Enhancing the Mechanical Properties of Recycled Aggregate Concrete: A Comparative Study of Basalt- and Glass-Fiber Reinforcements

Abstract

Recycled aggregate concrete (RAC) holds significant promise for reducing the environmental impact of the construction industry. However, the poor mechanical properties of RAC compared to conventional concrete are mainly due to the porous and soft nature of recycled aggregates. While fiber reinforcement has been proposed as a promising method to address this issue, existing studies primarily focus on steel and polypropylene fibers, with limited systematic comparison of alternative fiber types and dosages. In particular, the mechanical enhancement mechanisms of basalt and glass fibers in RAC remain underexplored, and there is a lack of predictive models for strength behavior. This study evaluates the effects of basalt and glass fibers on RAC through uniaxial compression, splitting tensile, and three-point bending tests. Nine mixtures with varying fiber types and volume fractions (1.0–2.5%) were tested, and results were compared to plain RAC. Key properties such as strength, energy absorption, toughness, and flexibility were analyzed using load–displacement curves and advanced toughness indices. Both fiber types improved tensile and flexural properties, with glass fibers showing superior performance, particularly at 1.5% content, where the splitting tensile strength increased by up to 40% and the flexural strength improved by 42.19%. Basalt fibers dispersed more uniformly but were less effective in enhancing toughness and crack resistance. Excessive fiber content reduced matrix homogeneity and mechanical performance. Optimal fiber dosages were identified as 1–1.5% for glass fibers and 1–2% for basalt fibers, depending on the targeted property. Predictive formulas for the flexural strength of fiber-reinforced RAC are also proposed, offering guidance for the design of structural RAC elements.

1. Introduction

The annual surge in construction and demolition waste poses considerable burdens on environmental sustainability [1]. It is imperative to develop and implement strategies that protect and conserve natural resources while reducing carbon emissions [2,3,4]. In this context, recycled aggregate concrete (RAC), made from processed waste concrete materials, is increasingly valued for its potential to address environmental concerns by reducing the need for natural aggregate, recycling construction and demolition waste, and then minimizing the carbon footprint [5,6]. As a kind of green and low-carbon construction material, it provides a strategy for addressing the global warming problem. Therefore, the study on the workability, mechanical properties, and durability of RAC is of great significance in promoting the sustainable development of the construction and building industry.

Recently, studies on the physical and mechanical properties of RAC have been carried out worldwide to understand and improve the performance of RAC [7,8]. Wang et al. [7] reviewed the history, recycling, reuse, and manufacturing processes of RAC, as well as its inherent defects and material properties. The common properties of RAC, such as water absorption, density, and crushing value, as well as the effects of the effective water–cement ratio, aggregate–cement ratio, and the microstructure characteristics, were also analyzed [9]. Studies have shown that RAC generally requires more water to achieve the same workability as normal concrete. Job Thomas et al. [10] pointed out that when recycled aggregate (RA) is used to replace natural aggregate, the workability of concrete decreases, the density is reduced, and the slump of fresh RAC can be maintained between 50 and 100 mm without superplasticizers.

The mechanical properties and durability of RAC are key indicators of its feasibility as a structural material. It is reported that the mechanical properties of RAC may be generally lower than those of normal concrete (NC) [7]. The incorporation of recycled aggregates can lead to decreased mechanical properties, including lower compressive and tensile strengths and reduced durability. This is mainly due to the poor physical properties of RA, such as high water absorption and weak bonding of the interfacial transition zone [11,12,13]. In addition, the durability of RAC is another key factor in its application in practical engineering. Compared with ordinary concrete, the durability of RAC is generally weaker, mainly due to the attached mortar on the surface of RA [14]. Guo et al. [15] studied the durability properties of RAC, including the resistance to chloride ion penetration, water absorption, and acid attack, and reported that the attached mortar on the surface of RA causes higher inherent porosity, which reduces the durability of RAC.

With further research on RAC, more studies were focused on improving the mechanical properties, durability, and environmental adaptability of RAC [16,17]. A variety of methods have been proposed to improve the performance of RAC, including removing or strengthening the adhesive mortar on the surface of RA, and adding minerals, nanomaterials, fibers, and other modified materials [7,18]. The addition of fibers has been shown to significantly improve the mechanical properties of RAC, including compressive strength, flexural strength, tensile strength, and toughness [19]. For example, Wang et al. [20] studied the 3D microstructure characteristics and strength-softening mechanism of fiber-reinforced recycled aggregate concrete (FR-RAC). It was reported that fibers increased post-peak strength-softening resistance by up to 30%. The influence of steel fiber parameters, such as the fiber aspect ratio and content, on the fresh and hardened properties and fracture mechanical properties of RAC was also systematically studied [21,22]. In addition, the introduction of polypropylene fiber and glass fiber can improve the crack resistance, durability, toughness, and even the impact resistance of RAC [23].

Recent studies have also highlighted the potential of basalt fiber (BF) and glass fiber (GF) in enhancing the mechanical performance of recycled aggregate concrete (RAC) [24,25]. Basalt fiber, known for its high tensile strength, thermal stability, and corrosion resistance, has demonstrated significant improvement in the compressive and tensile strengths of RAC [24,25,26,27,28,29]. Fang et al. [26] found that RAC incorporating 0.1% BF achieved a 28-day compressive strength of 34.4 MPa and a splitting tensile strength of 3.89 MPa; however, the slump decreased significantly with increasing BF content, reflecting the negative impact of higher fiber dosage on workability. Katkhuda et al. [27] reported that the compressive strength of concrete containing 20% recycled coarse aggregate (RCA) increased by 9% when a 1% volume fraction of basalt fibers was added. Shoaib et al. [30] showed that at a basalt fiber reference of 1.5%, the tensile strength of recycled concrete was maximized, with up to 92% strength recovery. Shi et al. [31] investigated the effect of different recycled coarse aggregate (RCA) substitution rates on the load-bearing capacity and deformation capacity of BF-reinforced recycled aggregate concrete and explored the optimal BF admixture for different RCA substitution rates. Glass fiber has also been widely used due to its high aspect ratio, lightweight nature, and resistance to chemical attack [32,33,34]. In the research of Ali et al. [35], they found that the incorporation of 0.5% by volume glass fibers in concrete made with 100% recycled coarse aggregate (RCA) resulted in a 22% increase in matrix tensile strength and a 31.9% increase in flexural strength, but not a significant increase in compressive strength, and that the addition of a plasticizer was required to improve the RAC’s workability. Also, the replacement of natural coarse aggregate (NCA) with recycled coarse aggregate (RCA) reduced the compressive strength, splitting tensile strength, and flexural strength of concrete by approximately 12%, 11%, and 8%, respectively. Zhang et al. [32] investigated the effect of glass fiber content on the performance of recycled concrete and observed that 7.5 kg/m³ optimized the pore structure, while an increase to 10 kg/m³ led to a decline in mechanical properties and a notable reduction in workability.

In summary, fiber reinforcement is an effective means to improve the performance of RAC. While numerous studies have investigated the behavior of fiber-reinforced RAC, the majority have concentrated on conventional fibers such as steel or polypropylene, with limited systematic evaluation of mineral fibers like basalt and glass. Despite their favorable mechanical and chemical properties, the effects of basalt fiber (BF) and glass fiber (GF) on the strength, toughness, and durability of RAC remain insufficiently explored. Particularly, there is a lack of comprehensive studies that compare the performance of BF- and GF-reinforced RAC across a range of fiber volume fractions, especially at higher contents (1.0% to 2.5%). Moreover, the interactions between fiber type, content, and recycled aggregate characteristics, as well as their influence on failure modes and reinforcement mechanisms, have not been fully addressed. To further systematically evaluate the effects of different fiber types and contents on the properties of RAC, a series of laboratory experiments were conducted, including uniaxial compression tests, splitting tensile tests, and three-point bending tests. The strength, toughness, energy absorption, deformation capacity, and load–displacement curves of RAC specimens reinforced with basalt and glass fibers were comparatively analyzed, with high volume fractions increasing from 1% to 2.5%. Moreover, the failure patterns and the reinforced mechanism are also discussed. Finally, simplified empirical models for compressive toughness indexes and a flexural strength prediction model were established. The findings also offer practical guidance for selecting fiber types and dosages to enhance the mechanical performance of RAC in sustainable construction.

