a laser particle size analyzer(Mastersizer 3000, Malvern Panalytical Ltd., Malvern, UK), exhibiting a specific surface area of 420 m²/kg and an amorphous content of 35% (determined by XRD quantitative analysis). The 24 h water absorption rates (in the saturated surface-dry state) of NA, RBA, and RBP were 1.2%, 12.5%, and 8.3%, respectively.

2.2. Mix Proportion

In this study, both natural fine aggregate mortar (N) and recycled brick fine aggregate mortar (RBAM) were prepared. Mortar N was made from cement, river sand, and water. RBAM was prepared using cement, RBA, water, and RBP at different dosages, named P0, P10, P20, and P30 (indicating RBP substitution rates of 0%, 10%, 20%, and 30%, respectively). The binder-to-aggregate ratio was set at 1:1 by mass (cement content of approximately 1000 kg/m³) to simulate high-strength mortar applications in precast components. This mix design was chosen because high cement content can compensate for the strength loss caused by RBA and RBP, and it aligns with the research focus on high-value utilization of recycled materials in high-performance mortars. The water-to-binder ratio (w/b) of N was 0.35. Due to the high-level water absorption of RBA, the w/b of RBAM was adjusted to 0.5 to maintain similar workability. To verify the influence of w/b, an additional control group (P0-0.35) with w/b = 0.35 was prepared, and its performance was compared with P0 (w/b = 0.5) and N (w/b = 0.35). The specific composition of all mortars is displayed in

Table 2. Mixed proportions of mortars.

Code	Cement (kg/m3)	Sand (kg/m3)	RBA (kg/m3)	RBP (kg/m3)	Water/Binder Ratio	Polycarboxylate Superplasticizer (kg/m3)
N	1000	1000	0	0	0.35	2
P0	1000	0	1000	0	0.5	2
P10	900	0	1000	100	0.5	2
P20	800	0	1000	200	0.5	2
P30	700	0	1000	300	0.5	2

.

2.3. Specimen Preparation

Vertical mortar mixers (HJW-60, Nanjing Tianshi Testing Equipment Co., Ltd., Nanjing, Jiangsu, China) were used to prepare the mixtures. First, the fine aggregate (NA or RBA) and cementitious materials (cement and RBP) were dry-mixed for 2 min at room temperature. RBA was pre-wetted in water at 20 ± 2 °C for 24 h and drained to a saturated surface-dry state (water content 10.2%) before mixing to reduce effective w/b. Next, 80% of the water was added, and the mixture was stirred for 4 to 5 min. The remaining 20% of the water was then introduced, and mixing continued for an additional 2 to 3 min. After the mixing process, the fresh mortar was poured into molds of three different sizes: 40 mm × 40 mm × 160 mm prisms, 70.7 mm × 70.7 mm × 70.7 mm cubes, and 100 mm (diameter) × 100 mm (height) cylinders. The molds were placed on a vibration table (ZT-100, Jinan Testing Machine Group Co., Ltd., Jinan, Shandong, China) for compaction (30 s) and then stored in a controlled environment with a relative humidity exceeding 95% and a temperature of (20 ± 2) °C for 24 h. Following this initial curing phase, the specimens were demolded and subjected to further curing under the same temperature and humidity conditions until the testing age was reached.

2.4. Measurements

2.4.1. Physical Properties

Fluidity test: This test was conducted in accordance with GB/T 2419-2005 [36]. Fresh mortar was poured into a standard conical mold (height: 50 mm, top diameter: 70 mm, bottom diameter: 100 mm) placed on a flow table. The mold was carefully lifted, and the flow table was jolted 25 times within 15 s. The diameters of the spread mortar were measured in two mutually perpendicular horizontal directions, and the average value was recorded as the fluidity (consistently referred to as “flowability” throughout the text). The test apparatus is illustrated in Figure 3a.

