Mechanical Properties and Microstructure of Alkali-Activated Fiber-Reinforced Mortar Incorporating Red Mud and Fly Ash

Abstract

Red mud (RM) and fly ash (FA) were used as a 30% replacement of cement in a sodium silicate-activated system. Composite mortar specimens with RM/FA ratios of 0:30, 1:5, 1:2, 1:1, and 2:1 were prepared with polypropylene fibers (PPF) for toughness enhancement. The mechanical properties and microstructure of the fiber-reinforced mortar were systematically investigated. The results showed that RM20F10 (RM/FA = 2:1) exhibited the best overall mechanical performance among all tested proportions. At this ratio, the 28-day compressive, flexural, and splitting tensile strengths reached 32.4 MPa, 7.3 MPa, and 4.2 MPa, exceeding the control mortar by 12.5%, 15.9%, and 23.5%, respectively. The RM/FA ratio of 1:1 achieved the highest 7-day flexural-to-compressive strength ratio. At 28 days, autogenous shrinkage increased from 910 με to 1100 με as the RM/FA ratio rose from 0:30 to 2:1, and all RM-containing specimens exhibited higher water absorption than the control mortar. Microstructural analysis by SEM, XRD, and FTIR revealed a denser matrix with reduced porosity, attributed to the synergistic formation of C–S–H, C–A–S–H, and N–A–S–H gels. RM reduced early-age porosity by promoting C–A–S–H gel formation, while FA facilitated late-age densification through delayed activation. PPF effectively bridged microcracks via fiber pull-out, leading to a ductile failure mode.

1. Introduction

Alkali-activated materials (AAMs) and the utilization of industrial solid wastes like fly ash (FA) and red mud (RM) have garnered significant attention as potential pathways for developing sustainable construction materials [1,2]. However, the large-scale application of AAMs is often hindered by issues such as high brittleness and significant shrinkage [3]. In contrast, the direct incorporation of FA and RM into ordinary Portland cement (OPC) systems presents a more readily applicable approach for valorizing these byproducts. Among these wastes, red mud (RM) is an alkaline solid waste produced during alumina refining, with 1.5–2.5 tons generated per ton of Al₂O₃ [4]. The global annual emissions of RM reach approximately 60 million tons, yet its utilization rate remains low at around 15% [5]. FA, a silica-alumina-rich solid waste from coal-fired power plants, is produced in massive quantities. The indiscriminate stockpiling of RM and FA occupies land and poses risks of atmospheric and water pollution [6]. Ordinary Portland cement (OPC) production requires significant energy and consumes non-renewable resources. To address these ecological concerns, researchers are developing sustainable alternatives that reduce cement use. In this context, utilizing RM and FA as supplementary cementitious materials (SCMs) can effectively alleviate CO₂ emissions while promoting the resource utilization of these industrial wastes [7]. A limited number of previous studies have investigated the incorporation of RM, either alone or in combination with FA, into mortar and concrete systems. For instance, Kumar Kuldeep et al. [8] replaced cement with RM (0–24%) while maintaining FA at 8%, demonstrating that RM acts as an effective strength modifier in the presence of FA, thereby reducing cement demand and supporting sustainable construction. Hu Cheng et al. [9] compared two types of red mud (A and B) incorporated at 10% and 20% in mortar, finding that red mud B provided higher flexural and compressive strengths due to its finer particle size and pore-filling effect. Similarly, Maddi et al. [10] reported that 10% RM replacement yielded optimal overall mortar performance, with microstructural analysis confirming enhanced densification compared with conventional mortar. The synergistic use of RM and FA in mortar can promote the formation of additional hydration products, such as C-A-S-H and C-S-H gels, which enhance the mechanical properties [11,12]. However, the relatively low reactivity of RM and FA often limits their hydration. Chemical activators, such as sodium hydroxide and sodium silicate, are sometimes employed to stimulate pozzolanic activity [13,14]. For instance, Ke et al. [15] studied a cementitious material system involving FA and RM and found that its 28-day compressive strength could reach 9 MPa. Venkatesh et al. [16] reported that a 10% RM content could enhance mortar strength through pore structure modification. Despite these efforts, there are still significant research gaps. First, the synergistic effects and optimal proportioning of RM and FA within OPC-based systems, especially when combined with chemical activators such as sodium silicate, are not fully understood. Second, although the incorporation of fibers is a recognized method for mitigating brittleness in cementitious composites [17], studies on the fiber-bridging effects and the corresponding micromechanisms within RM-FA-OPC ternary systems activated by sodium silicate are scarce. The random distribution of fibers within the mortar matrix can absorb and disperse stress, limit microcrack propagation, and significantly improve crack resistance [18]. Polypropylene fibers (PPF), known for their low cost, high chemical stability, and ease of application, are widely used in cementitious materials [19]. However, studies on the synergistic effects of RM and FA, activated by sodium silicate, in an OPC system reinforced with PPF are still scarce.

