Mechanical Behavior of Self-Compacting Concrete Incorporating Rubber and Recycled Aggregates for Non-Structural Applications: Optimization Using Response Surface Methodology

Abstract

The accumulation of end-of-life tires and the rapid increase in demolition activities pose significant environmental and waste-management challenges. The redevelopment of construction materials incorporating this waste is a potentially promising strategy for minimizing environmental impact while promoting the principles of a circular economy. This study investigates the performance of self-compacting concrete (SCC) incorporating up to 20% rubber aggregates (sand and gravel) and 40% recycled concrete aggregate (RCA) for non-structural applications. A series of tests was conducted to assess fresh and hardened properties, including flowability, compressive strength, tensile strength, flexural strength, water absorption, and density. The results indicated that increasing RCA content reduced density and compressive strength, while tensile and flexural strengths were only moderately affected. Response surface methodology (RSM), utilizing a Box–Behnken design, was employed to optimize compressive, tensile, and flexural strength responses. Statistical analysis was used to identify the optimal mix proportions, which balance the mechanical performance and sustainability of SCC with recycled components. Mixtures incorporating moderate rubber content—specifically, 5–5.5% sand rubber and 0–6% coarse rubber—and 40% recycled-concrete aggregate (RCA) achieved the highest predicted performance, with compressive strength ranging from 20.00 to 28.26 MPa, tensile strength from 2.16 to 2.85 MPa, and flexural strength reaching 5.81 MPa, making them suitable for sidewalks and walkways. Conversely, mixtures containing higher rubber proportions (5.5–20% sand rubber and 20% coarse rubber) combined with the same RCA level (40%) showed the lowest mechanical performance, with compressive strength between 5.2 and 10.08 MPa, tensile strength of 1.05–1.41 MPa, and flexural strength from 2.18 to 3.54 MPa. These findings underscore the broad performance range achievable through targeted optimization. They confirm the viability of recycled materials for producing environmentally friendly SCC in non-structural applications.

1. Introduction

Rapid urbanization and population growth have resulted in a huge increase in raw material consumption, which has led to the accumulation of large amounts of solid and industrial waste, including used rubber tires. They are dangerous for the environment because they are not biodegradable [1,2]. Different types of studies have been conducted to make use of this waste in the concrete industry by partially replacing natural aggregate with recycled rubber. These studies seek to produce sustainable environmental and economic solutions by incorporating used tires in self-compacting concrete (SCC) [1,2,3].

Steel wire, synthetic and natural rubber, and other chemical agents make used tires more complex to dispose of. A set of investigations highlights the importance of recycling [3]. Low specific gravity, good absorption, and reduced solid waste are among the most important benefits of using recycled rubber in concrete [2,4]. However, this application is not devoid of fundamental drawbacks. The most noticeable one is poor compressive strength resulting from the weak transition zone and high porosity. This necessitates pre-treating the rubber or blending it with enhancing materials in order to increase its internal cohesion.

The significant accumulation of construction and demolition waste, particularly recycled-concrete aggregate (RCA), is considered a serious environmental challenge to the construction sector. As part of the drive towards sustainability, this aggregate has been used as a partial or full substitute for natural aggregate to reduce the use of non-renewable natural resources [5,6]. These projects are part of the circular economy of reusing materials and lowering the financial and environmental costs of producing concrete in the old-fashioned way [5,7,8].

In this regard, self-compacting concrete has drawn particular attention because it can integrate recycled materials, namely RCA, into its structure without the use of traditional compaction systems. It makes it possible to use waste materials more effectively and sustainably [6,8,9,10]. To guarantee the intended performance, careful mix design and water-to-cement ratio optimization are also essential [5,8].

However, the combined use of rubber and recycled-concrete aggregates in a single mix introduces complex interactions affecting both the fresh and hardened properties of concrete, including workability, compressive strength, and tensile strength. Although a decline in mechanical performance is often observed when both materials are used simultaneously, such composite mixtures remain promising for various non-structural applications. These include pavements, noise barriers, curb stones, sidewalks, and lightweight filling materials, where sustainability and impact resistance are prioritized over high load-bearing capacity [11,12].

In light of the heterogeneous nature of the mixture, response surface methodology (RSM) will be utilized to optimize the incorporation of two distinct types of waste materials with appropriate proportions into a key construction material—self-compacting concrete. This approach aims to enhance both the fresh and hardened properties of the concrete, with a particular focus on improving compressive strength and overall structural performance.

2. Background

Many studies have examined the incorporation of recycled materials into self-compacting concrete (SCC) and their effects on both mechanical and fresh properties. Guneyisi partially replaced fine aggregates with rubber at 5%, 15%, and 25% by volume. They noticed that a higher rubber content led to a reduction in slump flow diameter and passing ability, as well as increased T50 time and V-funnel time [13]. Mishra et al. replaced natural coarse aggregates with rubber chips at 5%, 10%, 15%, and 20% in SCC using two rubber sizes (5 mm and 10 mm), and reported that increasing the rubber content enhanced the filling and passing abilities, with slump flow increasing from 570 mm to 730 mm [14]. A similar trend was reported by Lv et al., who used 10% to 50% rubber replacement of fine aggregates in self-compacting rubberized lightweight concrete (SCRLC) mixes. They observed a consistent decrease in flowability with increased rubber content, as evidenced by reduced slump flow and higher V-funnel times, along with improved segregation resistance [15]. Sobuz et al. also found that increasing the content of waste tire rubber aggregates (WRTA) at 5%, 10%, and 20% resulted in reduced slump flow and increased T500 and V-funnel times, confirming a decline in both workability and passing ability [4].

In summary, numerous experimental studies of self-compacting rubberized concrete (SCRC) have documented that partially replacing natural aggregate with recycled rubber generally leads to reduced workability—particularly in terms of flowability, passability, and filling capacity. This negative impact becomes more pronounced with higher rubber content, even when superplasticizers or other additives are used to counterbalance the effects. The primary reason for this decline in fresh concrete performance lies in the inherent characteristics of rubber, being lightweight, flexible, and hydrophobic, which increase internal friction and reversely reduce cohesion within the mix. These effects have been consistently remarked across multiple experimental investigations [16,17,18,19].

As far as the mechanical properties are concerned, Zheng et al. replaced fine aggregates with rubber in various proportions and noted a 30–50% reduction in compressive strength at rubber contents above 15% [20]. Bušić et al. reported that substituting up to 30% of fine aggregates with rubber in SCC caused a significant decrease in compressive strength and flexural strength, despite the inclusion of silica fume. However, mixes with 15% rubber and 5% silica fume still achieved strengths above 30 MPa [21]. Mallek et al. found that increasing rubber content from 0% to 15% led to an approximate 38% reduction in compressive strength, although the values remained above 25 MPa, and the rubberized mixes exhibited more ductile behavior [22,23]. Similarly, El Marzak et al. reported that higher rubber contents reduced compressive and tensile strengths as well as modulus of elasticity. Nevertheless, acceptable mechanical performance was maintained at the replacement levels of 20% fine, 25% coarse, and 20% combined rubber [24].