2. Experimental Program and Setup

2.1. Sample Preparation

The mix proportion of the present recycled aggregate concrete is shown in

Table 1. Mix proportion of recycled aggregate concrete.

Sample	Cement	Fine Aggregate	Coarse Aggregate	Water	Fiber Type	Volume Fraction
P	1.0	2.0	2.75	0.55	/	/
B1.0	1.0	2.0	2.75	0.55	Basalt	1.0%
B1.5	1.0	2.0	2.75	0.55	Basalt	1.5%
B2.0	1.0	2.0	2.75	0.55	Basalt	2.0%
B2.5	1.0	2.0	2.75	0.55	Basalt	2.5%
G1.0	1.0	2.0	2.75	0.55	Glass	1.0%
G1.5	1.0	2.0	2.75	0.55	Glass	1.5%
G2.0	1.0	2.0	2.75	0.55	Glass	2.0%
G2.5	1.0	2.0	2.75	0.55	Glass	2.5%

. The mix ratio of cement, fine aggregate, and coarse aggregate is 1.0:2.0:2.75 and the water-cement ratio is 0.55. With the same mixture compositions and proportions, two different types of fibers were added with increasing volume fractions (1%, 1.5%, 2%, 2.5%). Ordinary Portland cement of grade 42.5 (OPC 42.5) was used in this study, and its properties are shown in

Table 2. Properties of ordinary Portland cement.

Specific Surface Area (m2/kg)	Setting Time (min)		Flexural Strength (MPa)		Compressive Strength (MPa)
	Initial Setting	Final Setting	3 d	28 d	3 d	28 d
355	233	293	5.5	8.2	26.6	50.4

. Natural river sand with a fineness modulus of 2.8 and a grain size of 0–2.4 mm was adopted as the fine aggregate. Recycled coarse aggregate was made of crushed concrete blocks collected from a demolition site in a residential area in Chongqing with a particle diameter range of 4.75–25.0 mm. The particle size distribution of fine and coarse aggregates is shown in Figure 1. The particle size distribution of fine aggregate and coarse aggregate is within the upper limit and lower limit of ASTM C33 [36].

The recycled aggregate selected for the experiment is shown in Figure 2. Compared with natural aggregate, the most obvious feature of recycled aggregate particles is that its surface is coated with hardened cement mortar. By observing and analyzing the surface of recycled aggregate, it is found that recycled aggregate particles can be mainly sorted into five types, i.e., gravel aggregate partially or wrapped by cement mortar, multi-gravel aggregate bonded by cement mortar, aggregate formed by cement mortar block and gravel, pure cement mortar block, and clean gravel aggregate.

The basic physical and mechanical properties of natural aggregate and recycled aggregate are shown in

Table 3. Physical and mechanical properties of aggregates.

Type	Apparent Density (kg/m3)	Water Absorption (%)	Moisture Content (%)	Crushing Index (%)	Needle Flake Content (%)
Natural aggregate	2670	0.82	0.72	8.36	17
Recycled aggregate	2630	3.55	0.45	18.53	11

. The apparent density of recycled aggregate is slightly smaller than that of natural aggregate. This is mainly because the density of hardened cement mortar in recycled aggregate is much smaller than that of gravel, which leads to the decreased density of recycled aggregate concrete. The water absorption and crushing value of recycled aggregate are much larger than those of natural aggregate, which has a negative impact on the strength of hardened concrete. The cement mortar wrapped with gravel and cement mortar block is widely distributed in the recycled aggregate, so the fragile cement mortar increases the crushing value in the aggregate test. The pores in the abandoned cement mortar will absorb more water when the concrete is mixed, which is the reason why the water absorption of the recycled aggregate is fairly higher than that of the natural aggregate. However, fortunately, after being subjected to the vibration load during the demolition process, the needle flake content of recycled aggregate is much lower than that of natural aggregate, which has a positive effect on the strength performance of hardened concrete. As shown in Figure 3, in terms of mix preparation, all dry components (cement, fine aggregate, and coarse aggregate) were first mixed for 5 min. Then, water was added in two stages and mixed for another 5 min. Finally, basalt or glass fibers were slowly introduced and blended for an additional 6 min to ensure uniform dispersion and to minimize fiber clustering.

A total of 81 samples, including plain concrete, were prepared to investigate the effects of fiber type and fiber content on the mechanical performance of recycled aggregate concrete under different loads. Samples of the compression and splitting tests are cubes with an edge length of 100 mm. Samples of the three-point bending test are prisms with a length of 160 mm and a cross-section size of 40 × 40 mm in accordance with the standard of GB/T 17671-1999 [37]. Three samples were tested for every kind of RAC under each loading condition to eliminate accidental errors.

Fibers used in the present experiment are shown in Figure 4. The addition of glass fiber and basalt fiber can improve the mechanical properties, durability, and crack resistance of renewable concrete [38,39,40,41]. Basalt fiber resists crack growth and improves durability through bridging, especially in terms of salt erosion resistance. Glass fiber enhances the early crack resistance of concrete. The use of these fibers also has economic and environmental advantages and promotes sustainable development [42,43]. The basic physical and mechanical properties of glass fiber and basalt fiber are exhibited in

Table 4. Properties of glass and basalt fiber.

Fiber Type	Density (g/cm3)	Elastic Modulus (GPa)	Tensile Strength (MPa)	Breaking Elongation (%)
Glass	2.68	80	1834	3.0
Basalt	2.64	75	1821	3.0

. Glass fiber has a density of 2.68 g/cm³, an elastic modulus of 80 GPa, a tensile strength of 1834 MPa, and a breaking elongation of 3.0%. In contrast, basalt fiber possesses a slightly lower density at 2.64 g/cm³, a similar elastic modulus of 75 GPa, a tensile strength of 1821 MPa, and a similar breaking elongation of 3.0%. These properties highlight the unique characteristics of each fiber type, with glass fiber showing greater tensile strength and basalt fiber exhibiting a higher modulus of elasticity.

To achieve good workability and uniform particle and fiber distribution, the mixing procedures for the current RAC were employed rigorously according to previous works [44,45]. After mixing, the fresh concrete was poured into 100 × 100 × 100 mm and 40 × 40 × 160 mm plastic molds and vibrated for 30 s. After 24 h, hardened concrete was demolded and then cured in a standard curing room for 28 days at a temperature of 20 ± 5 °C and relative humidity greater than 90%, abiding by ASTM C31 [46].

To investigate the microstructural characteristics of fiber-reinforced recycled aggregate concrete, SEM images were analyzed for specimens containing basalt fibers and glass fibers. As shown in Figure 5, the concrete matrix with basalt fibers (Figure 5a) demonstrates a dense microstructure with uniformly distributed recycled aggregates and well-anchored fibers. The basalt fibers show strong bonding with the surrounding cement paste, with minimal observable interfacial cracks or voids. This indicates effective stress transfer and crack-bridging capability, which helps improve the mechanical performance of RAC. In contrast, the specimen with glass fibers (Figure 5b) shows a relatively looser matrix with visible fiber pull-out and interfacial debonding. Several microcracks are observed around the fiber–matrix transition zones, and the recycled aggregates appear less tightly integrated.

