Optimization of Mechanical and Durability Properties of Manganese Slag Hybrid Fiber Concrete Using an L9 Orthogonal Design and Grey Relational Analysis

Concrete is a composite material composed of cement, fine aggregates, coarse aggregates, and water. Portland cement, a key ingredient in conventional concrete, functions as a binder that ensures the cohesion of aggregates within the mixture [1]. However, cement production is associated with significant environmental concerns due to its energy-intensive manufacturing process and substantial CO₂ emissions [2]. The calcination of limestone and the high-temperature clinkerization process in cement kilns contribute approximately 7–8% of global anthropogenic CO₂ emissions, making the development of alternative cementitious materials a critical research priority. For each ton of cement produced in 2020, approximately 0.59 tons of CO₂ were typically emitted [3]. Consequently, there is a growing emphasis on exploring supplementary cementitious materials (SCMs) that are energy-efficient and environmentally sustainable, while maintaining comparable or superior mechanical properties to Ordinary Portland Cement (OPC). One such promising SCM is manganese slag (MnS), an industrial by-product derived from the production of electrolytic manganese [4].

Electrolytic MnS is generated as a solid waste in filters following the sulfuric acid leaching of manganese carbonate ore, oxidative deferrization with MnO₂, and lime neutralization [5]. Recent studies have highlighted its potential as an SCM due to its pozzolanic and latent hydraulic activity, which facilitates cement hydration and enhances concrete properties. Infrared spectroscopy and other experimental techniques have confirmed that MnS exhibits reactivity similar to natural pozzolanic materials, enabling its partial replacement of cement in concrete production [6]. MnS demonstrates both hydraulic and pozzolanic behavior, with a reactivity range lower than that of ground granulated blast furnace slag (GGBFS), but higher than that of fly ash. Its chemical composition, which includes CaSO₄·2H₂O, SiO₂, Al₂O₃, Fe₂O₃, and MnO₂, allows it to interact with cement hydration products, improving the microstructure and durability of hardened concrete [7]. It is found that calcium aluminate silicate hydrate (CASH) gel is also the hydration product of manganese slag cement slurry, except for calcium hydrate crystal and CSH gel [8]. The sulphate in the electrolytic manganese slag can react with the aluminum minerals in the concrete, effectively improving the volume stability. The dense concrete structure can effectively solidify the heavy metal ions in the electrolytic manganese slag [9]. The active Al₂O₃ component in GGBFS reacts with a sulphate activator to create Aft [10]. As a result, MnS has expanded beyond traditional low-value applications, such as road base material and aggregate substitution in bricks and blocks, and is now being considered for high-performance concrete applications [11]. MnS enhances marine concrete's sulfate resistance: 30% replacement improved resistance by 33% after 180 days by densifying microstructure and reducing calcium hydroxide, though 40% replacement slightly decreased performance due to lower early strength [12]. High-temperature sintered MnS, possessing vitrified properties and high compressive strength, can be crushed and graded to serve as an aggregate substitute in concrete. Calcined MnS at 1100℃ exhibits large grains and surface pores, with CaSO₄ particles identified via elemental mapping. These CaSO₄ particles enhance cement hydration by releasing SO₄²⁻ to form C-S-H gel with cement components, densifying the concrete microstructure and improving flexural strength [13]. This advancement aligns with sustainable construction practices by integrating industrial waste into concrete, reducing cement dependency, and lowering the carbon footprint of construction materials [14].

Despite its advantages, conventional concrete suffers from inherent limitations, including high compressive strength but low flexural and tensile strength, leading to brittleness and reduced durability under mechanical loads. To address these weaknesses, fibers and industrial by-products with pozzolanic activity are often incorporated into concrete mixtures to improve flexural strength and fracture toughness [15]. Concrete is classified as a quasi-brittle material, meaning it exhibits limited plastic deformation before failure. The addition of fibers enhances the material’s toughness by bridging cracks and improving post-cracking behavior, thereby mitigating premature failure [16]. Among various fiber types, polypropylene fibers (PPF) are widely utilized due to their chemical stability in alkaline environments, availability, and cost-effectiveness. While PPF has a lower modulus of elasticity and strength compared to high-strength fibers such as steel, its ductility enables it to enhance the impact resistance and energy absorption capacity of concrete [17]. Research has demonstrated that incorporating PPF into concrete mixtures enhances stress resistance, fatigue life, thermal stability, and mechanical properties, particularly in applications requiring enhanced tensile and shear strength [18-20].