Capillary water absorption test: This test was performed following ASTM C1585 [37]. Cubic specimens (70.7 mm × 70.7 mm × 70.7 mm) cured for 28 days were dried in an oven at 45 °C to a constant weight, cooled to room temperature, and sealed with wax on all surfaces except one. The specimens were immersed in water at (20 ± 2) °C, with the water level 2–3 mm above the bottom. The mass was measured at 15, 30, 45, 60, 90, 120 min, and 4, 8, 12, 24 h. The water absorption rate (%) was calculated using the formula: (mt − m0)/m0 × 100%, where m0 is the dry mass and mt is the mass at time t. The dry density of each mixture was measured to analyze the effect of pore volume.

2.4.2. Mechanical Properties

Flexural and compressive strengths were tested at curing ages of 3, 7, 28, and 90 days. Flexural strength was determined using 40 mm × 40 mm × 160 mm prisms in a three-point bending test (ASTM C293 [38]). Compressive strength was tested on the fractured prism halves (40 mm × 40 mm × 40 mm) using a compressive testing machine. Three specimens were tested per group, and the average value was recorded (error bars representing standard deviation are included in the figures). The test apparatus is shown in Figure 3c.

2.4.3. Durability

Chloride ion penetration test: This test was conducted using the Rapid Chloride Migration (RCM) method in accordance with GB/T 50082-2009 [39]. Cylindrical specimens (100 mm × 50 mm) cured for 28 and 90 days were used. The anodic cell contained 0.3 mol/L NaOH solution, and the cathodic cell contained 3% NaCl solution. The test voltage was 60 V, and the test temperature was (20 ± 2) °C. The chloride ion diffusion coefficient (D) was calculated using Equation (1). Three specimens were tested per group, and the average value was recorded.

Carbonation test: Prismatic specimens (40 mm × 40 mm × 160 mm) cured for 28 days were placed in a carbonation chamber (temperature: 20 ± 2 °C, relative humidity: 70 ± 5%, CO₂ concentration: 5 ± 0.1%) in accordance with GB/T 50082-2009 [39]. Carbonation depth was measured at 3, 7, 14, and 28 days using phenolphthalein indicator. The unusual 0 mm carb19onation depth of N at 28 days is attributed to the dense surface layer formed by low w/b and sufficient hydration; this is discussed in detail in Section 4.

Equation (1): Chloride ion diffusion coefficient calculation formula:

$$D={\displaystyle \frac{RTL}{nFU}}\mathit{ln}\left(1+{\displaystyle \frac{nF∆C}{RTI}}\right)$$

(1)

where R = gas constant (8.314 J/(mol·K)); T = absolute temperature (K); L = specimen thickness (m); n = valence of chloride ion (1); F = Faraday constant (96,485 C/mol); U = applied voltage (V); ΔC = chloride ion concentration difference (mol/m³); and I = current (A).

2.4.4. Microscopic Performance

SEM analysis: This test was conducted using a Zeiss Supra55 SEM. Specimens (5 mm × 5 mm × 2 mm) cured for 28 days were immersed in anhydrous ethanol to stop hydration, dried in a vacuum desiccator, and gold-coated. SEM images were captured at magnifications of ×1000, ×5000, ×10,000, and ×15,000 to observe microstructure and interfacial transition zone (ITZ). Marked areas in images are annotated in captions to clarify key features.

XRD analysis: This analysis was performed using a Bruker APEX II DU XRD system (40 kV, 100 mA, scanning range: 5–80° 2θ, step size: 0.15° 2θ). Powder samples (<0.075 mm) cured for 28 days were analyzed. Thermogravimetric Analysis (TGA) was conducted to quantify Ca(OH)₂ content (evidence of pozzolanic reaction) using a Netzsch STA 449F3 thermogravimeter (heating rate: 10 °C/min, N₂ atmosphere). Energy-Dispersive X-ray Spectroscopy (EDX) was used to analyze elemental distribution in hydration products.

MIP analysis: This test was conducted using an AUTOPORE IV 9510 mercury porosimeter. Samples (4 mm × 4 mm × 4 mm) cured for 28 and 90 days were dried in a vacuum oven. Pore size distribution (6 nm–300 μm) was measured, and pores were classified as gel pores (<10 nm), fine capillary pores (10–50 nm), medium capillary pores (50–100 nm), large capillary pores (100 nm–10 μm), and large pores (>10 μm) [40,41,42].