2.2. Experimental Apparatus and Methodology

The compressive experiments were conducted according to ASTM C39 [47] using a computer-controlled electromechanical servo-hydraulic compression testing machine with a capacity of 2000 kN, as illustrated in Figure 6. Stress–strain data were recorded using an axially aligned linear variable differential transformer (LVDT). To minimize the effects of end friction, Vaseline was applied to the contact surfaces between the sample and the machine. Additionally, a spherically seated block, as shown in Figure 6a, often referred to as the upper bearing block or suspended block, was utilized above the sample. This block is designed to tilt, ensuring uniform pressure distribution across the top surface of the sample. The compressive strength f_c is calculated as follows:

$$f_c={\displaystyle \frac{F}{A}}$$

(1)

where F is the maximum load and A is the bearing area of the sample.

In the splitting test, a specialized device, as shown in Figure 6b, was used to determine the split tensile strength of concrete. It comprises a loading head that applies a concentrated force at the center of the sample and two side supports that secure the concrete test piece, preventing lateral movement. The test sample is typically a cube or cylinder, with dimensions specified according to testing standards, like ASTM C496 [48]. During the test, as the loading head gradually increases the pressure, the internal stress within the sample rises until it reaches the split tensile strength. At this point, a crack forms beneath the loading head and propagates along the diameter of the sample. Usually, a thin wooden slab is placed at the place where the sample is in contact with the device, which can make the shear stress zone at the end of the sample smaller and closer to the splitting from the midline when the sample is loaded so that the test results are more accurate and stable [49,50]. The splitting tensile strength f_t is calculated as the following equation:

$$f_t={\displaystyle \frac{2F}{\pi A}}$$

(2)

where F is the maximum load and A is the splitting surface area of the sample.

The three-point bending test was carried out in accordance with ASTM C78 [51] with a three-point flexure apparatus. The sample was simply supported on two supports with a concentrated load applied at the midpoint between the supports and the span is 90 mm. The midpoint load is applied at a constant loading rate of 0.1 mm/min until the sample fails, creating a bending moment that causes the beam to deflect and is typically marked by the formation of a crack and subsequent breakage. The flexural strength f_f of concrete is calculated using the following formula:

$$f_f={\displaystyle \frac{3FL}{2b{h}^2}}$$

(3)

where F is the maximum load, L is the span length, b is the section width of the sample, and h is the section height of the sample.

3. Results and Discussion

3.1. Compressive Performance Parameters

3.1.1. Compressive Strength

The mechanical property parameters of RAC through compressive testing can be obtained as shown in

Table 5. Compressive performance parameters.

Sample	Fiber Type	Vf	fc (MPa)	cov	εc (mm)	I0	I1
P	\	\	55.19	0.039	1.19	\	\
B1.0	Basalt	1.0%	46.70	0.025	1.31	2.36	1.31
B1.5	Basalt	1.5%	48.21	0.034	1.60	2.13	1.36
B2.0	Basalt	2.0%	49.92	0.033	1.30	1.78	1.49
B2.5	Basalt	2.5%	46.62	0.027	1.53	2.67	1.51
G1.0	Glass	1.0%	55.93	0.030	1.43	2.00	1.30
G1.5	Glass	1.5%	47.68	0.058	1.17	2.06	1.44
G2.0	Glass	2.0%	48.14	0.046	1.44	2.02	1.47
G2.5	Glass	2.5%	32.43	0.029	1.41	1.90	1.44

V_f is the fiber volume fraction, f_c is the compressive strength, ε_c is the peak displacement, and I₀ and I₁ are the toughness indices.

. It can be observed that fiber type and content have an obvious influence on the mechanical properties of the recycled aggregate concrete. The plain sample (P) exhibits the highest compressive strength, suggesting that adding fibers might not always enhance compressive strength. This phenomenon was also observed and reported by previous studies [46,52,53], especially when concrete has a high fiber volume fraction. The decrease in compressive strength with increasing fiber content occurs due to weakened fiber–matrix interfaces, uneven fiber distribution, and reduced matrix density. Higher fiber volumes can create weak zones and stress concentrations, leading to microcracks under load. Additionally, excessive fibers disrupt the homogeneity of the concrete matrix, altering stress distribution and causing premature failure. While fibers enhance tensile and flexural performance, excessive use reduces the cementitious material’s contribution to compression resistance, highlighting the need to optimize fiber content for balanced mechanical properties [27,29,33].

The specific effects of fiber type and content on the compressive strength normalized to the plain sample compressive strength are illustrated in Figure 7. The compressive strength of basalt-fiber-reinforced RAC remains relatively stable, showing a slight increase before the fiber volume fraction reaches 2.0%. Then a slight decline occurs at 2.5%, likely due to fiber clustering or reduced matrix homogeneity. For glass-fiber-reinforced RAC, the compressive strength decreases consistently with an increasing fiber volume fraction. As the fiber volume fraction reaches 2.5%, the drop is particularly pronounced, indicating a significant adverse effect of excessive glass fiber content. Similar observations were reported by Ali et al. [36], who found that higher dosages (>0.5%) of glass fibers were more detrimental to the durability of concrete compared to lower dosages (<0.5%). Although the threshold identified in their study was lower, the underlying mechanism of excessive fiber content impairing matrix integrity remains consistent.

Basalt-fiber-reinforced RAC exhibits stable normalized compressive strength overall and performs better when V_f exceeds 1.5% compared to glass-fiber-reinforced samples, which demonstrates that basalt fibers are more suitable for improving mechanical properties in recycled aggregate concrete under higher fiber dosages. This suggests that basalt fibers have better compatibility with the concrete matrix and a more effective fiber–matrix interaction compared to glass fibers. However, it is worth noting that 1% glass-fiber-reinforced RAC presents a 1.34% increase in compressive strength among all samples compared with plain RAC. Combining its sharp reduction in compressive strength beyond this point, it can be claimed that the optimum usage of glass fiber in RAC is within 1%.

3.1.2. Compressive Load–Displacement Curve

Figure 8 shows the load–displacement curves of the nine types of fiber-reinforced RAC samples under uniaxial compressive loads. The strength degradation and ductility improvement with fiber addition can be seen clearly. The plain samples do not have a post-peaking phase, while the fiber-reinforced ones show obvious post-peak bearing capacity.

The typical load–displacement curve of plain concrete has only two stages, namely the compaction stage and linear elastic stage (AB), which could be simplified into two linear stages as shown in Figure 9a. The compaction stage (OA) is where pores and microcracks in concrete are compressed. In the linear elastic stage (AB), the load–displacement relationship of concrete is usually linear, that is, the displacement is proportional to the load. The slope of this stage can reflect the initial stiffness of concrete.

However, with the addition of fibers, the concrete material property is improved, and the typical constitutive curve has three more stages as shown in Figure 9b, i.e., the elastic–plastic stage (BC), the degradation stage (CD), and the convergence stage (DE). Point B is the cracking point of the concrete, where the slope of the load–displacement curve begins to decrease. At this time, the micro-cracks appear, marking the concrete’s transformation from the linear elastic stage (AB) to the elastic–plastic stage (BC). The point C where the load reaches the maximum value on the curve is called the peak point, which usually corresponds to the maximum bearing capacity of concrete. After the peak point, the curve enters the degradation stage (CD). At this time, the load begins to decrease, and the displacement continues to increase, indicating that the bearing capacity of the concrete begins to decrease. The slope of the curve at this stage becomes negative, reflecting the degradation stiffness of concrete. Until point E, where the load–displacement curves of concrete start tending to be parallel to the x-axis. In the case of large strain, the load–displacement curves of concrete with different strength grades may converge together, indicating that the broken concrete samples can still maintain a certain bearing capacity.