In summary, increasing rubber content generally results in a decrease in mechanical resistance. This decrease is primarily due to poor adhesion between rubber particles and the cement paste compared to the suitable adhesion between natural aggregates and the paste, in addition to the lower stiffness of rubber relative to conventional aggregate, which reduces the mixture’s load-transfer capacity. The hydrophobic nature of the rubber surface also hinders the development of the interfacial transition zone (ITZ), increasing voids and porosity within the concrete structure and contributing to a reduction in the overall mechanical performance [21,25,26,27].

Several studies have examined the use of RCA in SCC. They have shown that RCA tends to increase water absorption and reduce density due to surface impurities. However, these effects can be partially mitigated by adding supplementary cementitious materials or applying pre-soaking techniques [25,26]. Khalaf and DeVenny observed a linear reduction in compressive strength with increasing RCA content. Nonetheless, they confirmed its applicability in non-structural elements, such as pavements and retaining walls [28]. Vivian et al. demonstrated that using a two-stage mixing approach improved the interfacial transition zone and enhanced compressive strength by 10–15%, accordingly increasing the viability of RCA concrete for broader applications [29].

In general, studies indicate that the effect of RCA in SCC is related to the presence of an old mortar layer on the aggregate surface, which contributes to increased porosity, water absorption, and reduced density. Furthermore, the transition zone between this type of aggregate and the cement paste is weaker compared to natural aggregate, leading to a decrease in compressive strength. However, some studies have shown that improving mixing methods, such as adopting a two-stage mixing approach, helps to achieve better paste distribution around the aggregate and enhances microstructure cohesion, positively impacting mechanical properties [25,26,27,30].

Few studies have investigated the simultaneous incorporation of RCA and crushed rubber into concrete mixes. In the work of Guo et al., 0, 25, 50, 75, and 100% recycled raw materials were used in the mix. Although the inclusion of these components resulted in reduced compressive strength and stiffness—likely due to increased porosity—the mixtures demonstrated notable improvements in ductility, crack resistance, and energy absorption. This reveals their suitability for non-structural applications or those subject to dynamic loading [11]. Aslani et al. evaluated SCC containing both RCA and crumb rubber, reporting reductions in workability and strength with higher replacement levels, yet confirming acceptable performance at moderate dosages [12]. Saberian et al. found that incorporating approximately 1% rubber into RCA mixes enhanced strength and reduced porosity [27]. In a subsequent study, they reported improvements in cohesion and shear strength, particularly with coarse rubber; however, performance declined when rubber content exceeded 0.5%. Their predictive model supported the use of RCA in pavement base and subbase layers [28].

Although some previous studies have addressed the combined use of recycled aggregate and rubber into concrete, they neither employed a systematic design of experiments (DOE) to evaluate the effects of the various factors nor determined the suitable optimal replacement levels. Moreover, the replacement percentages were often imprecisely defined, with most findings based on direct experimental observations rather than on statistical analysis tools such as ANOVA or response surface modeling [31].

Accordingly, this study aims to fill this gap by adopting a structured experimental design, which enables the determination of optimal mechanical performance, particularly compressive and tensile strengths, as crucial responses. This will be achieved through the identification of the corresponding ratios balancing the mechanical performance and the environmental sustainability in non-structural applications [11,12,27,30].

This study investigates SCC incorporating both fine and coarse rubber with RCA, using only local materials and a superplasticizer. Response surface methodology was adopted to evaluate fresh and mechanical properties. The results were statistically analyzed using STATISTICA 12 (StatSoft, Inc., 1984–2014) software to highlight the effective and combined effects of rubber and RCA, aiming to assess their suitability for non-structural applications.

This paper is structured as follows: it is initiated by a detailed description of the materials used and the experimental procedures. This is followed by a comprehensive discussion of the results. Based on the data presented in

Table 1. Recommended performance for non-structural applications.

Applications	Compressive Strength (MPa)	Tensile Strength (MPa)	References
Road sub-base or fill material: low-strength concrete or composites used as sub-base layers beneath roads or pavements	3–10	<1	[32,33,34,35,36]
Low-strength concrete basins (landscaping): decorative ponds, garden basins, or shallow water features, utility bedding or backfill for drainage systems	7–10	<1	[37,38,39]
Playground flooring	15–20	1.5–2	[40,41]
Non-structural slabs	10–20	1–2	[42,43,44,45]
Light industrial floors	20–25	2–2.5	[46,47,48]
Sidewalks and walkways	20–30	2–3	[49,50,51]

, the studied mixtures exhibit compressive and tensile strength levels, which qualify them for a wide range of non-structural applications. Their areas of use vary according to the extent to which the achieved values conform to the performance requirements specified in Table 1.

3. Experimental Study

This section includes a detailed explanation of the materials used, their physical and chemical properties, mixing ratios, experimental conditions, and implemented tests. In order to ensure homogeneity within the mixes, an initial dry mixing of the solid components (rubber and RCA) was performed for two minutes using a horizontal mixer, which helped achieve even distribution. It also reduced clumping resulting from density differences. Subsequently, cement and water were gradually added over an extended mixing time to ensure the stability of the lighter rubber particles and prevent them from floating. The mixture was visually inspected before pouring to guarantee uniformity, and the RCA was pre-screened to achieve the appropriate grain gradation as well as reduce uneven distribution.

During the pouring process, additional measures were taken to reduce any expected internal unevenness in particle distribution within the molds. The mixes were poured continuously in one direction without vibration. This was done in accordance with self-compacting concrete standards. A consistent pouring rate was maintained to prevent heavier material settlement or rubber separation. The process was visually monitored to maintain consistent particle distribution before curing. All these steps contributed to keeping the homogeneity of the mixtures and ensuring the accuracy of the experimental results.

The mixes were tested immediately after mixing so as to ensure they met the requirements for self-compacting concrete. After pouring, the samples were stored in curing tanks for 28 days. The samples were then evaluated for compressive, tensile, and flexural strengths in addition to density and water absorption.

3.1. Materials

3.1.1. Portland Cement

Portland cement type CEM I 42.5, as specified by European standard EN 197-1 [52], was used.

Table 2. Chemical composition of Portland cement CEM I 42.5N.

Chemical Composition	Percentage (%)
CaO	62.08
SiO2	19.74
MgO	1.14
Al2O3	4.1
SO3	2.95
Fe2O3	3.4
Na2O	0.30
K2O	0.41
Na2Oeq	0.54
Insoluble residue	0.25
Loss of ignition	2.19

presents its chemical composition along with the estimated proportions of its main compounds.