To further optimize concrete performance, hybrid fiber-reinforced systems have gained significant attention, as they combine fibers with different mechanical properties to achieve superior mechanical and durability characteristics. Low-modulus fibers such as polypropylene, polyamide, and basalt are commonly used to suppress early-age shrinkage cracks, improve resistance to temperature variations, and enhance concrete’s resistance to microcracking under mechanical loads [21, 22]. Experimental findings indicate that PPF reinforcement improves the microstructural integrity of concrete by preventing the propagation of microcracks, thereby enhancing resistance to shrinkage, freeze-thaw cycles, and sulfate attack [23, 24]. Additionally, research on high-strength concrete reinforced with fibers has shown that fiber incorporation significantly increases fracture toughness, mitigates brittle failure, and effectively dissipates fracture energy [25-27]. Given that concrete is a multi-phase composite material, strategically combining different fiber types can optimize performance at the microscopic, mesoscopic, and macroscopic levels, contributing to enhanced mechanical properties and long-term durability.

Among fiber types, steel fibers (SF) are widely recognized for their ability to enhance both strength and toughness. Research has demonstrated that incorporating SF into concrete at volume fractions of 0.5%, 1.0%, 1.5%, and 2.0% leads to significant improvements in compressive and flexural properties [28]. Experimental studies suggest that 1.5% SF reinforcement is optimal, yielding a compressive strength increase of approximately 15.3% while also significantly improving post-cracking strength and flexural ductility. The use of hybrid SF-PPF systems has been found to further enhance flexural properties, crack propagation resistance, and energy absorption capacity [29-31]. Comparative studies of fiber-reinforced concrete mixtures have shown that SF primarily enhances mechanical strength, while PPF contributes to improving toughness, durability, and resistance to environmental degradation [32, 33]. Additionally, the effect of PPF dosage on concrete performance under elevated temperatures indicates that incorporating 1 kg/m³ of fibers enhances heat resistance and residual strength, making fiber-reinforced concrete a viable material for extreme environmental conditions [18, 34].

Grey relational analysis (GRA) has been widely applied in material optimization studies, particularly for evaluating multi-factor performance in composite materials. Before pavement construction, the optimum asphalt content of hot mix asphalt (HMA) must be determined through mix design. A GRA-based algorithm was developed for HMA design optimization, where asphalt mixture specimens were comparatively tested to validate the algorithm's effectiveness [35]. Similarly, another study systematically investigated and optimized the influence of key parameters - specifically the ground granulated blast furnace slag replacement ratio - using grey relational analysis methodology [36]. The methodological framework provides a scientific basis for systematically evaluating the influence of various curing-age mix proportioning schemes on the overall performance of concrete. These applications demonstrate GRA's capability to handle complex multi-parameter optimization problems in material science.

This study systematically investigates the synergistic effects of MnS, SF, and PPF on MnS hybrid fiber-reinforced concrete (MHFC) through an integrated experimental and analytical approach. The research examines the influence of MnS-SF-PPF ratios on mechanical properties and durability indicators, establishes optimal mixture proportions using an L9 orthogonal experimental design coupled with grey relational analysis across multiple curing ages, and elucidates the microstructural mechanisms underlying performance enhancement, with particular focus on MnS's role as a supplementary cementitious material in pore structure refinement and calcium hydroxide reduction.

The findings provide an optimized mix design framework that enhances the mechanical properties and durability of concrete, while promoting MnS as a high-performance supplementary cementitious material. The framework reduces cement consumption and carbon dioxide emissions, and lowers production costs through industrial byproduct utilization. Furthermore, it provides practical solutions for offshore structures and chloride-rich environments, demonstrating both economic and environmental benefits for sustainable construction.

3.1 Test results of specimens

The test specimens were subjected to MHFC strength tests and chloride ion flux tests under standard conditions. with their compressive strength (F_c), flexural strength (F_r), and flux (Q) shown in Table 7. Specifically, F_c7 denotes the compressive strength of concrete at 7 days of age (unit: MPa), F_r28 denotes the flexural strength at 28 days of age (unit: MPa), and Q₅₆ denotes the flux at 56 days of age (unit: Coulomb). The meanings of other symbols are similar.