By observing the force–displacement curve, the failure mode of concrete can be analyzed. Brittle failure usually shows a sudden decline in the curve, while ductile failure shows a slow decline in the curve. The failure patterns of the load-bearing surface of the plain RAC and fiber-reinforced samples are depicted in Figure 10. Plain samples show a typical shear failure with an “X” crack under compressive load, and this brittle failure mode usually occurs suddenly with splashing concrete fragments. Obviously, with the addition of fibers, the load-bearing surface presents different failure patterns. The center of the fiber-reinforced specimen retains the columnar area, and the surrounding concrete is peeled off like onion layers due to the crack penetration and connection, showing ductile failure. With the further increase in fiber content, the area of the intact area expands, and the integrity of the sample is obviously improved. Kang et al. [54] also observed that the failure of basalt fiber modified recycled aggregate concrete was accompanied by fine cracks, exhibiting better overall integrity and a noticeable reduction in surface cracking. In the present study, basalt-fiber-reinforced specimens demonstrate the highest structural integrity at 2% fiber content. However, at 2.5%, fiber agglomeration led to weak zones in the concrete structure, resulting in lower post-failure integrity compared to the specimens with 1.5% fiber content. The failed glass-fiber-reinforced RAC specimens generally have better integrity than basalt-fiber-reinforced ones, showing the best structural integrity at 1.5% fiber content.

It is noteworthy that some fiber-reinforced RAC specimens emit audible sounds of fiber breakage during the post-peak stage in the compressive test. Specimens reinforced with 1.5% and 2% basalt fibers generated intense sounds immediately after reaching the peak load, with the 1.5% fiber content specimens exhibiting the highest frequency and loudest amplitude. However, the sound of glass-fiber-reinforced specimens appears after a longer delay following the peak load. These sounds were less intense, with lower frequency and amplitude compared to basalt fibers. As the glass fiber content increased, the sounds became progressively weaker and occurred only near the residual strength stage.

3.1.3. Compressive Toughness Evaluation

Concrete ductility refers to the capacity of concrete to sustain deformation after cracking. It can be quantitatively described as the region enclosed by the load–displacement curve and the x-axis. This paper proposes an advanced method for assessing toughness, considering ASTM-C1080 [55] and previous studies [56,57,58]. The peak displacement ε_p, and the λ₁ and λ₂ times of peak displacement are selected as the reference displacement, and the toughness was accordingly evaluated by the areas of A₁, A₂, and A₃, as shown in Figure 11. Points A and C in Figure 11 are generally identical to the fitting results of the cracking point and the convergence point on the load–displacement curves in Figure 6, and λ₁ = 0.7, λ₂ = 1.3. Additionally, the specimens had sustained damage and essentially forfeited their load-bearing capability under greater axial displacements. Consequently, further analysis and discussion in this context were deemed unnecessary.

In Figure 11, A₁ was the area under curve OA, A₂ was the area under curve AB, and A₃ was the area under curve BC. The compressive toughness of RAC was represented by I₀ and I₁, calculated as follows:

$$I_0={\displaystyle \frac{A_1+A_2}{A_1}}$$

(4)

$$I_1={\displaystyle \frac{A_1+A_2+A_3}{A_1+A_2}}$$

(5)

I₀ is the toughness index to evaluate the elastic–plastic deformation capacity of RAC, and I₁ is to evaluate the post-peak deformation capacity. The relationship between fiber volume content and the toughness index I₀ for RAC reinforced with basalt and glass fibers is illustrated in Figure 12. The toughness I₀ index for the glass fibers shows a more consistent but slightly declining trend as volume fraction increases from 1.0% to 2.5%, reaching the maximum value at 1.5% fiber content. While basalt-fiber-reinforced RAC samples decrease significantly as volume fraction increases to 1.0% and 2.0% and increase dramatically at 2.5%, showing a V shape. Experimental data in Table 5 indicate that the 2.5% basalt-fiber-reinforced specimen demonstrates lower strength and higher peak strain, resulting in an increased A₂, which in turn leads to a higher I₀. Nonetheless, this does not suggest that the 2.5% strain specimen has the most effective energy absorption efficiency. In fact, the results indicate the contrary; samples with 1% basalt fiber have the greatest elastic–plastic deformation capacity before failure. What can be concluded is that basalt fibers indeed have better improvement on the elastic–plastic deformation and post-peak deformation capacity of RAC under compressive loads with lower fiber content, especially before fiber volume fraction reaches 1.5%.

The toughness index I₁ quantifies the post-peaking deformation and energy absorption capacity, that is, the ability to continue bearing after failure without severe sudden deformation, directly reflecting the material’s ductility and toughness under compressive loads. Figure 13 illustrates the variation of toughness index I₁ of basalt- and glass-fiber-reinforced RAC with increasing fiber volume content. It can be seen that I₁ of basalt-fiber-reinforced samples increases consistently as fiber content rises from 1.0% to 2.5%, indicating a significant improvement in the post-peaking ductility provided by the basalt fibers. The toughness index I₁ of glass-fiber-reinforced samples increases steeply as volume fraction increases from 1.0% to 1.5%, then shows a gradual decline from 2.0% to 2.5%, suggesting a potential saturation effect or diminishing returns.

To further understand the effect of toughness indices I₀ and I₁ and provide insights to improve the ductility of RAC with optimum fiber type and content, simplified models are established. The derivation process and results are as follows:

The calculation of A₁ and A₁ in Figure 11 can be simplified as the area of triangle OAD and trapezoid ADEB. Then, I₀ in Equation (4) can be expressed as follows:

$$\begin{array}{ll}{I}_0& =1+{\displaystyle \frac{(1-{\lambda }_1){\epsilon }_p·(f_{{\lambda }_1{\epsilon }_p}+f_{{\epsilon }_p})}{{\lambda }_1{\epsilon }_p{f}_{{\lambda }_1{\epsilon }_p}}}\\ & ={\displaystyle \frac{1}{{\lambda }_1}}+\left({\displaystyle \frac{1}{{\lambda }_1}}-1\right)·{\displaystyle \frac{f_{{\epsilon }_p}}{f_{{\lambda }_1{\epsilon }_p}}}\end{array}$$

(6)

where ε_p is the peak displacement, λ₁ is the ratio of the cracking point displacement and peak displacement, $f_{{\epsilon }_p}$ is the peak compressive load or strength, and $f_{{\lambda }_1{\epsilon }_p}$ is the load or strength at the cracking point. Normally, both ${\displaystyle \frac{1}{{\lambda }_1}}$ > 1 and ${\displaystyle \frac{f_{{\epsilon }_p}}{f_{{\lambda }_1{\epsilon }_p}}}$ > 1, so that I₀ > 1. It can be found that the value of I₀ depends on the difference between the cracking load and peak load. The larger the difference, the larger the I₀ value, and the stronger the elastic–plastic deformation performance of the material.

The calculation of A₃ can be simplified as the area of trapezoid BEFC. Then, I₁ in Equation (5) can be expressed as follows:

$$I_1=1+{\displaystyle \frac{({\lambda }_2-1){\epsilon }_p·(f_{{\epsilon }_p}+f_{{\lambda }_2{\epsilon }_p})}{{\lambda }_1{\epsilon }_p{f}_{{\lambda }_1{\epsilon }_p}+(1-{\lambda }_1){\epsilon }_p·(f_{{\lambda }_1{\epsilon }_p}+f_{{\epsilon }_p})}}$$

(7)

where  is the load or strength at the convergence point, λ₂ is the ratio of the convergence point displacement and peak displacement, normally simplified as 2 − λ₁. Then I₁ can be further simplified as follows:

$$\begin{array}{l}{I}_1=1+{\displaystyle \frac{(1-{\lambda }_1)·(f_{{\epsilon }_p}+f_{{\lambda }_2{\epsilon }_p})}{(1-{\lambda }_1)·f_{{\epsilon }_p}+f_{{\lambda }_1{\epsilon }_p}}}\\ =2+{\displaystyle \frac{(1-{\lambda }_1)·f_{{\lambda }_2{\epsilon }_p}-f_{{\lambda }_1{\epsilon }_p}}{(1-{\lambda }_1)·f_{{\epsilon }_p}+f_{{\lambda }_1{\epsilon }_p}}}\end{array}$$

(8)

I₁ < 2 and the value of I₁ basically depends on the value of convergence load for a certain λ₁. A higher convergence strength results in a larger value of I₁ and leads to stronger bearing and deformation capacity after peak load, indicating better ductility of materials.