3.1.2. Crushed Coarse Aggregate

Two types of crushed coarse aggregate (CCA) were used: C1, with a nominal size of 9.5–12.5 mm and a dry specific gravity of 2720, and C2, with a nominal size of 4.75–9.5 mm, a dry specific gravity of 2720, and an absorption capacity of 0.9%. The gradation curves of both C1 and C2 are shown in Figure 1.

3.1.3. Washed River Sand

Washed river sand with a nominal maximum size of 4.75 mm, a dry specific gravity of 2.620, and an absorption capacity of 0.9% was used. Its gradation curve is shown in Figure 1.

3.1.4. Rubber Aggregate

Two categories of rubber aggregate were utilized (Figure 2): a fine type (SR) with particle sizes ranging from 0 to 4.75 mm, and a coarse type, with CR ranging from 4.75 to 9.5 mm. Their particle size distribution curves are shown in Figure 1. The rubber has a specific gravity of 1.18. It does not exhibit any moisture absorption. The elemental composition of the rubber particles used in the SCC mixes is presented in

Table 3. Elemental Composition of Rubber Particles Used in SCC Mixes.

Element	Percentage (%)
C	71
${\mathrm{C}}_{18}{\mathrm{H}}_{36}{\mathrm{O}}_2$	0.3
Zn	1
O	5
H	6
S	1.5
N	0.6
FE	14.5
X	0.1

.

3.1.5. Recycled-Concrete Aggregate

After being extracted from the demolition of an old structure, recycled-concrete aggregate (RCA) was processed to produce particle sizes between 9.5 and 12.5 mm and a dry specific gravity of 2.37 (Figure 3). Figure 1 also displays the particle size distribution.

3.1.6. High Performance Concrete Softener

The superplasticizer (SP) used in this study is a high-performance admixture based on polycarboxylate technology. It has a specific gravity of 1.05 at 25 °C and contains no chloride, fully complying with BS EN 934-2 [53]. This material was used in all SCC mixtures to maintain the required flowability, especially given the addition of rubber and RCA, which was mainly oriented to reduce workability due to the irregular surface texture of the rubber and the high absorption capacity of the RCA. Preliminary tests were conducted for all the mixtures to establish a baseline dosage range of superplasticizer necessary to achieve a slump flow between 650 and 750 mm.

3.1.7. Ground Calcium Carbonate

Ground calcium carbonate is essentially made of finely ground calcium carbonate, with a particle size of less than 10 μm and a specific gravity of 2.71.

3.2. Experimental Design

The experimental design in this study was conducted using response surface methodology (RSM) with a Box–Behnken design (BBD) to investigate the effects of different combinations of the three independent variables (factors), as listed in

Table 4. Real values and corresponding coded levels of the studied factors.

Level Factor	Low Level (−1)	Center Level (0)	High Level (+1)
Sand rubber content (%)—A	0	10	20
Coarse rubber content (%)—B	0	10	20
Recycled-concrete content (%)—C	0	20	40

—sand rubber content (denoted A, ranging from 0 to 20%), coarse rubber content (B, 0–20%), and recycled-concrete content (C, 0–40%)—on the mechanical properties of newly fabricated concrete, specifically its compressive, tensile, and flexural strength. The three factors selected for this experimental design were sand rubber content, coarse rubber content, and recycled-concrete aggregate content. They were chosen because of their significant influence on the rheological properties and the mechanical behavior of SCC. In fact, the fine rubber particles can significantly affect the workability of SCC. In contrast, the coarse rubber aggregates principally have a more pronounced effect on mechanical performance. Furthermore, RCA content influences strength, coupled with absorption and interfacial transition zones. For all factors, they are a common, alternative aggregate that partially or fully replaces natural aggregate.

Each factor was evaluated at three levels—low, center, and high—determined based on pre-experimental research, which reported that exceeding these thresholds often leads to a sharp degradation in strength and workability. These limits also provide a sufficient range for the response surface methodology to effectively model potential nonlinear interactions.

All mixes were designed using a cement content of 370 kg/m³ and a water-to-cement ratio of 0.5.

Table 5. Mixtures formulation designed using Box–Behnken design.

Run	Design Mixes	Sand Rubber (0–4.75) %	Coarse Rubber (4.75–9.5) %	Recycled Concrete %	Cement	Ground Calcium Carbonate	Water	Sand	C1	C2	SP	Sand Rubber	Rubber Aggregate	Recycled-Concrete Aggregate
SCC0	SCC	0	0	0	370	80	185	880	440	440	7.3	0	0	0
SCC1	SCC(SR0CR0RC20)	0	0	20	370	80	185	880	440	352	7.3	0	0	78.7
SCC2	SCC(SR20CR0RC20)	20	0	20	370	80	185	704	440	352	7.3	77.9	0	78.7
SCC3	SCC(SR0CR20RC20)	0	20	20	370	80	185	880	352	352	7.3	0.0	39.3	78.7
SCC4	SCC(SR20CR20RC20)	20	20	20	370	80	185	704	352	352	7.3	77.9	39.3	78.7
SCC5	SCC(SR0CR10RC0)	0	10	0	370	80	185	880	396	440	7.3	0.0	19.6	0.0
SCC6	SCC(SR20CR10RC0)	20	10	0	370	80	185	704	396	440	7.3	77.9	19.6	0.0
SCC7	SCC(SR0CR10RC40)	0	10	40	370	80	185	880	396	264	7.3	0	19.6	157.4
SCC8	SCC(SR20CR10RC40)	20	10	40	370	80	185	704	396	264	7.3	77.9	19.6	157.4
SCC9	SCC(SR10CR0RC0)	10	0	0	370	80	185	792	440	264	7.3	39.0	0.0	157.4
SCC 10	SCC(SR10CR20RC0)	10	20	0	370	80	185	792	352	440	7.3	39.0	39.3	0
SCC11	SCC(SR10CR0RC20)	10	0	20	370	80	185	792	440	352	7.3	39.0	0.0	78.7
SCC12	SCC(SR10CR20RC20)	10	20	20	370	80	185	792	352	352	7.3	39.0	39.3	78.7
SCC13	SCC(SR10CR10RC20)	10	10	20	370	80	185	792	396	352	7.3	39.0	19.6	78.7
SCC14	SCC(SR10CR10RC20)	10	10	20	370	80	185	792	396	352	7.3	39.0	19.6	78.7
SCC15	SCC(SR10CR10RC20)	10	10	20	370	80	185	792	396	352	7.3	39.0	19.6	78.7

presents the mix formulations for all mixes in kg/m³.

The mixing process followed a four-step process. First, all the coarse aggregates, rubber, recycled concrete, and sand were mixed for half an hour to achieve a homogeneous mixture. Next, 2/3 of the required water was added while the mixing process continued for an additional minute. Then, the ground calcium carbonate and cement were introduced, and the mixing was continued for another minute. Finally, the superplasticizer and the remaining water were added, and the mixture remained in process for two more minutes. Once the act of mixing was over, the samples were cast for the subsequent tests. After demolding the specimens for one day, they were placed in water for curing over the following days.