This study addresses the research gaps by incorporating red mud, fly ash, and polypropylene fibers into a Portland cement-based mortar activated by sodium silicate. The influence of red mud and fly ash on the mechanical properties was systematically investigated, and the matrix performance was optimized through compositional adjustment. The novelty of this work lies in three aspects: (1) systematic evaluation of the combined effect of RM/FA ratio (from 0:30 to 2:1) in this specific system; (2) identification of a balanced proportion that achieves mechanical performance exceeding ordinary mortar; and (3) elucidation of the microstructural mechanisms governing strength development. Multi-scale characterization (X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and scanning electron microscopy with energy dispersive spectroscopy (SEM-EDS)) was employed to elucidate the interfacial interaction mechanisms between fibers and hydration products. Furthermore, structure–property relationships linking macro-scale performance to microstructural evolution were established. It should be noted that, although sodium silicate is used as an activator, the system is not a purely alkali-activated material but rather a hybrid cementitious system (70% OPC with 30% RM/FA), where OPC remains the dominant binder. These findings provide theoretical guidance for solid waste utilization and the development of sustainable fiber-reinforced waste-based mortars.

2. Materials and Methods

2.1. Materials

The materials used for the preparation of the fiber-reinforced mortar included P·O 42.5 ordinary Portland cement (OPC), which conformed to the Chinese standard GB/T 175-2023 [20]. Class F fly ash (FA) was obtained from the Huaneng Shang’an Power Plant, (Shang’an District, Shijiazhuang, Hebei, China), and Bayer-process red mud (RM) was supplied by Yicheng Materials Co., Ltd. (Zibo, China). The chemical compositions of OPC, FA, and RM are listed in

Table 1. Chemical compositions of the RM and FA (wt.%).

Materials	SiO2	Al2O3	Fe2O3	CaO	Na2O	MgO	SO3	MnO	Loss in Ignition
OPC	19.1	3.5	5.0	65.5	0.1	1.4	3.9	0.28	-
RM	14.60	22.60	44.70	0.87	3.90	0.13	0.69	0.09	-
FA	43.00	25.50	10.50	9.60	1.06	3.15	1.50	0.15	3.10

, and their particle size distributions are shown in Figure 1. The manufactured sand had an apparent density of 2552 kg/m³ and a bulk density of 1345 kg/m³. Tap water was used for mixing the samples. Polypropylene fibers (PPF) with a length of 6 mm were used with a volume fraction of 0.1%, and their properties are listed in

Table 2. Geometric and mechanical properties of the PP fibers.

Diameter (μm)	Length (mm)	Tensile Strength (MPa)	Density (g/cm3)	Elastic Modulus (GPa)	Max. Elongation (%)
31	6	560	0.9	4.8	16

. The chemical activator was liquid sodium silicate (Tianjin Huasheng Chemical Co., Ltd., Tianjin, China), which was added at 10% by mass of the total binder (OPC + RM + FA). The sodium silicate had a modulus (SiO₂/Na₂O ratio) of 3.2, density of 1.36 g/cm³, Na₂O content of 8.5%, and SiO₂ content of 27.0%.

2.2. Mix Proportions

This study focuses on optimizing the RM/FA ratio within the hybrid system (OPC + RM + FA + sodium silicate + PPF). In this study, a water-cement ratio of 0.5 was used to investigate the effects of PPF, RM, and FA on the mechanical properties, shrinkage rate, water absorption rate, and mass loss of fiber-reinforced mortar under conditions where liquid sodium silicate was used as an alkali activator. Both RM and FA were used with a total replacement of 30% of the OPC by mass, as presented in

Table 3. Mixing proportions of the fiber-reinforced mortar.

Sample	RM (g)	FA (g)	Sodium Silicate Solution (g)	PPF (g)	OPC (g)	Water (g)	Sand (g)	Water/Cement Ratio
OM	0	0	0	0	660	330	1650	0.5
RM0F30	0	198	66	0.66	462	330	1650	0.5
RM5F25	33	165	66	0.66	462	330	1650	0.5
RM10F20	66	132	66	0.66	462	330	1650	0.5
RM15F15	99	99	66	0.66	462	330	1650	0.5
RM20F10	132	66	66	0.66	462	330	1650	0.5

. RM and FA were combined at different replacement ratios to form the following mixtures: 0:30, 1:5 (5:25), 1:2 (10:20), 1:1 (15:15), and 2:1 (20:10), which were named RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10, respectively. Ordinary mortar (OM) without RM, FA, PPF, or sodium silicate was prepared as a reference. A sodium silicate dosage of 10% of total binder was adopted, which effectively activates the pozzolanic reactions of FA and RM without compromising workability or causing excessively rapid setting. A PPF dosage of 0.1% by volume was adopted considering the workability of fiber-reinforced mortar.