3.1.4. Summary of Compressive Performance Results

The compressive strength of recycled aggregate concrete (RAC) is significantly influenced by both the type and volume fraction of fibers. Basalt-fiber-reinforced RAC shows relatively stable compressive performance, with a slight improvement as fiber content increases up to 2.0%, followed by a minor decline at 2.5% due to fiber agglomeration and reduced homogeneity. This trend is consistent with findings by Fang et al. [27], who reported that the incorporation of BF led to a decrease in the strength of the concrete matrix, with a decreasing trend in the axial compressive strength before an increase in the axial compressive strength. Similarly, Yang et al. [29] observed improved compressive strength at moderate fiber content but noted that excessive fibers led to decreased matrix uniformity and strength. In contrast, glass-fiber-reinforced RAC exhibits a consistent decrease in compressive strength with increasing fiber content, with a marked reduction at 2.5%, highlighting the detrimental effect of excessive glass fiber. This observation aligns with Zhang et al. [33], who noted that increasing glass fiber content from 7.5 kg/m³ to 10 kg/m³ adversely affected the mechanical performance of RAC due to compromised workability and fiber dispersion. Among all samples, 1% glass-fiber-reinforced RAC achieves the highest improvement (1.34%) over plain concrete, suggesting an optimal dosage threshold.

Failure mode analysis reveals that fiber addition enhances ductility and structural integrity, with fiber-reinforced specimens showing peeling and layered cracking rather than the sudden shear failure seen in plain RAC. Basalt fibers improve the post-peak behavior and toughness most effectively at lower contents (around 1–1.5%), while higher contents compromise homogeneity. Toughness indices (I₀ and I₁) further confirm that basalt fibers, especially at 1%, enhance elastic–plastic deformation capacity more efficiently than glass fibers. Audible fiber breakage during post-peak failure also supports the stronger interaction of basalt fibers with the matrix.

3.2. Tensile Performance Parameters

3.2.1. Tensile Strength

The mechanical property parameters of RAC through the splitting test can be obtained as shown in

Table 6. Tensile performance parameters.

Sample	Fiber Type	Vf	ft (MPa)	cov	εt (mm)	Et (J)	ft/fc
P	\	\	4.62	0.027	3.06	94.64	0.08
B1.0	Basalt	1.0%	5.04	0.035	3.37	120.65	0.11
B1.5	Basalt	1.5%	5.27	0.024	3.46	119.92	0.11
B2.0	Basalt	2.0%	5.64	0.023	3.44	116.69	0.11
B2.5	Basalt	2.5%	4.55	0.037	2.74	88.77	0.10
G1.0	Glass	1.0%	6.17	0.030	3.86	132.13	0.11
G1.5	Glass	1.5%	6.45	0.048	3.88	152.03	0.13
G2.0	Glass	2.0%	5.12	0.042	3.03	103.53	0.11
G2.5	Glass	2.5%	4.44	0.039	3.29	100.75	0.14

f_t is the splitting tensile strength, ε_t is the peak axial displacement, E_t is the fracture energy, and f_t/f_c is the tension–compression ratio.

. The tensile performance parameters for basalt- and glass-fiber-reinforced RAC exhibit distinct trends with the addition and increase in fiber content. For both types of fiber-reinforced RAC, the splitting tensile strength and peak displacement show a trend of increasing first and then decreasing with the fiber content increasing from 1% to 2.5%. However, the optimum dosage of the two fibers is slightly different, RAC with 2% basalt fibers has the highest splitting tensile strength, while samples with 1.5% glass fibers have the best tensile performance, which may be related to the uniformity of fiber distribution, the interface bonding between fiber and matrix, and the interaction between fibers. It can be observed that glass fiber is more effective than basalt fiber in improving the material properties of RAC, especially at a lower fiber content.

The effects of fiber type and content on the splitting tensile strength of RAC are illustrated in Figure 14. It can be clearly observed that the splitting tensile strength of RAC increases first and then decreases with the fiber volume fraction increasing from 1.0% to 2.5%. The splitting strength of basalt-fiber-reinforced samples increases up to 22% when the fiber volume fraction reaches 2.0%, and that of glass-fiber-reinforced ones increases up to 40% when the fiber volume fraction reaches 1.5%. As the fiber volume content is within 1.0% and 1.5%, glass fiber exhibits higher improvement in splitting tensile strength than basalt fibers. This could be due to the inherent material properties of glass fibers, which might provide better reinforcement at this low volume fraction. When fiber volume fraction is at 2.0% and 2.5%, basalt-fiber-reinforced samples show a slightly higher splitting tensile strength compared to glass-fiber-reinforced ones, and the splitting tensile strength tends to be comparable as the fiber content continuously increases. It is noted that strength drops are observed in both fiber-reinforced samples when the fiber volume fraction exceeds 2.0%. The decrease in splitting tensile strength for glass fibers at 2.5% could be attributed to several factors, such as fiber agglomeration, poor dispersion, or increased stress concentration at higher fiber concentrations, which indicates that the maximum fiber content of both fibers should not exceed 2%.

Peak axial displacement is a critical parameter, which indicates the specimen’s ability to deform under loads, representing ductility and energy absorption capacity. Figure 15 illustrates the variation of peak axial displacement with fiber volume content in the splitting test for plain concrete and concrete reinforced with basalt and glass fibers. The peak axial displacement of basalt-fiber-reinforced RAC increases slightly as fiber content increases from 1.0% to 2.0% and goes down significantly at higher fiber contents (2.0–2.5%). Glass-fiber-reinforced specimens show a different trend, with peak axial displacement staying relatively stable when fiber content is between 1.0% and 1.5%, peaking at approximately 3.88 mm. Then, a steep decline follows at a 2.0% fiber content, with a slight recovery at 2.5%. The reduction in peak axial displacement for both fibers at higher fiber content can be attributed to reduced workability and fiber clustering, leading to non-uniform stress distribution and premature failure. Both types of fibers improve ductility compared to plain concrete at lower contents, especially before the volume fraction reaches 1.5%. Glass fibers present more pronounced effects, which is likely due to better fiber–matrix interaction and effective crack-bridging at lower contents.

3.2.2. Tensile Load–Displacement Curve

Through the splitting tensile test of RAC, the full tensile stress–strain curve of specimens is shown in Figure 16, and the tensile performance parameters are given in Table 6. It can be clearly seen the increases in peak load, peak displacement, maximum displacement, and fracture energy with the increasing fiber content, except for the fiber content of 2.5%. Fiber addition improved the ductility and deformation capacity before failure and changed the post-peak behavior. The extent of deformation after peak load reflects that the post-failure ductility is improved. The gradual failure crack instead of sudden cracking indicates that the crack resistance of RAC is enhanced.

The damaged specimens exhibit an axial crack running through the center of the specimens, with the crack’s width being the largest at the midpoint and gradually decreasing towards the ends. This indicates that the crack initiates from the center, and the central tensile stress of the specimen is the largest when subjected to splitting load, which aligns with the “center cracking” hypothesis of the Brazilian splitting test and verifies the effectiveness of the splitting test in this study.

It should be mentioned that two cork strips were placed on the upper and lower loading heads of the splitting device (shown in Figure 6) and the contact surfaces of the sample, which was reported to reduce the stress concentration at the loading ends and eliminate the triangular edge damage zones near the top and bottom loading points [44,59]. From the failure mode of the samples after this experiment, there is no stress concentration area, which verifies the scientific reliability of this method, enhancing the accuracy of the splitting test results.

