Performance and Microstructural Characteristics of Ultra-Early High-Strength Cement-Based Grouting Materials Modified with Accelerating and Retarding Agents

Abstract

To balance ultra-early strength development and workable time in cement-based grouting materials for rapid repair applications, an ultra-early high-strength grout system was developed by regulating the dosage of an accelerating agent (CF), retarder content, and water-to-binder ratio (w/b). The effects of these parameters on setting behavior, workability, mechanical properties, volumetric stability, and durability were systematically investigated. X-ray diffraction (XRD) and scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM/EDS) were further conducted to qualitatively evaluate the hydration characteristics and microstructural evolution of the optimized system. The results showed that CF accelerated early hydration and promoted the rapid formation of ettringite (AFt), which contributed to the development of ultra-early strength. The incorporation of a retarder effectively prolonged the workable time and improved slurry workability. Increasing the w/b ratio enhanced flowability and toughness, although excessive w/b reduced compressive strength. The optimal mixture contained 30% CF, 0.02% retarder, and a w/b ratio of 0.19. Under this condition, the grout exhibited a flowability of 312 mm and compressive strengths of 81.4 MPa at 1 h and 121.3 MPa at 28 d. In addition, low air shrinkage (0.027% at 28 d) and excellent chloride penetration resistance (12 C at 28 d) were achieved. Microstructural observations suggested that the dense structure formed by AFt and C–S–H gel contributed to the improved macroscopic performance. This study provides an engineering-oriented reference for the mix design and performance optimization of ultra-early high-strength cement-based grouting materials for rapid repair applications.

1. Introduction

Ultra-early high-strength cement-based grouting materials have been widely used in rapid repair engineering, emergency reinforcement, bridge maintenance, tunnel repair, and prefabricated structure connections due to their rapid setting, high early strength, and convenient construction characteristic [1,2]. Compared with conventional cementitious materials, ultra-early high-strength grouts can significantly shorten construction periods and reduce traffic interruption time, which is particularly important for emergency repair and infrastructure maintenance applications [3,4]. Therefore, the development of high-performance ultra-early high-strength cement-based grouting materials has attracted increasing attention in recent years [5].

The performance of ultra-early high-strength cement-based grouting materials is closely related to the hydration kinetics and microstructural evolution of the cementitious system [6,7]. Accelerating agents are commonly incorporated to promote the rapid hydration of aluminate and silicate phases, thereby significantly improving early-age strength development [8]. In particular, the rapid formation of ettringite (AFt) has been considered one of the primary reasons for rapid setting and high early strength in rapid-hardening cementitious materials [9,10]. However, excessive acceleration of hydration may also lead to insufficient workable time, uneven microstructure development, internal stress concentration, and reduction in later-age performance [11]. Therefore, balancing ultra-early strength development and workable time remains a major challenge in the design of rapid-repair cement-based grouting materials [12].

To improve workable time, retarders are often introduced to regulate the hydration process and delay the rapid precipitation of hydration products [13,14]. Previous studies have shown that retarders can prolong setting time mainly through adsorption and calcium complexation effects, thereby reducing the hydration rate at early ages. Meanwhile, an appropriate retardation effect may contribute to more uniform hydration and improved later-age microstructure [15]. However, excessive retarder incorporation may significantly suppress early hydration and reduce early-age strength development [16]. In addition, the interaction between accelerating agents and retarders in ultra-early high-strength grout systems has not yet been fully clarified, especially regarding their combined influence on hydration behavior and microstructural evolution [17].

The water-to-binder ratio (w/b) is another key factor affecting the rheological behavior, pore structure, hydration process, and mechanical properties of cement-based grouting materials [18]. A lower w/b ratio is generally beneficial for improving strength and compactness, whereas excessively low w/b may deteriorate workability and reduce structural uniformity [19]. In contrast, increasing the w/b ratio can improve flowability and facilitate construction but may also increase porosity and reduce compressive strength [20,21]. Therefore, optimizing the w/b ratio is essential for achieving a balance among workability, strength development, and durability [22].