2.3. Sample Preparation

The preparation process for the fiber-reinforced mortar specimens is shown in Figure 2: Step 1. OPC, FA, and RM were poured into the mixing pot and mixed at a low speed of (140 ± 5) rpm for two minutes. Step 2. Water and liquid sodium silicate were added to the mixing pot, and mixing was continued at the original speed for 3 min until the mixture reached a flowing state. Step 3. The PPF and sand were poured in and mixed at a high speed of (285 rpm ± 10) rpm for 5 min. Step 4. The fluidity test of the fiber-reinforced mortar was conducted immediately according to GB/T 17671-2021 [21] “Test Method for Strength of Cementitious Sand (ISO Method).” The mixture was molded into 40 mm × 40 mm × 40 mm, 70.7 mm × 70.7 mm × 70.7 mm, 25 mm × 25 mm × 285 mm, and 40 mm × 40 mm × 160 mm with three specimens each, vibrationally molded, and covered with plastic film. Step 5. Samples were demolded 24 h after casting, cured at (20 ± 2) °C and relative humidity (RH) ≥ 95%, and tested at the corresponding age.

2.4. Testing Methods

2.4.1. Compressive, Splitting Tensile, and Flexural Strengths

According to GB/T 17671-2021, the specimens cured for 3 days (3 d), 7 days (7 d), and 28 days (28 d) were tested for compressive, flexural, and splitting tensile strengths, and the results were the average of three specimens.

2.4.2. Saturated Water Absorption

Three 40 mm × 40 mm × 40 mm cubic specimens were selected for this test, and the water absorption test was conducted after 3 d and 28 d of maintenance, following the relevant provisions of ASTM C642 (2013) [22]. Finally, the water absorption (α) rate of each test piece was calculated using the following formula:

$$\alpha ={\displaystyle \frac{m_w-m_d}{m_d}}\times 100\%$$

(1)

where m_d is the specimen weight after a given immersion time in water (g), and m_w is the dry mass of the specimen after a constant mass was reached (g).

2.4.3. Drying Shrinkage and Autogenous Shrinkage

Effects of RM, FA, and PPF on the shrinkage of fiber-reinforced mortar were investigated according to ASTM C596 [23]. Fiber-reinforced mortar with dimensions of 25 mm × 25 mm × 285 mm was used to determine the drying shrinkage. When the specimen was demolded, it was immersed in water for 48 h and then moved to an environment with a temperature of 25 ± 2 °C and a humidity of 50 ± 4%. The initial length (L₀) was measured using a length meter. The autogenous shrinkage specimens were placed in an environment at 23 °C. The autogenous shrinkage of the specimens was measured using a φ 29 mm × 420 mm bellow, according to ASTM C1698 [24]. The lengths and masses of the drying and autogenous shrinkage specimens were recorded from 1 to 28 d. The drying shrinkage rate of the fiber-reinforced mortar at various ages was calculated using the following formula:

$$\lambda =\left(L_0-L_f\right)\times {\displaystyle \frac{{10}^6}{L_0}}$$

(2)

where λ is the drying shrinkage of the fiber-reinforced mortar (με), and L_f is the length of the fiber-reinforced mortar specimen block at the specified measurement time.

2.4.4. Microscopic Testing

The fragmented samples after compressive damage were removed by soaking in anhydrous ethanol for 72 h to terminate hydration and then placed in an oven (Model DHG-9030A, Electric Blast Constant Temperature Drying Oven, Shanghai Guangdi Instrument Equipment Co., Ltd., Shanghai, China) at 60 °C for 48 h before testing. After processing, a scanning electron microscope (SEM) (ZEISS Sigma 300, ZEISS, Oberkochen, Germany) and EDS were used to observe the microscopic morphology, elemental species, and fiber-reinforced mortar contents at 28 d. The net pulp samples maintained for 28 d were ground and sieved (particle size ≤ 0.75 μm), and the physical phases of the hydration products were determined using XRD (Rigaku Ultima IV, Rigaku Corporation, Tokyo, Japan). The chemical bonds of the hydration products were characterized using an infrared spectrometer (FTIR) (Thermo Scientific Nicolet iS20, Waltham, MA, USA) at room temperature.

3. Results and Discussion

3.1. Flowability

Figure 3 shows the flowability of the fiber-reinforced mortar. The OM mixture exhibited the highest flowability (210 mm), indicating good workability for plain cement mortar. The flowability of the composite mortars decreased markedly with the addition of RM, FA, PPF, and sodium silicate. A further reduction was observed as the RM-to-FA ratio increased. When the RM/FA ratio was 0:30 (RM0F30), the flowability was 192 mm; when the ratio increased to 2:1 (RM20F10), the flowability decreased to 145 mm, representing a 24.5% reduction relative to RM0F30 and a 30.9% reduction relative to OM. This reduction may be associated with (i) the irregular shape and high specific surface area of the RM particles, as shown in Figure 1, which increases interparticle friction and water demand; (ii) the inherent water absorption by RM; and (iii) the potential flow restriction imposed by the PPF network, though the individual effect of PPF was not directly tested in this study. The decreased FA content also reduces the beneficial “ball bearing” effect of spherical FA particles [25]. For practical applications, the reduced flowability (down to 145 mm at RM/FA = 2:1) may require the use of superplasticizers or adjustments in water content to maintain adequate workability.