In addition, a two-dimensional scanning of the top view of the damaged specimens was carried out through a multi-probe array ultrasonic tomography scanner produced by Beijing IST Technology Development Co., Ltd., Beijing, China (shown in Figure 17). Compared with basalt fiber, glass fiber can indeed maintain better integrity and decrease the crack after damage. Combined with the better performance of glass fiber in improving splitting strength and peak strain, it can be concluded that glass fiber can better improve the tensile properties of RAC than basalt fiber.

3.2.3. Fracture Energy

Fracture energy is a measure of the energy required to propagate a crack, representing a material’s toughness and ability to resist fracture [60,61]. In the splitting test of concrete, fracture energy represents the total energy absorbed by the specimen from the onset of loading until complete failure, including crack initiation and propagation. It is a critical parameter to compare the performance of different fiber types and volume fractions in enhancing the tensile capacity and energy absorption of RAC. The fracture energy (E_t) in this study is defined by the area under the load–displacement curve up to failure.

Figure 18 illustrates the relationship between fiber volume content and the fracture energy of RAC for basalt, glass, and plain samples. Glass fibers generally are more effective than basalt fibers in improving fracture energy when V_f < 1.5%, and the fracture energy of glass-fiber-reinforced specimens peaks at V_f = 1.5% (approximately 150 N·m) and decreases noticeably with higher fiber content. While the fracture energy of basalt-fiber-reinforced samples does not show an obvious wave before V_f reaches 2.0%, it slightly declines with further fiber addition. Both basalt and glass fibers enhance the fracture energy significantly at a low fiber content, attributed to their crack-bridging effect and energy dissipation during fiber pullout and debonding. The fracture energy decreases at a higher volume fraction. This may be due to fiber agglomeration and reduced bonding efficiency, which creates weak points and leads to stress concentration.

3.2.4. Tension–Compression Ratio

The tension–compression anisotropy of materials like concrete arises from the coupling effect of pressure and the Lode angle [62,63], and the tension–compression ratio is explicitly introduced as a mechanical characteristic parameter, which is crucial for structural design and analysis.

The variation in the tension–compression strength ratio of RAC with different fiber volume contents is presented in Figure 19, comparing basalt fibers, glass fibers, and plain samples. The tension–compression ratio of plain RAC is the lowest at 0.08, indicating poor tensile strength relative to compressive strength in the absence of fibers. The tension–compression ratio of basalt-fiber-reinforced RAC specimens remains constant at 0.11 for fiber volume contents up to 2.0%, and then the ratio decreases at 2.5%, corresponding to the reduction in tensile strength shown in Figure 14. While for glass-fiber-reinforced specimens, the tension–compression ratio peaks at 1.5% fiber content, reaching 0.13, indicating the significant improvement in tensile strength. However, the maximum tension–compression ratio at 2.5% glass fiber content cannot be considered to have the best tensile properties. The increase in the tension–compression ratio at this point is attributed to the huge decline in its compressive strength. Thus, the glass-fiber-reinforced RAC with 1.5% content still has the largest tension–compression ratio, 18% higher than the value of the basalt-fiber-reinforced specimen. In addition, the contribution of fibers to the improvement in compressive strength was less significant than their enhancement of tensile strength, which is consistent with the findings reported by Guo et al. [30].

3.2.5. Summary of Tensile Performance Results

The tensile performance of RAC is significantly affected by the fiber type and volume fraction, exhibiting a general trend of increasing and then decreasing tensile strength as fiber content rises from 1.0% to 2.5%. Optimal tensile strength is achieved at 2.0% basalt fiber content and 1.5% glass fiber content, with glass fibers demonstrating greater effectiveness at lower dosages. Specifically, glass fibers enhance the splitting tensile strength by up to 40% at 1.5% volume, while basalt fibers achieve a 22% increase at 2.0%. These differences are attributed to fiber–matrix bonding efficiency, dispersion uniformity, and crack-bridging effectiveness. Moreover, studies such as Katkhuda et al. [64] have shown that basalt fibers tend to perform better in tension at slightly higher dosages (1.5% volume fraction), which supports the observed trend in our results.

Both fiber types improve ductility and the deformation capacity, as reflected by increased peak axial displacement and fracture energy, particularly at lower fiber contents. This enhancement has been similarly reported by Zhang et al. [33], who noted substantial gains in fracture energy and ductility in RAC with glass fiber additions below 10 kg/m³. Glass fibers show superior enhancement of peak displacement and fracture energy when V_f < 1.5%, but performance declines at higher contents due to clustering and non-uniform stress distribution. Post-peak behavior analysis shows that fiber reinforcement delays crack propagation and promotes gradual failure, indicating improved toughness. These improvements in post-peak performance are supported by prior work [65], which emphasized the role of fiber–matrix interactions in promoting energy absorption and strain redistribution in fiber-reinforced concrete.

The tension–compression strength ratio further confirms the enhanced tensile behavior, especially for glass fibers. At 1.5% fiber content, glass-fiber-reinforced RAC reaches a maximum ratio of 0.13, representing an 18% improvement over basalt-fiber-reinforced specimens. However, this value declines at 2.5% due to a sharp reduction in compressive strength. Comparable trends have been noted in experimental studies by Khudhair et al. [66], where optimal glass fiber dosages improved the tension-to-compression ratio, but excessive additions led to reduced mechanical uniformity. Overall, fiber reinforcement (especially glass fiber reinforcement) was found to be more beneficial in improving tensile properties than compressive strength, which supports similar conclusions drawn from previous studies.

3.3. Bending Performance Parameters

3.3.1. Bending Strength

The mechanical properties parameters of RAC through the three-point bending test can be obtained as shown in

Table 7. Bending performance parameters.

Sample	Fiber Type	Vf	ff (MPa)	cov	df (mm)	Tf (J)	ff/fc
P	\	\	9.26	0.020	0.48	0.64	0.17
B1.0	Basalt	1.0%	10.37	0.025	0.41	0.67	0.22
B1.5	Basalt	1.5%	9.43	0.030	0.43	0.65	0.20
B2.0	Basalt	2.0%	9.17	0.027	0.42	0.61	0.18
B2.5	Basalt	2.5%	8.22	0.038	0.50	0.61	0.18
G1.0	Glass	1.0%	10.64	0.021	0.43	0.84	0.19
G1.5	Glass	1.5%	10.16	0.026	0.48	0.91	0.21
G2.0	Glass	2.0%	10.11	0.034	0.48	0.74	0.21
G2.5	Glass	2.5%	4.93	0.045	0.53	0.54	0.15

f_f is the flexural strength, d_f is the peak deformation, T_f is the bending toughness index, and f_f / f_c is the bending–compression ratio.

. The bending strength of both fiber-reinforced specimens decreases with the increase in fiber content and achieves the peak value at 1% for basalt fiber and 1.5% for glass fiber. The toughness index T_f of the basalt-fiber-reinforced samples shows a consistently decreasing trend with the increasing fiber content, while the T_f of the glass-fiber-reinforced samples increases first and then decreases. It can be observed that the bending–compression ratio of the fiber-reinforced RAC samples is basically higher than that of the plain samples, indicating that the addition of fiber is capable of improving the flexural performance of RAC material, especially with a low fiber volume fraction.

As shown in Figure 20, the effect of fiber type and content on the bending strength of RAC can be seen clearly. At a 1% fiber volume fraction, basalt- and glass-fiber-reinforced samples present increases of bending strength up to 12% and 15%, respectively, compared to plain samples. Then, basalt-fiber-reinforced samples gradually decrease with increasing fiber volume fraction, while glass-fiber-reinforced samples stabilize between a 1.5 and 2.0% volume fraction but plummet dramatically when the volume fraction reaches 2.5%, showing the most severe reduction. Glass fibers outperform basalt fibers before the volume fraction reaches 2.0% and show higher bending strength, while the stability of basalt fibers still performs as well as in the compression test.