Table 7. Test results of MHFC specimens at 7d, 28d, 56d, and 91d

The ANOVA results for the three factors—M_MnS (A), V_SF (B), and V_PPF (C)—at curing ages of 7, 28, 56, and 91 days are presented in Tables 8 to 10. In these tables, k₁, k₂, and k₃ represent the sum of test results at levels 1, 2, and 3 of each factor, respectively; μ₁, μ₂, and μ₃ are their corresponding mean values. R denotes the range, and S² is the variance.

3.2 Range and variance analysis of compressive strength at different ages

According to the data analysis in Table 8, the addition of MnS, SF, and PPF significantly influences the compressive strength of MHFC, albeit to varying degrees compared to the control group. Range and variance analyses indicate that the order of influence on compressive strength is as follows: V_SF (B) > M_MnS (A) > V_PPF (C). This suggests that steel fiber has the most pronounced effect on enhancing compressive strength, followed by MnS, while PPF has the least impact.

Table 8. Range and variance of the MHFC compressive strength at 7d, 28d, 56d, and 91d

The results show that the compressive strength of conventional concrete without MnS and fiber at 28 days is 35.88 MPa, meeting the C30 strength requirement. A closer examination of the data reveals that, with an M_MnS of 10%, the compressive strength of the specimens initially increases and then decreases as the V_SF rises. Specifically, as the V_SF increases from 0.5% to 1.0% and then to 1.5%, the compressive strength improves from 35.88 MPa to 40.21 MPa, 44.30 MPa, and 42.62 MPa, respectively—representing increases of 12%, 23%, and 18% compared to ordinary concrete. This indicates the presence of an optimal V_SF, beyond which excessive fiber addition may result in diminishing returns in strength enhancement.

Furthermore, the compressive strength of specimens increases with higher MnS content, reaching its maximum value at 20% MnS replacement. Similarly, increasing the SF content initially enhances compressive strength, peaking at 1.0% SF, beyond which a decline is observed. In contrast, PPF has only a marginal effect on compressive strength enhancement. Additionally, the most significant strength gain occurs between 7 and 28 days, while the increase from 56 to 91 days is comparatively minor. This suggests that the mechanical properties of the MHFC tend to stabilize over time as hydration and pozzolanic reactions progress.

The observed improvements in compressive strength can be attributed to the synergistic interaction between MnS and hybrid fibers, which together form a stable internal network structure within the MHFC matrix. The incorporation of SF and PPF enhances the bond between the aggregates and cement, thereby improving resistance to shear stress and inhibiting crack development. This reinforcement mechanism contributes to enhanced mechanical performance by hindering crack propagation and increasing overall structural integrity. However, when the fiber dosage exceeds a critical threshold, inadequate dispersion within the matrix may lead to fiber clumping. This results in inhomogeneous stress distribution, disruption of internal structural stability, and ultimately, a reduction in compressive strength. The findings highlight the importance of optimizing the proportions of MnS and fiber content to maximize compressive strength. Achieving a proper balance in the mix design ensures the formation of an interconnected and stable microstructure, prevents excessive fiber clustering, and maintains structural performance. These results provide valuable insights into the mechanical behavior of MHFC, offering guidance for future research and practical applications in sustainable construction materials.

3.3 Range and variance analysis of flexural strength at different ages

The experimental results in Table 9 indicate that incorporating MnS, SF, and PPF enhances the flexural strength of MHFC to varying degrees. Range and variance analyses reveal that the influence of the factors ranks in the following order: V_SF (B) > M_MnS (A) > V_PPF (C). This highlights that V_SF has the most significant effect on improving flexural strength, followed by M_MnS, while V_PPF exhibits a comparatively weaker impact.

Table 9. Range and variance of the MHFC flexural strength at 7d, 28d, 56d, and 91d

Data analysis further demonstrates that the introduction of MnS and mixed fibers into C30 concrete generally enhances the flexural strength of the specimens compared to the baseline group, although the extent of improvement varies. At 7 days of age, the flexural strength of the fiber control group was recorded at 4.64 MPa. Both MnS and fiber reinforcement were found to improve flexural strength, with the degree of enhancement depending on fiber content. As the SF volume fraction increased from 0.5% to 1.0%, the flexural strength of the specimens improved by 11% and 33%, respectively. However, when the SF content was further increased to 1.5%, the flexural strength increment dropped to 23%, suggesting the existence of an optimal fiber content beyond which additional reinforcement yields diminishing returns. The results indicate that flexural strength initially increases and then declines with rising SF content, reaching a peak at 1.0%. In contrast, increasing the PPF content from 0.5% to 2.0% does not result in a significant enhancement in flexural strength, reaffirming the relatively limited contribution of PPF to this property. Additionally, the increase in flexural strength from 7 to 28 days is more pronounced than the increment observed from 56 to 91 days, suggesting that the mechanical properties tend to stabilize over time as the material continues to mature.