3.2. Compressive Strength

The failure patterns and test results of the fiber-reinforced mortar in the compressive test are presented in Figure 4. As shown in Figure 4a, the mortar specimens with added PPF exhibited clean cracking without fragmentation at failure, whereas the OM specimens displayed a scattered failure state. Figure 4b shows that the OM mixture achieved its maximum compressive strength at 3 d and 7 d. At the early age of 3 d, mixtures with RM (e.g., RM20F10:21.7 MPa) showed higher strength than RM0F30 (13.6 MPa), which can be attributed to the combined alkaline activation from RM and sodium silicate, promoting early reactions. This may also be due to the relatively high fly ash content, where the amount of calcium hydroxide generated from cement hydration is insufficient to fully react with the fly ash, leaving a portion of the fly ash unreacted at this early stage [26].

By 7 d, RM0F30 achieved the highest compressive strength (23.9 MPa) among the PPF-reinforced specimens, surpassing RM20F10 by 6.7%. Extending the curing period from 3 d to 7 d resulted in relatively slow increases in the compressive strengths of RM5F25, RM10F20, RM15F15, and RM20F10 compared with RM0F30. Nevertheless, a notable strength inversion occurred at 28 d, with the RM20F10 mix exhibiting superior performance. Its compressive strength increased by 12.8%, 7.3%, 26.9%, 23.1%, and 40.1% compared with those of OM, RM0F30, RM5F25, RM10F20, and RM15F15. This inversion is attributed to the delayed activation of FA by the alkaline environment provided by RM and sodium silicate, which promotes additional gel formation at later ages. These results indicate that, for the specific mix design and curing regime employed here, the RM20F10 mixture (RM/FA = 2:1) achieved the highest compressive strength among the tested proportions. We propose that this proportion balances two mechanisms: sufficient RM provides alkali to activate FA, while excessive RM may hinder later-age densification. The filler effect of finer RM particles also promotes matrix densification and strength development [27]. The optimal proportion may vary with changes in raw materials, fiber type, or curing conditions.

3.3. Flexural Strength

Figure 5 shows the flexural strength and failure modes of the specimens tested. The OM mixture developed flexural strengths of 4.8 MPa, 5.9 MPa, and 6.3 MPa at 3 d, 7 d, and 28 d, respectively, exhibiting brittle fracture during failure. Compared to ordinary mortar specimens, the addition of PPF significantly improved the fracture behavior. Figure 5a,b demonstrate that the fiber-reinforced mortar specimens underwent flexural damage primarily through three sequential stages: crack emergence, expansion, and penetration. In fiber-reinforced mortars, the three typical distribution forms of PPF are fiber bridging, fracture, and pullout.

As depicted in Figure 5c, at the early age of 3 d, the flexural strength of all specimens containing PPF exhibited an increasing trend with age, with the enhancement magnitude correlating to the mixing ratio of RM to FA. At 7 d, the flexural strength of the RM0F30 specimen without RM addition exceeded that of specimens containing RM, second only to OM. At 28 d, the flexural strengths of the RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10 specimens were 7.0 MPa, 5.9 MPa, 6.4 MPa, 6.1 MPa, and 7.3 MPa, respectively. Compared with OM, these values changed by 11.1%, −6.3%, +1.6%, −3.2%, and 15.9%, respectively. This indicates that the RM20F10 specimen containing the combined RM, PPF, and FA exhibited the highest flexural strength. The RM20F10 specimen achieved a maximum flexural strength of 7.3 MPa at 28 d, outperforming ordinary cement mortar, confirming the synergistic reinforcement effect of RM, FA, and PPF. This synergistic enhancement can be ascribed to the combination of increased RM dosage, reduced FA dosage, elevated RM alkali content, and higher OH⁻ concentration in the mortar, which facilitates the dissolution of silica-alumina vitreous active components in FA. This process promotes the formation of AFt, C-S-H, and C-A-S-H gels, thereby enhancing the flexural strength [28].

3.4. Splitting Tensile Strength

Figure 6 presents the splitting tensile behavior of the fiber-reinforced mortar. In Figure 6a, fiber pullout or rupture consumes energy and redistributes stress, thereby delaying strength degradation and enhancing crack resistance. Examination of the fracture surface confirms that partially intact PPFs effectively bridged microcracks, highlighting their role in crack control. As depicted in Figure 6b, the splitting tensile strength of all fiber-reinforced mortars increased with curing age. The OM exhibited strengths of 2.2 MPa, 3.1 MPa, and 3.4 MPa at 3 d, 7 d, and 28 d, respectively. At 28 d, RM20F10 achieved the highest strength (4.2 MPa), representing a 23.5% improvement over OM, followed by RM0F30 at 4.0 MPa. In contrast, RM0F30 showed the lowest early strength (1.2 MPa at 3 d). Before failure, the load was mainly resisted by the bond between the hydration products and the PPF–matrix interface. With continued loading, degradation of the interfacial transition zone led to bond failure, after which stress was predominantly carried by the PPFs. Thus, the spatial orientation of the fibers significantly influenced both strength development and crack propagation in the mortar.