Table 5 presents the BBD used in the current study, detailing fifteen different tested conditions and the composition of the newly fabricated concretes. The last three mixes (Runs 13 to 15) share the same composition, as required by the BBD, to evaluate the repeatability and robustness of the experimental results at the center point of the experimental domain. In addition to the designed mixtures, a reference SCC mix, fabricated without any additives, is presented in the first row of Table 5. The rubberized and recycled-concrete mixes were labeled as SCC (SRxCRxRCy), with x as 0, 10, or 20, representing rubber replacement rates, and with y as 0, 20, or 40, representing recycled-concrete replacement rates (Table 5).

3.3. Fresh-State Tests

As shown in Figure 4, four tests were implemented on fresh self-compacting concrete in accordance with the European standards: the slump flow test, L-box test, V-funnel test, and sieve stability test.

The slump flow test, conducted in conformity to EN 12350-8 [54], evaluated the flowability of concrete. The result was reported as the diameter of the spread after the Abrams cone was lifted. With reference to the standard, the spread diameter is defined as no less than 600 mm.

The L-box test, performed following EN 12350-10 [55], assesses the passing ability of SCC through reinforcing bars. The evaluation was based on the ratio of the concrete height at the end of the horizontal section to its initial height in the vertical section. A ratio close to 100% indicated a noticeable passing ability, while the minimum acceptable value defined by the standard is 80%.

The V-funnel test was conducted in accordance with EN 12350-9 [56]. The viscosity and flow rate of the self-compacting concrete (SCC) were determined by measuring the time required for the mixture to completely flow out of the V-funnel apparatus. The standard specifies that this time should not exceed 9 s.

The sieve stability test, based on EN 12350-11 [57], evaluates the segregation resistance of SCC to ensure that the mixture remains uniform during its placement. It was assessed by measuring the percentage of the material passing through a 5 mm sieve relative to the total mass. The standard stipulates that this value should be less than 15%.

3.4. Hardened-State Tests

A comprehensive set of tests was conducted to evaluate the mechanical and physical properties of the hardened concrete mixtures. Measurements of water absorption and hardened density, in addition to evaluations of compressive, tensile, and flexural strength, were included in these tests.

European standard EN 12390-3 was used as a reference while evaluating the compressive strength [58]. Three 15 × 15 × 15 cm³ cube examples were made for every batch of concrete. They were cured for 28 days in a moist environment. A hydraulic servo-controlled machine with a maximum load capacity of 2000 kN was the tool for testing. Tensile strength was measured in accordance with EN 12390-6 [59] as the guideline. For every mix, three cylindrical specimens, 15 cm in diameter and 30 cm in height, were made. A 2000 kN servo-controlled hydraulic machine was used to evaluate the samples through 28-day wet curing.

Flexural strength testing was carried out in accordance with EN 12390-5 [60]. The concrete was cast in prism-shaped molds (10 × 10 × 40 cm³), and after 28 days of wet curing, three specimens per mix were tested using a dedicated flexural-strength apparatus.

Hardened density was determined by referring to EN 12390-7 [61]. Cube specimens (15 × 15 × 15 cm³) were prepared and cured in water for 28 days. For each mix, three samples were selected. Their mass was measured using a high-precision digital balance, while their dimensions (length, width, and height) were accurately recorded by calibrated measuring tools.

Water absorption was evaluated following European standard BS 1881-122 [62]. Concrete was cast into 15 × 15 × 15 cm³ cubes and cured under wet conditions for a 28-day period. After drying, the samples were weighed, then immersed in water for 1, 2, 3, 4, and 10 min. After each interval, the saturated weight was recorded. Water absorption was calculated using Equation (1).

$$Water Absorption (\%)={\displaystyle \frac{W_s-W_d}{W_d}}\times 100$$

(1)

where $W_s$ is the weight of the saturated sample (kg) and $W_d$ is the weight of the dry sample (kg).

Table 6. Testing plan.

Test	Sample	Size	Number of Samples	Age at Test	Standard
Compressive strength	cube	15 × 15 × 15 cm × cm × cm	3	28	EN 12390-3 [58]
Tensile strength	cylinder	30 × 15 cm × cm	3	28	EN 12390-6 [59]
Flexural strength	beam	10 × 10 × 40 cm × cm × cm	3	28	EN 12390-5 [60]
Hardened density test	cube	15 × 15 × 15 cm × cm × cm	3	28	EN 12390-7 [61]
Water absorption	cube	15 × 15 × 15 cm × cm × cm	One cube was used, and the test was repeated 6 times on the same cube.	28	BS 1881-122 [62]

shows the hardened-state tests conducted for all mixes, including the age of each sample and the number of specimens used in each test.

4. Results and Data Analysis

This section provides a meticulous description, thorough interpretation, and analysis of the findings of the current investigation. The measured values of the three strength types—compressive, tensile, and flexural (MPa)—following the Box–Behnken design are shown in

Table 7. Mechanical properties of the tested mixtures.

Run	Design Mixes	Compressive Strength (MPa)	CV (%)	Tensile Strength (MPa)	CV (%)	Flexural Strength (MPa)	CV (%)
SCC0	SCC	38.32	-	3.259	-	6.4	-
SCC1	SCC(SR0CR0RC20)	36.13 ± 0.61	1.69	3.15 ± 0.08	2.54	6.16 ± 0.25	4.06
SCC2	SCC(SR20CR0RC20)	18.93 ± 0.43	2.27	2.20 ± 0.03	1.36	4.20 ± 0.12	2.86
SCC3	SCC(SR0CR20RC20)	16.68 ± 0.66	3.96	1.99 ± 0.04	2.01	4.10 ± 0.02	0.49
SCC4	SCC(SR20CR20RC20)	5.52 ± 0.39	7.07	1.25 ± 0.01	0.80	2.16 ± 0.01	0.46
SCC5	SCC(SR0CR10RC0)	26.14 ± 0.36	1.38	2.45 ± 0.14	5.71	5.58 ± 0.16	2.87
SCC6	SCC(SR20CR10RC0)	8.28 ± 0.59	7.13	1.71 ± 0.21	12.28	3.37 ± 0.10	2.97
SCC7	SCC(SR0CR10RC40)	21.77 ± 0.27	1.24	2.06 ± 0.05	2.43	5.16 ± 0.15	2.91
SCC8	SCC(SR20CR10RC40)	7.76 ± 0.08	1.03	1.24 ± 0.07	5.65	3.39 ± 0.23	6.78
SCC9	SCC(SR10CR0RC0)	23.59 ± 0.54	2.29	2.05 ± 0.05	2.44	5.27 ± 0.08	1.52
SCC10	SCC(SR10CR20RC0)	7.51 ± 0.60	7.99	1.56 ± 0.01	0.64	3.37 ± 0.23	6.82
SCC11	SCC(SR10CR0RC20)	24.11 ± 0.54	2.24	2.68 ± 0.08	2.99	5.45 ± 0.15	2.75
SCC12	SCC(SR10CR20RC20)	8.01 ± 0.94	11.74	1.41 ± 0.01	0.71	3.17 ± 0.09	2.84
SCC13	SCC(SR10CR10RC20)	13.34 ± 0.58	4.35	1.65 ± 0.26	15.76	4.38 ± 0.18	4.11
SCC14	SCC(SR10CR10RC20)
SCC15	SCC(SR10CR10RC20)

. The average ± standard deviation is used to report the results.