Although considerable research has been conducted on rapid-hardening cementitious materials, most previous studies mainly focused on individual factors such as accelerating agents, mineral admixtures, or expansive components [23,24]. Systematic investigations on the synergistic regulation of accelerating agents, retarders, and w/b ratio in ultra-early high-strength cement-based grouting systems are still limited [25]. In particular, the relationship between macroscopic performance and microstructural evolution has not yet been fully understood [26,27].

In this study, an ultra-early high-strength cement-based grouting material was developed by incorporating an accelerating agent (CF) and a retarder into a Portland cement-based system [28]. The effects of CF replacement ratio, retarder dosage, and w/b ratio on flowability, setting behavior, mechanical properties, volumetric stability, and durability were systematically investigated [29]. Furthermore, X-ray diffraction (XRD) and scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM/EDS) were employed to analyze the hydration products and microstructural evolution of the optimized system [30]. The objective of this study is to provide experimental and microstructural insights into the design and optimization of ultra-early high-strength cement-based grouting materials for rapid repair applications.

2. Experimental Program

2.1. Material

Unless otherwise specified, all raw materials and chemical admixtures used in this study were supplied by SINO-SINA Building Materials Co., Ltd. (Zhengzhou, China). The raw materials used in this study mainly included ordinary Portland cement (PO, P·O 52.5 grade), an accelerating agent (CF), silica fume (SF), fly ash (FA), quartz sand, a water-reducing agent, a retarder, a defoaming agent, a plastic expansion agent, and mixing water. The cement used was commercially available P·O 52.5 ordinary Portland cement. CF was a commercially available accelerating admixture used to promote early hydration and enhance early-age strength development. Silica fume (SF) and fly ash (FA) were used as mineral admixtures. The specific surface area of SF was 16 m²/g. The FA used in this study was Class I fly ash according to GB/T 1596-2017 [31]. Quartz sand was used as the fine aggregate, with particle sizes of 20–40 mesh and 40–70 mesh. The superplasticizer was a polycarboxylate-based high-range water reducer in powder form, with a water-reduction rate greater than 25%. The retarder used was analytically pure tartaric acid. The defoamer was a polyether-based surfactant in powder form, while the expansive agent was an azo-based expansive additive in powder form. Ordinary tap water was used as the mixing water. CF was incorporated as a partial replacement for Portland cement (PO) on a mass basis. The mix proportions of grout mixtures with different CF replacement ratios are presented in

Table 1. Mix proportions of grout materials with different CF dosages (kg).

PO	CF	SF	FA	Quartz Sand (20–40 Mesh)	Quartz Sand (40–70 Mesh)	Superplasticizer	Retarder	Water
450	0	30	20	280	220	5	0.5	95
400	50	30	20	280	220	5	0.5	95
350	100	30	20	280	220	5	0.5	95
300	150	30	20	280	220	5	0.5	95

. The total binder content (PO + CF + SF + FA) was maintained constant for all mixtures. The water-to-binder ratio (w/b) was calculated based on the total mass of cementitious materials.

2.2. Experimental Design and Methods

This study systematically investigated the effects of key parameters, including CF dosage, retarder dosage, and water-to-binder ratio (w/b), on the flowability, setting behavior, and mechanical properties of ultra-early high-strength grouting materials at different ages. In addition, the volumetric stability and durability of the optimized grout system were further evaluated. X-ray diffraction (XRD) and scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM/EDS) were conducted to qualitatively characterize the hydration products and microstructural characteristics of the material. The microstructural observations were used to interpret the evolution of macroscopic performance.

Based on preliminary experimental investigations, the basic mix proportion of the grouting material was determined. The binder-to-sand ratio was fixed at 1:1. Quartz sand with double gradation was used as the fine aggregate. The proportion of 20–40 mesh particles was 56%, while the remaining fraction consisted of 40–70 mesh particles. In the reference binder system, PO, SF, and FA accounted for 90%, 6%, and 4% of the binder mass, respectively. The dosages of the chemical admixtures were 1% superplasticizer, 0.10% retarder, 0.10% defoamer, and 0.02% expansive agent by binder mass. The water-to-binder ratio (w/b) was 0.19. Based on this reference mixture, three groups of comparative experiments were designed to systematically investigate the effects of CF replacement ratio, retarder dosage, and w/b ratio on the performance of the grouting material.