3.5. Flexural-to-Compression Ratio Analysis

Based on flexural and compressive strength data, the flexural-to-compressive ratios of different specimens at 7 d and 28 d are shown in Figure 7. As depicted in Figure 7, RM0F30 exhibited the lowest flexural-to-compressive ratio at both ages, followed by OM. At 7 d, RM20F10 showed the highest flexural-to-compressive strength ratio, exceeding those of OM and RM0F30 by 5.2% and 26.7%, respectively. By 28 d, RM15F15 achieved the maximum flexural-to-compressive strength ratio, surpassing OM and RM0F30 by 7.3% and 9.3%, respectively.

3.6. Drying Shrinkage and Mass Loss

Figure 8a indicates that the drying shrinkage of all fiber-reinforced mortars exhibited minimal growth from 1 d to 7 d, followed by a substantial increase from 7 d to 21 d, and slower growth from 21 d to 28 d. The OM mixture exhibited a 28 d drying shrinkage of 1750 με and mass loss of 5.1%. All composite mortars exhibited higher drying shrinkage than OM, with values for RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10 at 28 d being 1820 με, 1960 με, 2020 με, 1940 με, and 1900 με, respectively. RM10F20 showed the highest shrinkage, while RM0F30 was the closest to OM. Results in Figure 8b show mass loss rates of RM0F30, RM5F25, RM10F20, OM, RM15F15, and RM20F10 at 28 d were 5.4%, 5.6%, 5.8%, 5.1%, 5.0%, and 4.8%, respectively. From an application perspective, the increased shrinkage (up to 2020 με for RM10F20 vs. 1750 με for OM) should be considered, and measures such as shrinkage-reducing admixtures or proper curing may be necessary to mitigate cracking risk in field applications.

3.7. Autogenous Shrinkage and Mass Loss

Autogenous shrinkage primarily occurs owing to chemical shrinkage within the fiber-reinforced mortar, manifesting as an absolute volume reduction between the products and reactants during cement hydration. As shown in Figure 9a, the OM mixture exhibited 28 d autogenous shrinkage of 850 με and mass loss of 1.1%. The fiber-reinforced mortars exhibited higher autogenous shrinkage than OM. RM20F10 exhibited the highest autogenous shrinkage up to 7 d, reaching 790 με. Between 14 and 28 d, RM0F30 demonstrates the lowest autogenous shrinkage, with values at 28 d measuring 910 με, 1040 με, 1010 με, 1070 με, and 1100 με for RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10. The RM20F10 specimen containing RM exhibited the highest autogenous shrinkage, whereas RM0F30 without RM exhibited the lowest. The mass loss results shown in Figure 9b indicate that RM0F30 demonstrated the highest mass loss rate from 1 to 28 d. At 28 d, the mass loss rates for RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10 are 1.4%, 1.2%, 1.1%, 1.0%, 0.65%, and 0.85%, respectively.

3.8. Water-Absorbing Properties

Figure 10a,b present the water absorption of OM and fiber-reinforced specimens at 3 d and 28 d. OM exhibited the lowest absorption at both ages, which can be attributed to two factors: (i) the rapid formation of a dense C–S–H gel matrix during ordinary Portland cement hydration, and (ii) the absence of PPF, which avoids the additional interfacial transition zones (ITZs) that fibers introduce into the matrix. In contrast, the incorporation of solid waste initially increased capillary porosity, especially in RM0F30, which exhibited the highest absorption at 3 d owing to the limited early participation of FA in hydration. With the addition of red mud (RM), early-age absorption decreased relative to RM0F30, as RM possesses higher reactivity and rapidly reacts with sodium silicate to form amorphous C–S–H and C–A–S–H gels, filling the pores and enhancing the density. By 28 d (Figure 10b), RM0F30 displayed lower absorption than RM-blended mixes, indicating a sustained activation of FA by the alkali solution. Unreacted FA particles release Si⁴⁺ and Al³⁺, promoting the formation of C–A–S–H and N–A–S–H gels through ion substitution, which further densify the matrix and reduce porosity. It can be seen from Figure 10c that the saturated water absorption of fiber-reinforced mortars decreased with increasing RM/FA ratio at the 3 d curing age. At 28 d, among the composite mortars, RM0F30 exhibited the lowest saturated water absorption rate, whereas RM5F25 exhibited the highest.