As mentioned previously with regard to the Box–Behnken design step, the last three data sets in Table 7 correspond to repeated tests conducted on the same blend composition in order to assess the repeatability and robustness of the experimental procedure. Although each of these three replicates gave slightly different values, a single representative (mean) value is reported in Table 7 for clarity reasons in the presentation of the dataset used for modeling via response surface methodology (RSM). Additionally, to validate the reliability of using three measurements per specimen, coefficients of variation (CV = (standard deviation/average value) × 100) were calculated for selected mixes, also shown in Table 7. The CV values remained within acceptable ranges, indicating statistical consistency and supporting the adequacy of the chosen sample size.

4.1. Fresh-State Test Results

As shown in Figure 5, the slump flow test results reveal a slight decrease in mixture fluidity with increasing amounts of rubber and recycled-concrete ratio (RCA). The reference mix (SCC) achieved the highest flow value of 72 cm. For instance, SCC(SR20CR10RC40), which contained 20% fine rubber and 40% RCA, recorded a slump flow of 68.29 cm—a 5.1% reduction compared to SCC. Similarly, SCC(SR20CR20RC20), with 20% fine and coarse rubber and 20% RCA, exhibited a slump flow of 68.38 cm. More importantly, it confirmed the negative impact of rubber inclusion on flowability. This reduction can be attributed not only to the lower density of rubber and RCA, but also to their irregular particle shapes, high surface roughness, and porous structure. These characteristics increase internal friction, absorb a portion of the mixing water, disrupt the uniform movement of particles, and ultimately hinder the flowability of the self-compacting concrete mixes.

Figure 6 illustrates that the viscosity, as measured by the V-funnel test, increases with the incorporation of rubber and RCA, indicating a more cohesive mixture with greater resistance to flow. The reference mix, SCC, exhibited a V-funnel time of 6 s, whereas rubber-modified mixtures generally showed higher values. For example, SCC(SR20CR20RC20) recorded a V-funnel time of 8.5 s, while SCC(SR20CR10RC40) exhibited a similar trend with a time of 7.9 s. This increase in V-funnel time is likely to refer to the enhanced internal friction and interactions among the rubber particles, which tend to absorb part of the mixing water due to their porous structure, resulting in a denser mix. Added to that, RCA contributes to this effect through its rough surface texture and high water-absorption capacity.

Figure 7 illustrates that the passing ability, as measured by the L-box test, does not exhibit a clear trend with the rubber incorporation. The reference mix SCC had an L-box ratio of 0.86, while most rubber-modified mixtures maintained values between 0.83 and 0.87, indicating that the addition of rubber did not modify the mixture’s ability to pass through obstacles exemplified by reinforcement bars. However, the presence of RCA had a more noticeable effect on the passing ability. For instance, SCC(SR10CR20RC20) displayed an L-box ratio of 0.85, slightly lower than SCC. This reduction can probably be explained by the angularity and the rough texture of the recycled aggregates, which slightly restricted the flow through narrow spaces.

The results of the sieve stability test, illustrated in Figure 8, indicate that the addition of rubber aggregates improved the resistance of the self-compacting concrete (SCC) to segregation. The reference mix exhibited a sieve stability ratio (SR%) of 13%. In contrast, mixtures incorporating rubber, such as SCC(SR0CR10RC0) and SCC(SR20CR10RC0), achieved lower SR% values of 8% and 9%, respectively, reflecting enhanced stability.

This improvement is probably the result of rubber’s special physical properties, particularly its lower density and elastic behavior, which enhance its internal cohesiveness and reduce its paste separation when the material is still fresh. Nevertheless, mixes that contained recycled-concrete aggregates (RCA) showed a modest decrease in stability. SCC (SR10CR0RC20), for instance, had an SR% of 10%. This decrease might be explained by the RCA’s high water-absorption capacity, which could lead to uneven moisture distribution and less mixture uniformity.

4.2. Compressive Strength

The kind and percentage of rubber and RCA added to the mixes have a considerable impact on the compressive strength of the SCC mixtures under examination, as illustrated in Figure 9. Taking all the mixes into consideration, the SCC (SR20CR20RC20) mixture showed the lowest compressive strength, dropping 85.6% from the reference mix (SCC), whereas SCC (SR20CR0RC20) demonstrated a lower reduction of 50.6%, attaining a compressive strength of 18.93 MPa. This discrepancy implies that the coarse rubber’s larger porosity and poorer bond efficiency may have a more negative impact on the compressive strength. These effects have been confirmed by previous microstructural studies showing poor adhesion between rubber particles and the cement matrix as well as the presence of microcracks and voids in recycled-concrete aggregates [23].

Moreover, SCC (SR10CR0RC20), which had fine rubber and RCA, showed moderate strength losses of 37.1%. However, SCC (SR20CR10RC0) and SCC (SR10CR20RC0), which both contained coarse and fine rubber without RCA, showed decreases in compressive strength of 80.4% and 78.4%, respectively. These findings highlight that the simultaneous use of both fine and coarse rubber significantly contributes to strength reduction. Two primary mechanisms contributing to this decrease are (a) the poor bonding between rubber and cement paste compared to the stronger bond typically formed with natural aggregates, and (b) the weak development of the interfacial transition zone (ITZ), which is due to the smoother surface texture of rubber particles relative to the rougher texture of natural aggregates.

Mixtures such as SCC(SR20CR20RC20), SCC(SR20CR10RC0), and SCC(SR10CR20RC0), which recorded compressive strengths below 10 MPa, may be suitable for low-strength applications such as sub-base layers or landscaping concrete, provided that compressive strength is not a critical performance requirement and durability standards are met [37,38,39]. Mixes like SCC(SR20CR0RC20), SCC(SR10CR0RC0), and SCC(SR10CR0RC20) (with strengths ranging from 18 to 25 MPa) are more appropriate for non-structural slabs or playground surfaces [40,41]. Meanwhile, mixtures approaching or exceeding 20 MPa could be considered for light-duty pavements or sidewalk applications [46,48,49,50].