3.3.2. Bending Load–Deflection Curve

In Figure 21, the peak deflection of both types of fiber-reinforced RAC samples shows similar trends with increasing fiber content. The peak deflection reaches the smallest values at 1% fiber content and then increases with fiber content increase and achieves a significant improvement at 2.5% fiber volume content. But considering the huge bending strength degradation shown in Figure 20, it suggests that the ductility is increased so much that the RAC becomes softer when the fiber content exceeds 2%. This phenomenon is also illustrated in the load–deflection curves, as shown in Figure 22, especially the curves of B2.5 and G2.5. This indicates that excessive fiber addition is turning brittle materials such as concrete and rock into plastic materials like low-carbon steel and copper, which is obviously not consistent with improving the mechanical properties of RAC material. Thus, it can be claimed that the maximum fiber addition should be less than 2%, ensuring no bending performance degradation. And the optimum fiber content is between 1% and 1.5%.

It is worth noting that when the fiber volume fraction is at 1%, the bending strength of basalt- and glass-fiber-reinforced samples increases by 12% and 15% compared to the plain sample. Generally, fiber addition can improve the bending strength of concrete materials by bridging cracks in the concrete matrix, delaying crack propagation, and improving load-bearing capacity under bending [67]. However, as shown in Figure 21, the peak deflection of fiber-reinforced RAC at a 1% volume fraction is smaller than that of the plain RAC, which is different from the reported results of steel-fiber-reinforced concrete [68,69]. The addition of fibers to concrete can lead to an increase in bending strength but a decrease in ductility due to the following reasons:

• Bending Strength Increase: the crack-bridging effect of basalt and glass leads to higher flexural strength as fibers effectively resist tensile stresses at the microcrack level, and fibers distribute stress more evenly across the composite, enhancing the resistance of concrete to bending forces.
• Ductility Decrease: (1) Stiffness of Fibers: Glass and basalt fibers are relatively stiff. As shown in Figure 22, they increase strength by resisting crack propagation but decrease the ductility by enhancing the flexural modulus of fiber-reinforced RAC. The slope of the P sample (λ_P) as the deflection between 0.35 and 0.45 mm is 19.5, while the slopes of B1.0 (λ_B1.0) and G1.0 (λ_G1.0), as the deflection between 0.30 and 0.40 mm, are 21.2 and 19.8, respectively. They limit the ability of the concrete matrix to deform under load, leading to reduced ductility, especially shown in the load–deflection curve of B1.0 and G1.0. (2) Matrix Modification: the fiber three-dimensional network structure reduces the free movement of aggregates and the concrete matrix, making the composite less flexible, which is reflected by the decreased deflection of concrete under a 1 KN axial load in the curves.

At a moderate fiber content (e.g., 1.0–1.5%), the balance between bending strength and ductility is typically optimized as fibers enhance strength without overly restricting matrix deformation. However, at higher dosages, excessive fibers may cause clustering or reduce matrix integrity, destroying bending performance.

The failure patterns of samples in the bending test are shown in Figure 23, and the same device as the splitting test was used in scanning failure surfaces. The widest portion of the crack is located at the bottom of the specimen, and the width of the crack progressively decreases as the crack propagates upwards towards the loading point, which is in line with the stress distribution on the fracture plane. With the increased fiber content, no obvious change is found in the crack width of samples reinforced with glass fibers, while an increasing trend in crack width is observed on basalt-fiber-reinforced samples. Furthermore, compared with the basalt-fiber-reinforced RAC specimen, the crack width of the glass-fiber-reinforced specimen is larger in general, which is also consistent with the results presented in Figure 21, showing the superior improvement in deformation performance and toughness of glass fiber to avoid sudden brittle failure.

3.3.3. Bending Toughness Evaluation

The bending toughness of concrete refers to the material’s ability to absorb energy and undergo deformation beyond the elastic limit without fracturing. It is a critical property for evaluating the performance of concrete in flexural applications, such as beams and slabs, where the material is subjected to bending stresses. Bending toughness is directly related to a concrete structure’s resistance to cracking under service loads. A material with high bending toughness can better resist crack propagation, which is essential for maintaining the structural integrity and durability of concrete elements. Various standards and methods have been established to evaluate the flexural toughness of steel-fiber-reinforced concrete, including the energy method and strength method [70,71].

In the present study, the energy method is adapted to assess the bending toughness by calculating the area under the load–deflection curve, and the relationship between the fiber volume fraction and the bending toughness of RAC is illustrated in Figure 24. With the inclusion of basalt fibers, the toughness of RAC samples decreases gradually with increasing fiber content, and the peak value is only 5% higher than the plain samples, showing limited enhancing ability of basalt fibers. Energy absorption of glass-fiber-reinforced RAC increases first and then decreases with further fiber addition, and the peak value is 42% higher than plain ones, demonstrating a superior improvement in flexural toughness compared to basalt fiber.

3.3.4. Bending–Compression Ratio

The bending–compression ratio is defined as the ratio of the flexural strength to the compressive strength of the material. For fiber-reinforced concrete, this ratio is influenced by the presence of fibers, which can significantly enhance the material’s performance in bending compared to plain concrete [72]. Figure 25 depicts the effect of fiber volume fraction on the bending–compression strength ratio for RAC reinforced with basalt and glass fibers, compared to plain RAC. It can be observed that the fiber-reinforced samples generally have higher values than the plain ones except for a 2.5% fiber volume fraction, due to the catastrophic degradation of bending strength. Before fiber content reaches 2.0%, the ratio of glass-fiber-reinforced samples demonstrates a steady increase with the peak ratio of 0.21 at 1.5% fiber content. While the samples with basalt fiber have a maximum ratio of 0.22, the ratio declines evenly with increasing fiber content. All in all, glass fibers are more effective in enhancing the flexural strength of RAC by bridging cracks, redistributing stresses, and improving energy absorption, despite the negligible maximum value difference with basalt fibers.

Considering the overall bending performance and expenses, glass fibers provide effective enhancements in the bending properties, making them suitable for applications requiring balanced flexural and compressive performance. Basalt fibers are more suitable for applications emphasizing cost-efficiency and lower fiber volumes.

3.3.5. Flexural Strength Prediction Model

The flexural performance of fiber-reinforced concrete is jointly affected by the fiber–matrix synergistic effect and the interfacial deterioration mechanism [73]. Several predictive models for the flexural strength of fiber-reinforced concrete have been proposed in previous studies, incorporating factors such as fiber type, content, and matrix properties to enhance accuracy [74,75]. Based on the multi-scale mechanical theory, this paper proposes an intrinsic model that can distinguish fiber types and quantify the competition mechanism between enhancement and deterioration, which provides a theoretical basis for the design of mixed-fiber concrete.

The model is based on several basic assumptions:

(1)

The concrete matrix is assumed to be an isotropic linear elastic material;

(2)

It is assumed that at higher fiber volume fractions, the strength of the matrix decays due to agglomeration of fibers with increased porosity;

(3)

The strengthening and deterioration mechanisms of different fiber types are described by independent parameters.

Experimental results demonstrate that within the volume fraction range of 1% to 2.5%, an increase in the fine fiber content used in this study leads to a decline in the flexural strength of the specimens. This indicates that within this range, a higher fiber dosage weakens the flexural performance of the concrete. Accordingly, the proposed predictive model for overall flexural strength consists of a matrix degradation term, a fiber reinforcement term, and a deterioration term, as expressed in Equation (9).

$${\sigma }_{FRC}={{\sigma }^{\prime }}_p+{\eta }_i\Delta {\sigma }_i-{\eta }_i\Delta {\sigma }_d$$

(9)

where ${\sigma }_{FRC}$ is the flexural strength of fiber-reinforced concrete, ${{\sigma }^{\prime }}_P$ is the decayed matrix strength, ${\eta }_i$ is the fiber type factor, $\Delta {\sigma }_i$ is the additional strength provided by fibers, $\Delta {\sigma }_d$ is the loss of strength induced by fiber agglomeration and interfacial slip, and $i$ refers to the type of fiber.