3.9. Microstructural Analysis

3.9.1. XRD

Figure 11 illustrates the XRD spectra of RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10 net pulp samples at 28 d. The diffraction peaks of the five samples exhibited similar patterns, with primary crystalline phases including calcite (CaCO₃), alite (Ca₃SiO₆), belite (Ca₂SiO₄), and quartz (SiO₂). The hydration products of RM0F30, RM5F25, RM10F20, and RM15F15 exhibited more pronounced diffraction peaks than those of RM20F10. Notably, all five sample groups displayed a distinct dispersion peak within the 25–35° range in the XRD spectra, indicating the presence of an amorphous gel phase in the hydration products. This phenomenon is attributed to the dissolution of reactive Si-Al substances in FA and RM under alkaline conditions, wherein Al³⁺ and Si⁴⁺ react with [SiO₄] and [Al(OH)₄]⁻, Ca²⁺, SO₄²⁻, and Na⁺ in the mortar matrix to form C-S-H and C(N)-A-S-H gels, subsequently creating an amorphous gel phase [29,30]. Using the OriginPro 2024b software for calculation, it was determined that the samples RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10 exhibited amorphous gel phase contributions (estimated by the integrated peak area in the 25–35° range) of 17.1%, 16.7%, 15.0%, 14.5%, and 18.2%, respectively. Since the mechanical properties of the fiber-reinforced mortars are positively correlated with the polymerization degree of the hydration products [31], the higher amorphous gel contribution of RM20F10 (18.2%) is consistent with its superior 28 d compressive and splitting tensile strength. It should be noted that the reported amorphous gel contributions reflect the combined influence of raw material proportion differences and reaction products, as varying the FA/RM ratio leads to different baseline crystallinity values in the unreacted mixtures. Therefore, these values are best interpreted as comparative indicators rather than absolute measures of reaction extent.

The incorporation of liquid sodium silicate, combined with RM’s higher alkalinity of RM, accelerates cement hydration. The fiber-reinforced mortar exhibited both cement hydration and alkali activation reactions, resulting in the rapid formation of hydration products, including C-S-H, AFt, Ca(OH)₂, and aluminosilicate gel (N-A-S-H). Consequently, the early-stage mechanical properties of RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10 generally improved with increasing RM content. In the later stages, under RM and liquid sodium silicate activation, the continuous hydration of C₃S and C₂S, combined with the complete dissolution of previously unreacted FA and active oxides in the RM, led to extensive ionic combinations and reactions, generating substantial amounts of AFt and C-S-H gels [32].

3.9.2. FTIR

The chemical bonding of the five groups of mortar specimens was characterized using FTIR, and the results are shown in Figure 12. The absorption peak at 3445 cm⁻¹ corresponds to the O-H bond asymmetric stretching and bending vibrations; carbonate C-O bonding characteristic peaks appear at 1651 cm⁻¹ and 1426 cm⁻¹, which, combined with Ca(OH)₂ presence in XRD, indicates carbonate formation through Ca(OH)₂ carbonation [33]; [SiO₄]⁴⁻ or [AlO₄]⁻ binding Si-O-Si(Al) characteristic peaks were identified between 800~1200 cm⁻¹ [34]; absorption peaks near 976 cm⁻¹ result from Si-O-Si and Si-O-Si-O bonds in Al due to asymmetric stretching vibrations, primarily from FA reactions [35]. The Si-O vibration peak in RM20F10 shifted from 970 cm⁻¹ to 976 cm⁻¹, indicating that the Si-O bond energy was enhanced, which helped improve its structural stability. RM doping enhances hydration reaction rates, with peak position and intensity reflecting gel quality and structure from alkali excitation reactions; the 460 cm⁻¹ characteristic peak relates to Si-O-Si (Al) bending vibration [36].

Figure 13 shows the fitting results of the resonance peak SiOⁿ in the range of 800 cm⁻¹ to 1200 cm⁻¹ for five mortar samples. The five mortar groups correspond to SiQ⁰, SiQ¹, SiQ², SiQ³, and SiQ⁴ at 850 cm⁻¹, 950 cm⁻¹, 1000 cm⁻¹, 1050 cm⁻¹, and 1100 cm⁻¹, respectively, where n in Qⁿ represents the number of coordinating bridge oxygen atoms around Si, and Qⁿ represents the relative area of the corresponding resonance peak [37]. By analyzing the relative changes in the areas of these characteristic peaks, the formation and transformation processes of aluminosilicate gel in the mortar matrix can be studied. Since changes in the number of bridging oxygen atoms can reflect the relative extent of polymerization or depolymerization reactions in the system, the relative bridging oxygen (RBO) value serves as an indicator of the degree of polymerization of the aluminosilicate gel network, with higher RBO values suggesting a more crosslinked, compact gel structure. The polymerization degree of SiQⁿ can be calculated using the RBO Formula (3) [38], with the results shown in

Table 4. The relevant parameters of Si-O-Si in RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10.