(1)

Effect of CF replacement ratio: Under identical testing conditions, the CF replacement ratio was set at 0%, 10%, 20%, 30%, and 40% to evaluate its effects on the setting behavior, flowability, and mechanical properties of the grouting material.

(2)

Effect of retarder dosage: In the preliminary experiments, the mixture without retarder exhibited excessively rapid setting behavior and insufficient workable time after CF incorporation. Therefore, the retarder dosage investigation focused on mixtures containing 0.02%, 0.06%, 0.10%, 0.14%, and 0.18% retarder to ensure practical operational time for engineering applications. Based on the optimal CF dosage determined in Scheme (1), the influence of retarder dosage on the setting behavior and strength development of the grout was systematically investigated.

(3)

Effect of water-to-binder ratio (w/b): After determining the optimal CF replacement ratio and retarder dosage, the w/b ratio was adjusted from 0.18 to 0.22 to evaluate its influence on flowability, mechanical properties, and toughness.

After determining the optimal mixture proportion, the volumetric stability and durability of the grout system were further evaluated through shrinkage and electric flux tests. In addition, XRD and SEM/EDS analyses were conducted to qualitatively characterize the hydration products and microstructural characteristics of the optimized system. The microstructural observations were used to interpret the corresponding macroscopic performance.

The ultra-early high-strength grouting material was prepared using a laboratory-scale planetary mortar mixer. First, PO, SF, FA, quartz sand, and the chemical admixtures, including CF, superplasticizer, retarder, defoamer, and expansive agent, were dry-mixed to ensure uniform dispersion of all solid components. Subsequently, the predetermined amount of mixing water was added. The mixture was first stirred at low speed for 210 s to obtain an initial homogeneous slurry. High-speed mixing was then conducted for an additional 90 s to further improve the uniformity and flowability of the grout.

The initial flowability of the ultra-early high-strength grouting material was measured in accordance with the Technical Specification for Application of Cementitious Grouting Materials (GB/T 50448-2015) [32]. The test was conducted using a truncated cone mold with a top diameter of 70 mm, a bottom diameter of 100 mm, and a height of 60 mm. The mold was placed on a smooth horizontal glass plate [33], as shown in Figure 1a. Before testing, the glass plate and mold were thoroughly cleaned. A small amount of water was sprayed onto the glass plate to maintain a moist surface without visible free water. The freshly mixed grout was then poured into the mold in a single operation without layering. The top surface was gently leveled with a trowel, and excess material was removed to ensure that the slurry surface was flush with the upper edge of the mold. After filling, the mold was vertically lifted within approximately 10 s to minimize disturbance to the slurry flow. The grout then spread freely under its own weight on the glass plate. After the flow stopped, the maximum spread diameters in two perpendicular directions were measured. The average value was taken as the flowability of the grout. Each test was repeated at least twice, and the average value was reported as the final result.

The setting time of the ultra-early high-strength grouting material was determined in accordance with the Test Methods for Basic Properties of Building Mortar (JGJ/T 70-2009) [34]. The test was conducted using a mortar penetration resistance apparatus, as shown in Figure 1b. The testing environment was maintained at a temperature of (20 ± 2) °C and a relative humidity of no less than 50%. Freshly mixed grout was immediately poured into the prepared mold. The specimen was then lightly vibrated and surface-leveled to ensure adequate compaction and a smooth surface. After filling, the mold was placed beneath the penetration resistance apparatus. The penetration needle was adjusted to just contact the specimen surface, and this moment was recorded as the starting time of the test. Penetration measurements were subsequently conducted at regular time intervals, generally every 5 min. For mixtures exhibiting rapid setting behavior, the testing interval was appropriately shortened. During each measurement, the needle was allowed to penetrate freely, and the penetration depth was recorded. As hydration proceeded, the penetration depth gradually decreased. When the penetration depth reached approximately (4 ± 1) mm from the bottom of the mold, the material was considered to reach the initial setting state. The test was then continued until the needle left only a slight mark on the specimen surface or could barely penetrate the grout. This condition was defined as the final setting state. The elapsed time from water addition to the initial and final setting states was defined as the initial setting time and final setting time, respectively. Each test was repeated at least twice, and the average value was reported as the final result.