The observed improvement in flexural strength can be attributed to the formation of a micro-reinforcement network within the matrix, facilitated by the uniform dispersion of MnS and hybrid fibers. This network enhances the internal microstructure, mitigates crack initiation and propagation, and ultimately strengthens the concrete matrix. Steel fibers play a critical role in reinforcing weak zones within the specimens due to their high shear resistance, which effectively improves structural integrity and further enhances flexural strength. These findings underscore the importance of optimizing fiber content to maximize the flexural performance of MHFC.

3.4 Range and variance analysis of the electric flux of chloride ion corrosion in MHFC at different ages

Through range and variance analysis in Table 10, the influencing factors on the electric flux of MHFC are ranked in the order: M_MnS (A) > V_SF (B) > V_PPF (C). It indicates that the incorporation of MnS and fibers into the MHFC matrix contributes to a reduction in electric flux, thereby enhancing resistance to chloride ion penetration.

Table 10. Range and variance of the MHFC chloride ion flux at 7d, 28d, 56d, and 91d

In detail, M_MnS consistently plays a dominant role in enhancing the chloride ion penetration resistance of MHFC, as evidenced by its significantly higher R and S² values compared to other factors, indicating a strong and stable influence. V_SF shows a relatively obvious promoting effect at early curing ages, but its effect gradually weakens as the curing age extends. However, once the V_SF surpasses a certain threshold, the electric flux begins to increase, indicating a decline in chloride ion resistance. For instance, at 28 days of curing, when the M_MnS remains constant at 10%, increasing the SF content from 0.5% to 1.0% results in a 24% reduction in electric flux. However, when the V_SF is further increased from 1.0% to 1.5%, the electric flux increases by 17%, leading to a drop in chloride ion penetration resistance from Q-II to Q-I classification. In contrast, V_PPF had a weaker effect at all ages, with consistently lower R values and S² values. Furthermore, as the curing age increases, the electric flux values gradually decline, indicating a continuous improvement in chloride resistance. The reduction in electric flux from 7 days to 28 days is more pronounced than the reduction observed between 56 days and 91 days, suggesting that the MHFC structure stabilizes over time.

3.5 Grey relational analysis of NHFC comprehensive performance at different ages

A grey relational analysis was conducted on 36 sets of experimental data for NHFC at 7, 28, 56, and 91 days. For each curing age, the 9 experimental datasets were treated as an independent grey system. The optimal values within each system were designated as the ideal reference sequence (X₀), representing the target performance, while the remaining experimental data served as comparison sequences (Xᵢ). After the data were normalized, gray correlation analysis was performed. The correlation coefficients and correlation degrees were calculated using formulas (1)-(5), and the correlation degrees of the four age stages were ranked separately. The results of the gray correlation analysis are shown in Tables 11-14.

Table 11 provided the results for NHFC at 7 curing ages. The MnS₂₀SF₁PPF₀.₅ group achieved the highest grey relational grade (GRG) of 0.618, indicating superior early-age performance due to its optimal combination of 20% manganese slag (MnS₂₀), 1% silica fume (SF₁), and 0.5% polypropylene fiber (PPF₀.₅). In contrast, MnS₅SF₀.₅PPF₀.₅ ranked lowest (GRG: 0.346), suggesting that low slag and silica fume content hinder early strength development. Notably, flexural strength (Fᵣ) and compressive strength (Fc) had higher correlation coefficients than Q (e.g., workability or permeability), emphasizing mechanical properties as critical for early performance.

Table 11. GRA of 7-Day MHFC Properties

In Table 12, MnS₂₀SF₁PPF₀.₅ maintained dominance (GRG: 0.781), with its GRG increasing by 26.4% from 7 to 28 days, highlighting rapid strength gain. The MnS₁₀SF₁PPF₂ group (GRG: 0.657) emerged as a strong alternative, suggesting that higher fiber content (PPF₂) compensates for reduced slag (MnS₁₀). The lowest-performing group, MnS₅SF₀.₅PPF₀.₅ (GRG: 0.432), further validated that 5% slag inadequately supports mid-term hydration.