Sample	Relative Content/%					RBO	R2
	SiQ0	SiQ1	SiQ2	SiQ3	SiQ4
RM0F30	5.14	44.26	25.32	1.73	21.55	47.52	0.97
RM5F25	4.30	55.99	17.29	3.07	19.29	44.26	0.99
RM10F20	4.17	50.82	21.54	3.29	20.18	46.12	0.98
RM15F15	4.81	49.73	19.03	3.09	22.31	47.01	0.98
RM20F10	5.70	44.65	19.08	20.35	14.20	48.23	0.97

. It can be seen that the order of RBO values is RM20F10 > RM0F30 > RM15F15 > RM10F20 > RM5F25. Among these, RM0F30 and RM20F10 have relatively higher polymerization degrees, with values of 47.52 and 48.23, respectively. The exceptionally high RBO value of 48.23 for RM20F10 indicates its highly crosslinked gel structure, which is consistent with its superior mechanical properties. This consistency across multiple samples supports the validity of the RBO analysis as a comparative, rather than absolute, measure of polymerization degree.

$$RBO={\displaystyle \frac{1}{4}}\times {\displaystyle \frac{\mathsf{\Sigma }n\times Q^n}{\mathsf{\Sigma }{Q}^n}}$$

(3)

3.9.3. SEM-EDS

Figure 14a shows the SEM images and EDS spectra of RM0F30, RM5F25, RM10F20, RM15F15, and RM20F10 at 28 d. The EDS data were obtained by area scanning (map scanning) of the entire field of view in each SEM image.

Table 5. Results of elemental distribution in SEM images of samples.

Sample	C	O	Si	Na	Fe	Ca	Ca/Si
OM	7.93	48.20	11.48	-	2.23	24.12	2.10
RM0F30	7.48	43.27	12.04	1.55	2.27	25.72	2.14
RM5F25	7.55	35.92	13.08	1.49	5.08	31.43	2.40
RM15F15	6.10	37.49	13.07	3.35	8.72	29.45	2.25
RM20F10	7.84	36.30	12.30	2.22	8.11	25.57	2.08

presents the elemental content of the EDS spectra corresponding to the SEM and the Ca/Si ratio. As observed in Figure 14a, the OM matrix exhibits a relatively dense structure with no significant fatal defects. The gel phases in the SEM images of most fiber-reinforced mortars appear in clusters. For instance, numerous hydration products exist on the surface of needle-and-stick AFt in RM0F30, while RM5F25 exhibits abundant flocculent C-S-H gel and needle-and-stick AFt. RM0F30 had a high water absorption rate at 3 d due to the presence of many unreacted FA particles and loose gel distribution. However, continuous hydration reactions at 28 d led to structural densification, which was consistent with the downward trend in water absorption rate and upward trend in strength. Substantial gel-like C-S-H and needle-and-stick AFt appear on the FA particles’ surface in RM15F15, attributed to the FA particles being closely encapsulated by highly polymerized C-S-H gels after alkali activation. RM20F10 contains a large amount of hydration products, which are closely interlocked with each other, filling the harmful pores in the structure, reducing harmful voids, and making the microstructure more compact, corresponding to its low water absorption rate and high strength at 28 d.

From the EDS results of the five groups of fiber-reinforced mortar in Table 5, it can be seen that OM, RM0F30, RM5F25, RM15F15, and RM20F10 primarily contain C, O, Si, Ca, and Na elements, with relatively high O, Si, and Ca content, indicating that the gel in fiber-reinforced mortar is primarily C-S-H. It can be observed that the Ca/Si ratio among the five groups of fiber-reinforced mortar follows the order: RM5F25 > RM15F15 > RM0F30 > OM > RM20F10. As the RM content increases, the Ca/Si ratio decreases, leading to a reduction in the concentration of dissolved Ca²⁺ in the fiber-reinforced mortar system. As a result, only a small portion of the Na⁺ ions in the N-A-S-H gel formed by FA can be replaced by Ca²⁺, resulting in a decrease in the formation of C-S-H gel and an increase in N-A-S-H gel.

The presence of unreacted FA particles and pores in fiber-reinforced mortar results in weak interfacial transition zones (ITZs) between PPF and mortar matrix [39]. Figure 14b reveals the distribution characteristics and failure mechanisms of PPF within the fiber-reinforced mortar matrix. Under external loading conditions, the PPFs predominantly exhibited tensile fracture and pull-out failure modes. This fiber pull-out-dominated failure mode indicates superior ductility in the fiber-reinforced mortar system, consistent with the observations by Wang et al. [40]. Further microscopic examination reveals that only limited hydration products adhere to the PPF surfaces, with pull-out failure being the predominant mode. This observation has provided microstructural evidence that the relatively weak interfacial bonding between PPFs and the mortar matrix fundamentally causes these failure patterns. More significantly, when fiber-reinforced mortar fractured, the PPFs experienced stretching and twisting during loading, increasing the effective fiber-matrix contact area. As macroscopic cracks propagated to PPFs, these deformed fibers altered crack paths through mechanical interlocking and induced secondary microcracks, thereby substantially enhancing the composite’s energy dissipation capacity. These phenomena indicate that the PPF not only provides bridging effects but also absorbs energy during specimen damage, effectively inhibiting crack initiation and propagation, thereby enhancing the fiber-reinforced mortar toughness [41]. SEM observations in this study are primarily qualitative. Additionally, the samples were obtained from fractured specimens after strength testing, which may introduce cracks that are not representative of the original material structure. Quantitative techniques such as mercury intrusion porosimetry (MIP) are needed to directly link pore structure to mechanical performance, and will be incorporated in future work.