The flexural and compressive strengths of the ultra-early high-strength grouting material were measured in accordance with the Method of Testing Cements—Determination of Strength (ISO Method) (GB/T 17671-2021) [35]. Mechanical tests were conducted using an HYE-300F-D universal testing machine (Hebei Sanyu Testing Machine Co., Ltd., Cangzhou, China). Freshly mixed grout was poured into a three-compartment mold with dimensions of 40 mm × 40 mm × 160 mm. The mold was filled in two layers. After placing each layer, the mold was vibrated on a vibrating table for approximately 60 vibrations to ensure adequate compaction. After vibration, the specimen surface was leveled with a trowel to make it flush with the top of the mold, and excess grout was removed. After casting, the specimens were stored at a temperature of (20 ± 2) °C and a relative humidity of at least 90% for approximately 24 h before demolding. The demolded specimens were then cured in water at (20 ± 1) °C until the designated testing ages. Flexural and compressive strength tests were conducted at 1 h, 3 h, 1 d, 3 d, 7 d, and 28 d. For ultra-early-age specimens, appropriate curing conditions were adopted according to the experimental requirements, and testing was performed immediately after the target age was reached. The flexural strength test was performed using a three-point bending method. During testing, the specimen was placed on the supports of the flexural testing machine and loaded at a specified rate until failure, as shown in Figure 1c. The failure load was recorded and used to calculate the flexural strength. The two fractured halves obtained from the flexural test were subsequently used for compressive strength testing. The specimens were placed in a compression testing machine and loaded uniformly until failure, as shown in Figure 1d. The failure load was recorded and used to calculate the compressive strength. At least three specimens were prepared for flexural testing and six specimens for compressive strength testing in each group. The reported results represent the average values.

The volumetric stability of the ultra-early high-strength grouting material was evaluated using the shrinkage test method specified in the Test Methods for Basic Properties of Building Mortar (JGJ/T 70-2009) [34]. Freshly mixed grout was poured into shrinkage molds with dimensions of 40 mm × 40 mm × 160 mm and filled in layers [36]. After placing each layer, slight vibration was applied to ensure adequate compaction and to remove visible air bubbles. After casting, the specimen surface was leveled with a trowel. Measuring studs (gauge points) were then installed at both ends of the specimen for subsequent length-change measurements, as shown in Figure 1e. The specimens were stored under standard curing conditions at (20 ± 2) °C and a relative humidity of at least 90%. The specimens were demolded after approximately 1 h. Immediately after demolding, the initial length of each specimen was measured and defined as the reference length. The specimens were subsequently subjected to two curing conditions: (1) water curing at (20 ± 2) °C and (2) air curing at (20 ± 2) °C and a relative humidity of (60 ± 5)%. Length changes were measured at 1 d, 3 d, 7 d, and 28 d. The shrinkage rate was calculated based on the measured length variation at each age, and the corresponding equation is given in Equation (1). At least three specimens were prepared for each test group, and the reported results represent the average values.

$$\epsilon ={\displaystyle \frac{L_{\mathrm{t}}-L_0}{L_0}}\times 100\%$$

(1)

where

ε—shrinkage rate (%);

L₀—initial length of the specimen (mm);

L_t—length of the specimen at age t (mm).