3. Results

3.1. Fluidity of Mortar

Figure 4 provides a comparison of the flowability test results for natural and recycled mortars. The flow value of N (w/b = 0.35) was 152 mm, while P0 (w/b = 0.5) had a similar flow value of 148 mm. The flow value of P0-0.35 (w/b = 0.35 with superplasticizer) was 150 mm, confirming that w/b adjustment effectively maintains workability. As the RBP substitution rate increased, the flowability of RBAM decreased: P10 (128 mm, −13.5% vs. P0), P20 (130 mm, −12.4% vs. P0), and P30 (118 mm, −20.4% vs. P0). The flow values of P10 and P20 were similar, indicating no significant change in flowability for RBP substitution rates of 10–20%. The quantitative statistical results of the flowability tests for all mortar groups are shown in Figure 5, which clearly reflects the slight difference in flow values between P10 and P20, indicating no significant change in flowability for RBP substitution rates of 10–20%.

3.2. Capillary Water Absorption of Specimens

Figure 6 shows the capillary water absorption of N and RBAM. During the first 12 h, all specimens absorbed over 80% of their total water absorption. The water absorption of RBAM was significantly higher than that of N: P0 (3.2%), P10 (3.6%), P20 (3.8%), and P30 (4.3%) vs. N (2.1%) at 24 h. The dry density of RBAM (1850–1920 kg/m³) was lower than that of N (2050 kg/m³), indicating that increased water absorption is partially due to higher total pore volume. As the RBP content increased, water absorption increased, attributed to the higher specific surface area and porosity of RBP.

3.3. Mechanical Properties

3.3.1. Compressive Strength

Figure 7 illustrates the compressive strength of N and RBAM. The 3- and 7-day compressive strengths of N (w/b = 0.35) and P0 (w/b = 0.5) were similar (3-day: 42.5 MPa vs. 41.2 MPa; 7-day: 51.3 MPa vs. 49.8 MPa), but P0 had lower strengths at 28 and 90 days (28-day: 62.5 MPa vs. 48.4 MPa; 90-day: 68.3 MPa vs. 51.1 MPa). P0-0.35 (w/b = 0.35) had higher strengths than P0 (28-day: 56.8 MPa; 90-day: 61.5 MPa), confirming the negative impact of high w/b. As the RBP substitution rate increased, early-age (3, 7, 28 days) compressive strength decreased, but 90-day strengths increased: P10 (90-day: 58.3 MPa, +14.1% vs. P0), P20 (90-day: 63.5 MPa, +24.3% vs. P0), and P30 (90-day: 57.2 MPa, +11.9% vs. P0). The 90-day compressive strength of P20 approached that of N (68.3 MPa).

3.3.2. Flexural Strength

Figure 8 displays the flexural strength of N and RBAM. P0 had lower flexural strengths than N at all ages (3-day: 5.8 MPa vs. 6.6 MPa; 7-day: 7.2 MPa vs. 7.6 MPa; 28-day: 8.4 MPa vs. 10.1 MPa; 90-day: 8.9 MPa vs. 11.7 MPa). P0-0.35 had higher flexural strengths than P0 (28-day: 9.2 MPa; 90-day: 10.3 MPa). As the RBP substitution rate increased, early-age flexural strengths decreased slightly, but 90-day strengths increased: P10 (90-day: 10.2 MPa, +14.6% vs. P0), P20 (90-day: 11.3 MPa, +27.0% vs. P0), and P30 (90-day: 10.0 MPa, +12.4% vs. P0). The 90-day flexural strength of P20 exceeded the 28-day flexural strength of N (10.1 MPa).