At a high fiber dosage, matrix compactness decreases, and the decayed flexural strength can be expressed by Equation (10).

$${{\sigma }^{\prime }}_P={\sigma }_P(1-\beta V_f)$$

(10)

where ${\sigma }_P$ is the flexural strength of plain concrete, $\beta$ is the decline degree parameter of the matrix caused by porosity increment, and $V_f$ is the fiber volume fraction.

Fibers provide additional strength by bridging cracks, and their contribution is first enhanced and then saturated with the increase in $V_f$, which can be expressed by Equation (11).

$$\Delta {\sigma }_i=k_{1i}{V}_f{e}^{-{\lambda }_i{V}_f}$$

(11)

where $k_{1i}$ is the reinforcement index of fiber and ${\lambda }_i$ is the attenuation coefficient of the dispersion efficiency of fiber.

With the increase in the fiber volume fraction, fiber agglomeration and interfacial bonding failure between fiber and matrix due to uneven fiber distribution will lead to a decrease in the flexural strength of the sample. A quadratic term form shown in Equation (12) can be used to capture the synergistic amplification effect of defects at high volume additions.

$$\Delta {\sigma }_d=k_{2i}{V}_f^2$$

(12)

where $k_{2i}$ is the deterioration index of fiber.

Thus, the prediction formula for the flexural strength of fiber-reinforced concrete can be expressed as follows:

$${\sigma }_{FRC}={\sigma }_P(1-\beta V_f)+{\eta }_i[k_{1i}{V}_f{e}^{-{\lambda }_i{V}_f}]-{\eta }_i[k_{2i}{V}_f^2]$$

(13)

Based on the experimental data, the values of the coefficients in Equation (13) were obtained by nonlinear least squares fitting, as shown in

Table 8. Fitting results of the flexural strength prediction formula.

Fiber Type	$\mathit{\beta }$	${\mathit{k}}_{1\mathit{i}}$ (MPa/%)	${\mathit{k}}_{2\mathit{i}}$ (MPa/%)	${\mathit{\lambda }}_{\mathit{i}}$ (%−1)
Basalt	0.00	18.70	1.16	4.48
Glass		10.70	0.00	14.33

.

Based on the fitting results, $\beta =0$ indicates that the matrix strength remains unaffected by the incorporation of fibers, suggesting that the addition of fibers does not significantly increase porosity or compromise the matrix’s density. The condition $k_{1B}>k_{1G}$ suggests that basalt fibers exhibit a stronger reinforcement effect, with their volume fraction leading to a better enhancement of the matrix’s bending strength. On the other hand, $k_{2B}>k_{2G}$ indicates that the overall degradation effect of glass fibers is weaker. Experimental data show that when the fiber volume content ranges from 1% to 2%, the strength loss caused by increasing glass fiber content is relatively small. The value of ${\lambda }_B$ being much smaller than ${\lambda }_G$ implies that the reinforcement efficiency of basalt fibers decreases more slowly as their volume content increases. This might be due to a more uniform dispersion of basalt fibers in the matrix or superior interfacial mechanical coupling, allowing them to effectively bridge cracks at higher content levels, thereby minimizing the matrix strength degradation.

A comparison between the experimental results and the predicted values from the model is presented in Figure 26. The model predictions exhibit good agreement with the experimental data, with coefficients of determination R_B² = 0.96 and R_G² = 0.86. Notably, the model provides more accurate predictions for the flexural strength of basalt-fiber-reinforced concrete. Therefore, it can serve as a reference method to guide the design of flexural structures incorporating this type of fiber-reinforced concrete.

3.3.6. Summary of Flexural Performance Results

The flexural performance of RAC is influenced significantly by fiber type and dosage, exhibiting peak bending strength at 1.0% basalt fiber and 1.5% glass fiber content. Both fiber types improve flexural strength at low volume fractions due to enhanced crack-bridging and stress distribution effects. Similar findings were reported by Wang et al. [76], who found that moderate fiber contents improve the load transfer and flexural strength of RAC by bridging microcracks and delaying crack propagation. Glass fiber reinforcement shows superior flexural strength enhancement before the fiber content reaches 2.0%, achieving a maximum increase of 15%, slightly higher than the 12% gain from basalt fibers. However, at 2.5% volume content, both fibers lead to a sharp reduction in bending strength, highlighting the adverse effects of excessive fiber addition, such as clustering and decreased matrix integrity. These negative effects at high dosages are in agreement with previous research [44], which attributed the decline to poor fiber dispersion and void formation.

While bending strength increases at low fiber contents, ductility shows a complex trend. At 1.0% fiber content, peak deflection is reduced compared to plain concrete, indicating increased stiffness due to fiber inclusion. As fiber volume increases to 2.5%, deflection increases significantly, suggesting that over-reinforcement softens the RAC matrix, reducing structural rigidity and turning brittle failure into plastic deformation. This transformation is undesirable for structural applications and highlights that optimal fiber content should remain below 2.0%.

Bending toughness results, evaluated through energy absorption under the load–deflection curve, show that glass fibers provide greater toughness improvements, with a 42% increase compared to plain RAC, while basalt fibers offer more limited enhancements. This observation is supported by the findings of Jan et al. [77], who emphasized that increased glass fiber content improved toughness indices and energy absorption under flexural loading, with 4% fiber content showing the best performance. Similarly, the bending–compression strength ratio improves with fiber addition and peaks at 1.5% content for glass fibers and 1.0% for basalt fibers but declines at higher contents due to strength deterioration.

Overall, glass fibers exhibit better performance in enhancing the bending behavior of RAC, particularly in terms of strength, energy absorption, and ductility control at moderate contents. However, basalt fibers offer more stable performance across a broader range of contents and are more cost-effective.

4. Conclusions

In this study, a series of compressive, splitting, and three-point bending tests were conducted to examine the effects of fiber type and content on the mechanical performance of recycled aggregate concrete (RAC). The strength, energy absorption, toughness, and failure patterns of glass and basalt-fiber-reinforced RAC under different loading conditions were analyzed with volume fractions ranging from 1.0% to 2.5%. Based on the test results, the following conclusions can be drawn:

(1)

RAC with 1.5% glass fiber exhibits the best tensile and flexural performance, though it slightly reduces the compressive strength. In contrast, 1% glass fiber offers slightly lower tensile and flexural performance but marginally improves compressive strength. Therefore, the optimal glass fiber content ranges between 1% and 1.5%.

(2)

The compressive strength of RAC is reduced by the addition of basalt fibers. At a fiber content of 2%, specimens achieve the best compressive and tensile performance, while obtaining the highest flexural performance at 1%. Thus, the dosage of basalt fibers should be determined based on the specific performance requirements for the intended application.

(3)

Both types of fibers can improve the tensile and flexural performance of RAC, especially glass fibers, but a barely positive effect is observed on the compressive performance. In addition, improved models for calculating compressive toughness and flexural strength are proposed, which more accurately reflect the relationship between toughness, crack resistance, and load-bearing performance, demonstrating good accuracy and practical applicability for flexural structural design.

(4)

Fiber addition alters failure modes, improving ductility and preventing brittle failure. Glass fibers exhibit better bonding but tend to agglomerate at high contents, whereas basalt fibers are more uniformly distributed but less effective in crack resistance and energy absorption due to their lower strength and modulus, resulting in weaker bridging effects.

(5)

This study mainly focuses on mechanical performance under static loading. Dynamic performance, long-term durability, and hybrid fiber combinations remain to be explored. Moreover, the study did not include a natural aggregate concrete (NAC) control group, which limits the ability to directly quantify the performance differences introduced by using recycled aggregates. Investigations into fiber dispersion control and environmental influences on fiber–RAC interaction are also recommended to further optimize FR-RAC performance for practical applications.