4.3. Tensile Strength

As presented in Figure 10, the tensile strength of the concrete mixtures decreases significantly as natural aggregates are partially replaced with rubber and recycled-concrete aggregates. The reference mix, SCC, exhibited the highest tensile strength of 3.259 MPa. This strength gradually decreased with the increasing levels of substitution, reaching minimum values of 1.30 MPa and 1.25 MPa in SCC(SR20CR10RC40) and SCC(SR20CR20RC20), respectively. This reduction is likely attributed to the elastic behavior and lower stiffness of the rubber particles, which tend to weaken the internal bonding within the cementitious matrix.

Nonetheless, several mixes—such as SCC(SR0CR0RC20) (3.11 MPa), SCC(SR10CR0RC20) (2.75 MPa), and SCC(SR0CR10RC0) (2.52 MPa)—maintained tensile strength levels which remained acceptable for certain non-structural applications. According to previous studies and practical guidelines, concrete mixes with tensile strengths above 2.5 MPa are probably considered suitable for sidewalks and light-duty pavements [49,50]. Mixes with tensile strengths ranging from 2.0 to 2.5 MPa (e.g., SCC(SR20CR0RC20), SCC(SR0CR10RC0), SCC(SR0CR10RC40)) could be appropriate for light industrial floors or non-structural slabs [43,46,48,49,50]. Meanwhile, those with values below 1.5 MPa (e.g., SCC(SR20CR20RC20), SCC(SR20CR10RC40), SCC(SR10CR20RC20)) are more appropriate for sub-base layers or landscaping applications where tensile strength is not a critical performance requirement [33,38,39].

4.4. Flexural Strength

Figure 11 illustrates a gradual reduction in flexural strength as natural aggregates were increasingly replaced by rubber and recycled-concrete aggregates. The control mixture (SCC) recorded the highest flexural strength at 6.40 MPa, highlighting the strong bond between the cement matrix and the natural aggregates. However, with higher replacement ratios, a noticeable decline in flexural strength was observed across all the modified mixes. SCC (SR20CR20RC20), the most impacted mix, showed the lowest strength value at 2.16 MPa, a significant drop of almost 66.25% when compared to the control mix. This decline is linked to the reduced mechanical properties of the rubber particles, specifically stiffness and elasticity, which reduced the effectiveness of adhesion and bonding between the components. These properties contributed to an uneven stress distribution within the mixture. The high porosity of recycled concrete, coupled with its high water-absorption capacity, may also lead to the weakening of the bond between the aggregate and cement paste, particularly when subjected to flexural tests.

Notwithstanding these difficulties, certain mixes, such as SCC(SR0CR0RC20) (6.16 MPa) and SCC(SR0CR10RC0) (5.85 MPa), showed respectable flexural strength values. This implies that the adverse effects of rubber and RCA inclusion can be lessened by using improved additives such as steel fibers or pozzolanic materials. They enhance the internal bonding and gap reduction, which positively affect the mechanical properties and durability of the self-compacting concrete [12]. Furthermore, the relative strength retention observed in some mixes may be due to the uniform dispersion of some rubber particles within the matrix, which would reduce localized stress concentrations and minimize early cracking.

Based on these findings, improving the flexural performance of rubberized concrete may involve optimizing the mix design through better rubber distribution, incorporating mineral additives to enhance interfacial bonding, and adjusting the water-to-cement ratio to achieve a balance between workability and mechanical strength.

4.5. Hardened Density Test

The results clearly demonstrate a progressive reduction in density with the increasing substitution of the natural aggregates by rubber and recycled aggregates, as represented in Figure 12. The reference mix of SCC displayed the highest density at 2342.77 kg/m³. This was clearly expected, since it contained only natural aggregates without any replacement by lighter materials. As rubber and recycled-concrete aggregates were introduced into the mix, a remarkable decline in density was observed, with some mixtures registering values below 2100 kg/m³. For instance, the SCC(SR20CR10RC40) mix recorded a significantly lower density of 1907.21 kg/m³. This reduction can primarily be attributed to the inherently lower densities of rubber and recycled concrete compared to natural aggregates, which significantly affected the overall weight and compactness of the concrete matrix.

4.6. Water Absorption

The experimental results reveal a consistent and gradual increase in water absorption over time across all tested mixtures, highlighting the dynamic nature of moisture movement within the concrete’s porous network. As shown in Figure 13, SCC0 exhibited the lowest absorption values throughout all measured intervals, reaching a final absorption rate of 6.46% after 10 min. In contrast, the SCC(SR20CR10RC40) mixture demonstrated the highest absorption rate, reaching 8.44%, indicating a significant variation in permeability among the different mix designs. Rubberized concrete exhibits higher porosity compared to conventional concrete, as the replacement of natural aggregates with rubber introduces additional voids within the concrete matrix.

This effect was particularly pronounced in mixtures containing only rubber, such as SCC(SR0CR10RC0), SCC(SR20CR10RC0), and SCC(SR10CR0RC0). They recorded significantly higher water-absorption rates than the reference mixture, even in the absence of recycled aggregates. The elevated absorption is essentially explained by the weak interaction between the rubber particles and the cement paste. Rubber, being an elastic material with poor binding properties, leads to the formation of weak interfacial transition zones (ITZs) between the rubber and the surrounding cement matrix. Water could enter these ITZs more easily because they are less cohesive and more porous than the bulk cement paste. The concrete’s compactness is compromised by the weak link between the cement paste and rubber, which makes it more permeable and vulnerable to moisture intrusion.

SCC(SR0CR10RC40) and SCC(SR20CR10RC40), two mixes that contained recycled aggregates, showed the highest water-absorption values. After 10 min, SCC(SR20CR10RC40) reached 8.44%, the highest of all the tested mixes. The high absorption is primarily attributed to the porous structure of the recycled aggregates, which often includes remnants of old, unreacted cement paste. These remnants, which retain partial moisture, were the key factor in enhancing the aggregate’s water-absorption capacity. In addition, the high porosity introduces additional voids into the concrete structure, increasing water movement within the mixture.

Consequently, these mixes are likely to become saturated with water, making them more susceptible to moisture-related problems over time. High permeability could also contribute to reduced long-term durability by increasing susceptibility to freeze–thaw cycles, corrosion of the reinforcing steel, and penetration of harmful chemicals into the concrete. Ultimately, these elements may jeopardize the concrete’s service life and structural integrity. As seen in SCC(SR0CR10RC40) (8.13%) and SCC(SR10CR10RC20) (8.05%), the overall porosity of the concrete is further aggravated when rubber and recycled aggregates are combined, leading to a significant rise in the water-absorption ratio. The poor interfacial bond between the rubber particles and the cement paste, as well as the recycled aggregates’ natural porosity, are both determining elements in this rise.