3.4. Chloride Penetration Resistance

Figure 9 presents the chloride ion diffusion coefficient (D) of N and RBAM. After 28 days of curing, N exhibited a lower D value (2.1 × 10⁻¹² m²/s) compared to RBAM (P0: 5.8 × 10⁻¹² m²/s; P10: 6.3 × 10⁻¹² m²/s; P20: 6.5 × 10⁻¹² m²/s; P30: 7.2 × 10⁻¹² m²/s). After 90 days of curing, the D values of all groups decreased significantly: N (0.4 × 10⁻¹² m²/s, a reduction of 81.0%), P0 (0.6 × 10⁻¹² m²/s, a reduction of 89.7%), P10 (0.3 × 10⁻¹² m²/s, a reduction of 95.2%), P20 (0.3 × 10⁻¹² m²/s, a reduction of 95.4%), and P30 (0.9 × 10⁻¹² m²/s, a reduction of 87.5%). P20 showed the most remarkable improvement in chloride ion resistance, which was attributed to its optimized pore structure and sufficient pozzolanic reaction. The visual test results of chloride ion penetration for mortar specimens at 28 d and 90 d are displayed in Figure 10, which intuitively shows the significant reduction in chloride ion penetration of RBAM after long-term curing, with P20 showing the most remarkable improvement in chloride ion resistance, attributed to its optimized pore structure and sufficient pozzolanic reaction.

3.5. Carbonization Resistance

Figure 11 quantifies the carbonation depth of N and RBAM. N had a carbonation depth of 0 mm at 28 days, while P0 had a small carbonation depth (0.8 mm). As the RBP content increased, carbonation depth increased: P10 (2.3 mm), P20 (3.5 mm), and P30 (4.8 mm) at 28 days. However, the carbonation depth of all RBAM samples remained relatively small, indicating that moderate RBP addition does not significantly reduce carbonation resistance. The actual carbonation test characterization results of each mortar group are presented in Figure 12, which visually verifies the low carbonation degree of RBAM with RBP substitution rates within 20%, further confirming the good carbonation resistance of the mortar under this mix proportion.

3.6. Microscopic Performance

3.6.1. SEM Analysis

Figure 13 presents the microstructure of N and P0 after 28 days. N had a dense microstructure with a compact ITZ and uniform hydration products (Figure 13a,b). P0 had more pores and cracks, a wider ITZ, and uneven hydration product distribution (Figure 13c,d). Figure 14 shows the microstructure of P10, P20, and P30 after 28 days. With increasing RBP content, porosity increased, but P20 showed a relatively dense structure at high magnification (×15,000), with fine C-S-H gel filling pores (Figure 14e). The ITZ of P20 was more compact than that of P0 and P30, indicating a synergistic effect between RBP and cement hydration.

3.6.2. XRD and TGA

Figure 15 presents the XRD spectra of N and RBAM after 28 days. The primary hydration products of N were C-S-H gel, Ca(OH)₂, ettringite (AFt), and portlandite. The XRD spectra of RBAM were similar to those of N, with no new phases detected. However, the intensity of Ca(OH)₂ diffraction peaks decreased with increasing RBP content, indicating pozzolanic reaction (consuming Ca(OH)₂). TGA results (Figure 16) showed that the Ca(OH)₂ content of P20 (3.2%) was lower than that of P0 (5.8%), confirming the pozzolanic activity of RBP. EDX analysis revealed that the C-S-H gel in P20 had a higher Si/Ca ratio (0.85) than that in P0 (0.62), indicating enhanced pozzolanic reaction.

3.6.3. MIP Analysis

The porosity, average pore diameter, and median pore diameter of N cured for 28 days (N-28d), P0 cured for 28 days (P0-28d), P20 cured for 28 days (P20-28d), and P20 cured for 90 days (P20-90d), as measured by MIP, are displayed in Figure 17. Overall, the porosity of N was significantly lower than that of RBAM. The porosity of P20 with RBP addition was slightly higher than that of P0 without RBP. As the curing age increased, the porosity of P20 decreased and became slightly lower than that of P0. However, the average pore diameter of P20 was smaller than both P0 and N, and it further decreased with prolonged curing. Additionally, the median pore diameter of P20 cured for 90 days was also lower than that of N. Combining the results of porosity, average pore diameter, and median pore diameter, it is evident that the pores in RBAM with RBP addition tended to be concentrated in smaller sizes. With extended curing time, the porosity, average pore diameter, and median pore diameter of RBAM decreased, likely due to the continuous hydration of cement and ongoing interaction with RBP, which gradually improved the pore structure of the mortar.