3.10. Alkali Activation Mechanism

The specific process of this reaction mechanism is shown in Figure 15. Liquid sodium silicate establishes an alkaline environment for fiber-reinforced mortar by releasing Na⁺ and OH⁻ ions. These alkaline conditions facilitate the dissolution of Si and Al oxides in the raw material by breaking the Si-O and Al-O bonds. During the initial strength formation phase, cement hydration in the fiber-reinforced mortar generates C-S-H gel for strength development while simultaneously producing Ca(OH)₂ that releases Ca²⁺. Additionally, CaO in RM reacts with water to form Ca(OH)₂, which provides additional Ca²⁺. Some Ca²⁺ reacts with silica–oxygen tetrahedral monomers to form C-S-H gel, while the increased concentration of silica–oxygen and aluminum–oxygen tetrahedra in the liquid phase forms silica-aluminate oligomers such as Al(OH)₄⁻ through activation by liquid sodium silicate’s Na⁺ and OH⁻ [42]. During the intermediate hydration stage, C₃S and C₂S continue hydrating, while Al₂O₃ in RM and SiO₂ in FA are activated to form [Al(OH)₄]⁻ and SiO₃²⁻, respectively. The activated Al₂O₃ and SiO₂ subsequently react with Ca(OH)₂ to produce C-S-H and C-A-H gels. The system’s high Ca²⁺ and Na⁺ content enables free Ca²⁺ to combine with Al/Si monomers, forming interconnected three-dimensional C-A-S-H gels, while Na⁺ forms N-A-S-H gels. This process yields hydration products, including AFt and Ca(OH)₂, which fill structural pores. In the final hydration stage, C-S-H, C-A-S-H, and N-A-S-H gels progressively solidify and harden over time, filling matrix voids, enhancing pore structure density [43], and substantially improving the compressive, flexural, and split tensile strengths of the specimens [44]. In summary, the reaction equation involved in fiber-reinforced mortar can be expressed as follows [45]:

C₃S + H₂O → CH + C-S-H

(4)

C₂S + H₂O → CH + C-S-H

(5)

Na₂SiO₃ → 2Na⁺ + SiO₃²⁻ + OH⁻

(6)

SiO₂ + 2OH⁻ → SiO₃²⁻ + H₂O

(7)

Al₂O₃ + 2OH⁻ + 3H₂O → 2[Al(OH)₄]⁻

(8)

Ca²⁺ + SiO₃²⁻ + nH₂O → C-S-H

(9)

Ca²⁺ + [Al(OH)₄]⁻ + mH₂O → C-A-H

(10)

Na⁺ + Ca²⁺ + [Al(OH)₄]⁻ + SiO₃²⁻ + H₂O → C(N)-A-S-H

(11)

4. Conclusions

Based on the above study, the following conclusions can be drawn:

(1)

The incorporation of RM, FA, PPF, and sodium silicate reduced the flowability of fiber-reinforced mortar compared to OM. Flowability decreased with increasing RM/FA ratio.

(2)

Within the specific material system and curing conditions of this study, the fiber-reinforced mortar with an RM/FA ratio of 2:1 (RM20F10) demonstrated the best overall mechanical performance, achieving 28 d compressive, flexural, and splitting tensile strengths of 32.4 MPa, 7.3 MPa, and 4.2 MPa, respectively. These values represent improvements of 12.5%, 15.9%, and 23.5% over ordinary mortar.

(3)

The fiber-reinforced mortars generally exhibited higher drying and autogenous shrinkage than OM, with the maximum values observed for the RM10F20 and RM20F10 mixtures, respectively.

(4)

Microstructural analyses confirmed that the optimal mix (RM20F10) exhibited the highest gel product content and a favorable Ca/Si ratio, leading to a denser matrix. This is considered a key factor contributing to its superior compressive strength. Furthermore, SEM analysis reveals that PPF effectively bridges microcracks and that pull-out is the dominant failure mechanism. This mechanism is the primary reason for the significantly enhanced toughness, flexural, and splitting tensile strengths.

Future studies should extend the curing period beyond 28 days and include durability tests such as sulfate resistance and chloride permeability to further validate the long-term performance of this hybrid system.