Once both components were present at the same time, the concrete’s permeability greatly increased, which would make it easier for water to enter a building. This finding emphasizes that the combined impact of adding rubber and recycled aggregates results in a noticeable decrease in the concrete’s resistance to water permeability, in addition to a decrease in density. The water-absorption rate increases in proportion to the percentage of these substitute aggregates, which might seriously impair the concrete’s long-term durability. Because of this increased permeability, the concrete is more susceptible to moisture-related damage, which might include degradation under environmental stresses such as freeze–thaw cycles, corrosion of reinforcement, and deterioration of the cement matrix.

Furthermore, the formation of porous interfacial transition zones, which further increase permeability and moisture infiltration, is a result of the poor interaction between rubber and cement paste. These results corroborate how crucial it is to carefully balance the use of rubber and recycled aggregates in mix design in order to reduce their negative impacts on concrete performance. The long-term durability of concrete is significantly impacted by the variations in the absorption behavior between the mixes. Increased water absorption may cause the possibly easy entry of dangerous substances such as sulfates, chlorides, and moisture from freeze–thaw cycles. Over time, they could hasten the deterioration and jeopardize the structural integrity.

The use of rubber and recycled aggregates reduces building waste and promotes sustainability, bringing benefits for the environment. Nonetheless, it presents potential problems because of the increased permeability. To mitigate these effects, extra steps would be required, such as applying surface sealants, adding pozzolanic additives (such as fly ash or silica fume), or creating mix designs that were optimal. These techniques would guarantee the long-term performance of rubberized and recycled aggregate concrete and assist in maintaining sufficient durability under a range of exposure situations.

4.7. Statistical Analysis and Optimization of Mechanical Properties

The multivariable quadratic model used in this investigation for the three responses is shown in Equation (2):

$${\widehat{y}}_k=a_0+{\sum }_{i=1}^3{a}_i·x_i+{\sum }_{\begin{array}{c}i=1\\ j>i\end{array}}^3{a}_{ij}·x_i·x_j+{\sum }_{i=1}^3{a}_{ii}·x_i^2$$

(2)

where the answers’ expected values are denoted by ${\widehat{y}}_k$, where ${\widehat{y}}_1$ stands for compressive strength (MPa), ${\widehat{y}}_2$ for tensile strength (MPa), and ${\widehat{y}}_3$ for flexural strength (MPa). The intercept, linear, interaction, and quadratic terms are represented by the coefficients $a_0$, $a_i$, $a_{ij}$, and $a_{ii}$, respectively. The level of factor i in the kth experiment is shown by $x_{i_k}$.

STATISTICA 12 (StatSoft, Inc., 1984–2014, Tulsa, OK, USA) [63], a powerful program frequently used for regression analysis and the examination of variable interactions in experimental investigations, was used in this study for both the experimental design and statistical analyses. Predictive models to evaluate the mechanical behavior of SCC using recycled components were made easier by the software.

The influence of both individual and combination factors, as well as the reliability of the model, was evaluated using t-tests and analysis of variance (ANOVA). The quality and appropriateness of the finished models were assessed using statistical metrics like the root mean square error (RMSE), modified R², and coefficient of determination (R²). p-values below 0.05 were regarded as statistically significant. Furthermore, the constructed models’ predictive power and statistical robustness were validated by high F-values from Fisher’s testing [64].

Table 8. Recapitulation of estimated regression coefficients and Student’s t-test for all studied responses.

	Compressive Strength (MPa)	Tensile Strength (MPa)	Flexural Strength (MPa)
	Model Coefficients
Constant	17.04 ***	1.98 ***	4.28 ***
Sand rubber (%):A	−7.53 ***	−0.41 **	−0.99 ***
Coarse rubber (%): B	−8.13 ***	−0.48 **	−1.04 ***
Recycled concrete (%): C	−0.48 NS	−0.05 NS	−0.05 NS
A2	−1.54 **	−0.11 NS	0.04 *
B2	−1.47 **	−0.14 NS	0.07 **
C2	0.22 NS	0.001 NS	−0.04 NS
A × B	1.51 *	0.05 NS	0.01 NS
A × C	0.96 NS	−0.02 NS	0.11 *
B × C	−0.01 NS	−0.10 NS	−0.09 *
	Model Fit Quality
R2 (%)	99.44	94.10	99.89
R2Adj (%)	98.42	83.47	99.68
RMSE (MPa)	1.10	0.22	0.06
p-value	< 0.001	0.014	< 0.001

*** Strongly significant influence: p < 0.001; ** very significant influence: p < 0.01; * significant influence: p < 0.05; and ^NS: not significant.

presents the estimated regression coefficients determined using the least-squares method, along with the results of the student’s t-tests for the studied responses: compressive, tensile, and flexural strengths. Furthermore, Table 8 includes key statistical parameters used to assess the appropriateness of the determined model.

It seemed evident that the sand rubber content (A) and the coarse rubber content (B) had a highly significant negative linear effect on all tested strength types (p < 0.01). This effect was attributed to the highly prominent decrease in strength responses as both factors increased from lower to higher levels. This trend is clearly illustrated in Figure 14a,b,d,e,g,h for both Factors A and B, and in Figure 14c,f,i specifically for Factor B. However, the recycled-concrete content (C) did not exhibit a significant linear influence on any of the strength responses, as shown in Figure 14c,f,i.

Regarding the different interactions’ effects, the interaction between A and B (A × B) significantly influenced the compressive strength in isolation, with a positive coefficient (1.51; p < 0.05; Figure 14a). In contrast, the interactions A × C and B × C significantly affected only the flexural strength: A × C showed a positive coefficient (0.11; p < 0.05; Figure 14h), whereas B × C displayed a negative coefficient (−0.09; p < 0.05; Figure 14i). The percentages of sand rubber (A) and coarse rubber (B) exhibited highly significant negative quadratic effects on compressive strength (−1.54 and −1.45, respectively), resulting in an upward concave trend with a minimum within the studied experimental range (Figure 14a–c). Conversely, these two factors showed significant positive quadratic effects on flexural strength (0.04 and 0.07, respectively), indicating a downward concave trend with a maximum (Figure 14g–i). In contrast, the quadratic effect of the percentage of the recycled-concrete content (C) did not influence any responses (Figure 14c,f,i).

Equations (3)–(5) present the final relationships for compressive (${\widehat{y}}_1$), tensile (${\widehat{y}}_2$), and flexural (${\widehat{y}}_3$) strengths, respectively.

$$\begin{gathered} {\widehat{y}}_1=17.04-7.53·x_1-8.13·x_2-0.48·x_3+1.51·x_1·x_2+0.96·x_1·x_3-0.01·x_2·x_3-1.54·x_1^2-1.47 \\ ·x_2^2+0.22·x_3^2 \end{gathered}$$

(3)

$$\begin{gathered} {\widehat{y}}_2=1.98-0.41·x_1-0.48 ·x_2-0.05·x_3+0.05·x_1·x_2-0.02·x_1·x_3-0.10·x_2·x_3-0.11·x_1^2-0.14 \\ ·x_2^2+0.001·x_3^2 \end{gathered}$$

(4)

$$\begin{gathered} {\widehat{y}}_3=4.28-0.99·x_1-1.04 ·x_2-0.05·x_3+0.01·x_1·x_2+0.11·x_1·x_3-0.09·x_2·x_3+0.04 ·x_1^2+0.07 \\ ·x_2^2-0.04 ·x_3^2 \end{gathered}$$

(5)

where $x_1$, $x_2$, and $x_3$ represent the coded level (as −1, 0, or +1) of the three factors.