Figure 18 and Figure 19 show the incremental pore size distribution curves and cumulative pore size distribution curves of N-28d, P0-28d, P20-28d, and P20-90d. The pore size distribution curves illustrate the distribution of pores across different size ranges for each sample. Compared to RBAM, N had a relatively uniform pore distribution. In contrast, P20 with RBP addition had fewer large pores and more small pores than P0. As the curing age increased, the number of large pores in P20 continued to decrease, while the number of small pores increased. Furthermore, the cumulative pore volume of P20 was smaller than that of P0, and it continued to decrease with extended curing. This indicates that RBP had a refining effect on the pore structure of RBAM, and sufficient curing time enhanced this refining effect.

To further quantitatively characterize the pore size distribution of different mortars, the pores were classified into gel pores (<10 nm), fine capillary pores (10–50 nm), medium capillary pores (50–100 nm), large capillary pores (100 nm–10 μm), and large pores (>10 μm). As shown in Figure 20, although N had lower porosity, it contained a relatively high proportion of large pores (23.4%). In contrast, the proportion of large pores in P0 and P20 was lower (11.1–13.0%). Compared to P0, P20 exhibited an increased volume fraction of fine capillary pores, while the volume fractions of large pores, medium capillary pores, and large capillary pores decreased. When the curing age increased from 28 days to 90 days, the volume fractions of gel pores and fine capillary pores in P20 continued to increase, while those of large pores, medium capillary pores, and large capillary pores further decreased. Studies have shown that pores smaller than 100 nm can be considered harmless or minimally harmful. The volume fraction of harmful pores (>100 nm) in P0 was 38.7%, while that in P20 was 32.7%, indicating that RBP optimized the pore structure of RBAM. With an extended curing time, the volume fraction of harmful pores in P20 further decreased to 31.3%.

4. Discussion

4.1. Effect of RBP Substitution Rate on Mortar Performance

The incorporation of RBP significantly affects the performance of RBAM. At early ages (3, 7, and 28 days), the increasing RBP substitution rate leads to decreased mechanical strength and flowability, attributed to: (1) physical dilution of cement, reducing the amount of cementitious material; (2) high specific surface area of RBP, increasing water demand and reducing workability; and (3) an insufficient pozzolanic reaction, resulting in fewer hydration products. However, at 90 days, the strength of RBAM with RBP increases significantly, especially P20, which achieves the highest strength. This is because RBP undergoes pozzolanic reaction with Ca(OH)₂ (Equations (2) and (3)) to form additional C-S-H gel, densifying the matrix and ITZ. The 20% RBP substitution rate balances cement dilution and pozzolanic activity, achieving optimal performance. Excessive RBP (30%) leads to reduced strength due to excessive dilution and increased porosity.

Equation (2): Reaction between RBP (SiO₂) and Ca(OH)₂:

$${SiO}_2+{Ca\left(OH\right)}_2+H_{2}O\to C-S-H gel$$

(2)

Equation (3): Reaction between RBP (Al₂O₃) and Ca(OH)₂:

$${Al}_2{O}_3+{Ca\left(OH\right)}_2+H_{2}O\to C-A-H gel$$

(3)

4.2. Effect of Curing Time on Mortar Performance

Curing time has a significant impact on RBAM performance. Long-term curing (≥90 days) promotes cement hydration and pozzolanic reaction of RBP, leading to: (1) increased hydration products (C-S-H, C-A-H), filling pores and densifying the matrix; (2) reduced porosity and average pore diameter, improving impermeability and durability; and (3) enhanced ITZ bonding, increasing mechanical strength. The 90-day compressive strength of P20 approaches that of N, indicating that long-term curing is essential for RBAM with RBP. This is consistent with previous studies (Dai et al. 2024 [30]; Luga et al. 2025 [33]), which reported that RBP-containing cement-based materials require long curing to develop optimal performance.