Table 12. GRA of 28-Day MHFC Properties

From Table 13, it can be seen that MnS₂₀SF₁PPF₀.₅ achieved near-perfect alignment with ideal properties (GRG: 0.948), while MnS₁₀SF₁PPF₂ (GRG: 0.863) demonstrated diminishing returns from PPF₂, implying fiber dosages beyond 0.5% offer marginal benefits. The MnS₅SF₀.₅PPF₀.₅ group (GRG: 0.519) remained the weakest, reinforcing that low slag and silica fume content limit long-term durability.

Table 13. GRA of 56-Day MHFC Properties

Table 14 provided that MnS₂₀SF₁PPF₀.₅ (GRG: 0.929) and MnS₁₀SF₁PPF₂ (GRG: 0.881) exhibited sustained excellence, but MnS₂₀SF₁.₅PPF₁ (GRG: 0.786) underperformed, indicating excessive silica fume (SF₁.₅) may reduce long-term efficacy. The consistent poor performance of MnS₅ groups across all ages suggests that ≥ 10% slag is essential for viable NHFC formulations.

Table 14. GRA of 91-Day MHFC Properties

3.6 Discussion

The grey relational analysis identifies MnS₂₀SF₁PPF₀.₅ as the optimal NHFC mix, exhibiting consistently high GRG at 7–91 days (Tables 11–14), with exceptional synergy between compressive/flexural strength (Fc/Fᵣ) and durability (Q). This result is corroborated by orthogonal analysis, confirming its reliability for material design.

The observed trend can be attributed to three primary mechanisms. First, MnS generates a dense C-S-H gel through the pozzolanic reaction, enhancing hydration efficiency and producing an interconnected, compact network of C-S-H gel. The interaction between C-S-H gel and secondary hydration products (e.g., ettringite) further improves microstructural integrity. Concurrently, Ca(OH)₂ crystals become less pronounced, indicating that the pozzolanic reaction between MnS and cement effectively consumes free Ca(OH)₂, thereby optimizing the pore structure of concrete and enhancing its resistance to chloride ion penetration. SF reduces chloride ion intrusion pathways by bridging microcracks and inhibiting crack propagation in the early stages, but its long-term effectiveness may be limited by the stability of the interface between the fibers and the matrix. PPF, however, has a low elastic modulus and weak interaction with the matrix, resulting in a negligible improvement in resistance to chloride ion penetration. Secondly, MnS and hybrid fibers contribute to modifying the internal pore structure distribution within the MHFC, thereby enhancing durability. Thirdly, these materials actively absorb the cement hydration product Ca(OH)₂, leading to a gradual decrease in the concentrations of Ca²⁺ and OH⁻ ions within the matrix. This reduction in ion concentration subsequently lowers the overall conductivity of the MHFC, resulting in a progressive decrease in electric flux values. However, with excessive fiber content, localized fiber accumulation can lead to fiber clustering, promoting microcrack formation, which adversely affects chloride ion resistance. The most significant reduction in electric flux occurs between 7 and 28 days, coinciding with the strongest absorption effect of Ca(OH)₂ by MnS and fibers. In contrast, the reduction observed from 56 to 91 days is comparatively smaller, reflecting a more stabilized phase of hydration and microstructural development.

From an engineering application perspective, it is recommended that the M_MnS be controlled between 10-20% to ensure the long-term durability of concrete. V_SF should be increased in dosage (1.0-1.5%) during the early age stage (7-28 days) and appropriately reduced (0.5-1.0%) in the later stage (56-91 days) to achieve a balance between performance and cost. V_PPF can be reduced or omitted. In chloride-ion corrosive environments, special attention should be paid to the synergistic effect of manganese slag and steel fibers during the early stage (7–28 days), while ensuring that the manganese slag content serves as the core parameter for controlling the concrete's long-term resistance to chloride ions.

Based on the experimental results and systematic comparisons with similar studies, it is recommended to control the MnS content between 10-20% to ensure long-term concrete durability, while dynamically adjusting SF dosage (1.0-1.5% at 7-28 days, reduced to 0.5-1.0% at 56-91 days) for cost-performance balance, and minimizing or omitting PPF. Early-stage MnS-SF synergy is critical for chloride resistance, supported by slag-fiber studies. The MnS range aligns with cementitious mechanisms, while PPF reduction contrasts with some findings (e.g., 0.5% PPF benefits), likely due to hybrid-system interactions. Future work will integrate SEM/XRD to validate MnS-SF interface mechanisms, extending the macro-scale findings presented here.