The three models could be considered robust and effective in describing the three studied responses, as evidenced by their high R² and adjusted R² values, low RMSE, and particularly low p-values (Table 8). This strong predictive performance was further supported by Figure 14, which delineates minimal differences between the experimental data points (o) and the fitted 3D response surfaces.

4.8. Mechanical Properties Optimization for Common Non-Structural Applications

Since the three models for the mechanical properties (Equations (3)–(5)) were valid by exhibiting strong statistical significance, the Optimizer Tool in the STATISTICA software was employed to determine the optimal proportions of sand rubber, coarse rubber, and recycled concrete needed to achieve the desired compressive and tensile strength values for various common non-structural applications. The results are presented in

Table 9. Limit substitution values for several components for non-structural applications of the composite.

Applications	Compressive Strength (MPa)	Tensile Strength (MPa)	Determined Values of Several Components			Predicted Value of Responses
			Sand Rubber (%)	Coarse Rubber (%)	Recycled Concrete (%)	Compressive Strength (MPa)	Tensile Strength (MPa)	Flexural Strength (MPa)
Road sub-base low-strength concrete	3–10	<1	5.5–20	20	40	5.2–10.08	1.05–1.41	2.18–3.54
(Landscaping) utility bedding or backfill for drainage systems	7–10	<1	5.5–10	20	40	7.18–10.08	1.20–1.41	3.13–3.54
Playground flooring	15–20	1.5–2	0–0.5	20	40	14.77–15.32	1.75–1.79	3.88–3.92
Non-structural slabs	10–20	1–2	0–5.5	20	40	10.08–14.77	1.41–1.75	3.51–3.88
Light industrial floors	20–25	2–2.5	3.5–5.5	3.5–6	40	20.00–25.15	2.16–2.55	5.21–5.62
Sidewalks and walkways	20–30	2–3	5–5.5	0–6	40	20.00–28.26	2.16–2.85	5.21–5.81

. It is worth noting that, regarding the currently used optimization, the amount of recycled concrete was deliberately maximized to 40%.

Premised on the results given in Table 9, it could be noted that the lowest strength values were associated with the mixtures containing 5.5–20% sand rubber, 20% coarse rubber, and 40% recycled-concrete aggregate, as used in road sub-base applications. These exhibited compressive strengths as low as 5.2 MPa, tensile strengths of 1.05 MPa, and flexural strengths of 2.18 MPa. In contrast, the best performance was recorded in sidewalk and walkway applications, where the mixture included 5–5.5% sand rubber, 0–6% coarse rubber, and 40% RCA, achieving a compressive strength up to 28.26 MPa, tensile strength of 2.85 MPa, and flexural strength of 5.81 MPa.

The results of this study demonstrate that incorporating sand rubber, coarse rubber, and especially recycled concrete could produce sustainable SCC suitable for a wide range of non-structural applications. This fact highlights the significance of the work, which was based on a multivariable analysis using response surface methodology.

5. Conclusions

The current findings were corroborative of the fact that self-compacting concrete (SCC) incorporating rubber and recycled aggregate represents a sustainable alternative for producing non-structural concrete using recycled materials. Based on the experimental tests and analyses conducted, the following conclusions can be drawn: The workability of fresh concrete is slightly reduced by increasing the proportions of rubber and recycled aggregate. This reduction can be attributed to the high water-absorption capacity and rough surface texture of both materials, which reversibly increase the internal friction and reduce the flow itself. Nevertheless, all the mixtures remained within the acceptable performance limits for SCC.

Compressive strength significantly decreases with increasing rubber content, particularly when coarse rubber is incorporated. The decrease in compressive strength is due to the increased porosity in the mixture resulting from the porous composition of rubber and RCA, in addition to the weak bond between these materials and the cement paste surface, which reduces the efficiency of the stress transmission within the concrete. The differences in density and the mechanical behavior of these materials also increase the occurrence of internal weaknesses. However, mixes containing moderate proportions (e.g., 10% fine rubber and 20% recycled aggregate) achieved strengths above 20 MPa, rendering them suitable for applications such as pavements, playground surfaces, and subgrade layers. Additionally, the density of the hardened concrete gradually decreased with higher recycled aggregate content, due to its lower specific gravity compared to natural aggregate. Tensile and flexural strengths exhibit smaller reductions compared to compressive strength, suggesting that the inclusion of rubber enhances the ductility and crack resistance of hardened concrete. Water-absorption rates increase with higher recycled aggregate content, reflecting greater porosity. However, this does not critically impair the usability of the mixes for non-structural applications.

The findings support the feasibility of using SCC containing rubber and recycled aggregate as a sustainable material for non-structural applications. These mixtures meet the mechanical and workable requirements while contributing to waste reduction and minimizing the environmental footprint of the construction sector.

The optimization analysis confirmed that the mechanical performance of SCC incorporating recycled materials is highly dependent on component proportions. The lowest strength values were associated with the mixtures containing 5.5–20% sand rubber, 20% coarse rubber, and 40% recycled-concrete aggregate, as used in road sub-base applications. These exhibited compressive strengths as low as 5.2 MPa, tensile strengths of 1.05 MPa, and flexural strengths of 2.18 MPa. In contrast, the best performance was recorded in sidewalk and walkway applications where the mixture included 5–5.5% sand rubber, 0–6% coarse rubber, and 40% RCA, achieving a compressive strength up to 28.26 MPa, tensile strength of 2.85 MPa, and flexural strength of 5.81 MPa. These findings define practical upper and lower limits for each component, emphasizing that moderate rubber content and consistent RCA levels could effectively balance sustainability and structural adequacy.

6. Recommendations

Based on the results of this study, it is recommended that future research expand the examination of a wider range of factors within statistical optimization models, such as different curing degrees and additional percentages of recycled materials, to develop more effective self-compacting concrete mixes. It is also recommended to lead an exploration into the effect of adding admixtures such as fibers or pozzolanic materials, given their potential role in enhancing the mechanical properties and durability, particularly in non-structural applications such as sidewalks, walkways, and landscaping.

Although the current study did not include microscopic analyses such as scanning electron microscopy (SEM), conducting such analyses in future studies could contribute to a deeper understanding of the internal bonding behavior and porosity structure of concrete containing recycled aggregates, enhancing the interpretation of the results obtained at the macro level.