4.3. Microstructural Evolution Mechanism

The microstructural evolution of RBAM with RBP is governed by two competing mechanisms: (1) Filler effect: RBP particles fill small pores, reducing porosity and improving compactness. (2) Pozzolanic effect: RBP reacts with Ca(OH)₂ to form C-S-H gel, densifying the matrix and ITZ. At early ages, the filler effect dominates, while the pozzolanic effect becomes significant at long ages. SEM and MIP results show that RBP optimizes the pore structure of RBAM, reducing the number of large pores and increasing fine capillary pores. XRD and TGA results confirm the consumption of Ca(OH)₂ by RBP, indicating pozzolanic activity. The synergistic effect of filler and pozzolanic effects explains the improved performance of P20 at 90 days.

4.4. Interpretation of Unusual Results

Similar early-age strength of N and P0: Despite the higher w/b of P0 (0.5 vs. 0.35 for N), their early-age strengths are similar. This is because RBA absorbs water during mixing, reducing the effective w/b of P0. Additionally, the rough surface of RBA enhances mechanical interlocking with hydration products, offsetting the negative effect of high w/b. The P0-0.35 control group confirms that reducing w/b further improves strength.

0 mm carbonation depth of N: The dense surface layer of N, formed by low w/b and sufficient hydration, prevents CO₂ penetration. This is consistent with the low porosity and small average pore diameter of N, as measured by MIP.

Increased water absorption of RBAM: The higher water absorption of RBAM is due to higher total pore volume (lower dry density) and more open pore structure, as confirmed by MIP and dry density measurements.

4.5. Comparison with Existing Literature

The findings of this study are consistent with previous research. Ahmed et al. [23] reported that 100% RBA substitution reduces concrete strength by ~15%, which aligns with the 22.8% strength reduction in P0 at 28 days. Dai et al. [30] found that <20% RBP substitution maintains concrete strength, similar to the optimal 20% RBP substitution rate in this study. Ge et al. [24] highlighted the benefits of pre-wetting RBA, which is adopted in this study to reduce water absorption during mixing. However, this study is the first to systematically investigate the synergistic effect of 100% RBA and RBP on mortar performance, providing new insights into fully recycled brick mortar.

5. Conclusions

This study investigates the effect of the RBP substitution rate (0%, 10%, 20%, and 30%) on the performance of mortar with 100% RBA. The main conclusions are as follows:

(1)

The incorporation of 20% RBP achieves the optimal balance of performance. It maintains good flowability (130 mm), exhibits the highest 90-day compressive (63.5 MPa) and flexural (11.3 MPa) strengths, and improves chloride ion resistance (D = 0.3 × 10⁻¹² m²/s at 90 days) without significantly reducing carbonation resistance (3.5 mm at 28 days).

(2)

Long-term curing (≥90 days) is essential for RBAM with RBP. It promotes pozzolanic reaction and pore structure optimization, significantly improving mechanical strength and durability. The 90-day strength of P20 approaches that of natural mortar, indicating the potential for practical application.

(3)

RBP improves the microstructural characteristics of RBAM. It acts as a filler to reduce large pores and undergoes pozzolanic reaction to form additional C-S-H gel, densifying the matrix and ITZ. The 20% RBP substitution rate optimizes the pore structure, reducing the volume fraction of harmful pores (>100 nm) to 32.7%.

(4)

The synergistic effect of RBA and RBP is governed by filler and pozzolanic effects. RBA provides mechanical interlocking, while RBP optimizes the microstructure and enhances hydration. This study provides a theoretical basis for the high-value utilization of construction and demolition waste in cement-based materials.

It should be noted that besides the mechanical strength and durability focused on in this study, other properties, such as shrinkage and frost resistance, are also of great significance for the practical application of RBAM. Due to space limitations, this study only investigated the core performance indicators along with the microstructure. In future work, we will conduct in-depth research on properties, including shrinkage and frost resistance, to comprehensively evaluate the practical application potential of RBAM.