The chloride ion permeability of the ultra-early high-strength grouting material was evaluated using the rapid chloride permeability test (RCPT) in accordance with the Standard for Test Methods of Long-Term Performance and Durability of Ordinary Concrete (GB/T 50082-2009) [37]. The electric flux test was conducted using a rapid chloride permeability testing apparatus manufactured by Beijing Lugongjian Instrument Technology Co., Ltd. (Beijing, China). The testing apparatus is shown in Figure 1f. Freshly mixed grout was cast into cylindrical specimens with dimensions of φ 100 mm × 50 mm. After vibration and compaction, the specimens were cured under standard conditions until the designated testing age, generally 28 d. Before testing, both ends of the specimens were ground to obtain smooth and flat surfaces and then cleaned thoroughly. The specimens were subsequently subjected to vacuum saturation treatment. A vacuum of no less than −0.08 MPa was first applied for approximately 3 h. Deionized water was then introduced under vacuum and maintained for about 1 h. The specimens were further immersed in water for 18 h to ensure full saturation. After saturation, the specimens were mounted in the RCPT testing cell. One side of the specimen was exposed to 3.0% NaCl solution, while the opposite side was exposed to 0.3 mol/L NaOH solution. Proper sealing was ensured throughout the test to prevent leakage. A constant DC voltage of 60 V was applied across the specimen for 6 h. During the test, the current passing through the specimen was recorded at regular intervals, typically every 30 min. The total charge passed during the 6 h test period was calculated from the current–time curve and expressed in coulombs (Cs). The obtained electric flux value was used to evaluate the resistance of the material to chloride ion penetration. At least three specimens were tested for each group, and the reported results represent the average values.

To further elucidate the intrinsic mechanism underlying the performance evolution of the ultra-early high-strength grouting material, X-ray diffraction (XRD), scanning electron microscopy (SEM), and energy-dispersive spectroscopy (EDS) were employed to characterize the hydration products and microstructural evolution of the material [38]. XRD analysis was conducted using a Rigaku Smart Lab SE diffractometer (Rigaku Corporation, Tokyo, Japan). SEM and EDS analyses were performed using a Sigma 300 field-emission scanning electron microscope (Carl Zeiss AG, Oberkochen, Germany). Prior to testing, the specimens were crushed, and internal fragments were selected for analysis. The hydration reaction was terminated using anhydrous ethanol, followed by drying and grinding to prepare samples suitable for microstructural characterization. XRD was primarily used to identify the phase composition of the hydration products and analyze their evolution characteristics at different curing ages. SEM was employed to observe the morphology and compactness of the microstructure, while EDS was used to qualitatively analyze the elemental composition of the selected regions.

3. Experimental Results Analysis

3.1. Effect of Cf Dosage on Workability and Mechanical Properties

As shown in Figure 2a, the flowability of the grout increased slightly from 315 mm to 320 mm as the CF replacement ratio increased from 0% to 40%. The overall variation was relatively small, indicating that CF had a limited influence on the initial rheological behavior of the grout within the investigated dosage range. In addition, no obvious adverse effect on flowability was observed. The slight increase in flowability may be attributed to the influence of CF on particle dispersion and early hydration behavior. As illustrated in Figure 2b, the setting time decreased significantly with increasing CF replacement ratio. Compared with the control mixture (235 min), the setting time decreased to 50 min at a CF replacement ratio of 10%, corresponding to a reduction of approximately 79%. When the CF replacement ratio further increased to 40%, the setting time remained nearly constant at approximately 44–45 min. This result indicates that the accelerating effect of CF gradually approached saturation when the replacement ratio exceeded 10%. These results suggest that CF can effectively shorten the induction period of cement hydration and accelerate the transition of the system into the acceleration stage, thereby promoting rapid setting behavior.

CF exhibited a significant enhancement effect on the early mechanical properties of the grout, as shown in Figure 2c,d. At an age of 1 h, the control mixture exhibited almost no measurable strength. In contrast, the compressive strength reached 5.7 MPa when the CF replacement ratio was 10%. When the CF replacement ratio further increased to 40%, the compressive strength increased to 59.1 MPa. At 3 h, the compressive strength increased from nearly 0 MPa to a maximum value of 96.8 MPa at a CF replacement ratio of 30%, indicating that the material could achieve ultra-high early strength within a very short period. This phenomenon may be associated with the high contents of aluminate and sulfate phases in CF, which promote the rapid formation of hydration products such as ettringite (AFt) and C–S–H gel. In addition, the rapid precipitation of AFt may provide nucleation sites for the formation of C–S–H gel, thereby contributing to the development of early strength. At 1 d and 3 d, both flexural strength and compressive strength gradually increased with increasing CF replacement ratio. At 7 d, the compressive strength continued to increase, whereas the flexural strength showed a trend of initial increase followed by a slight decrease. At 28 d, both flexural and compressive strengths exhibited a similar trend. The strengths first increased and then decreased with increasing CF replacement ratio. The maximum compressive strength was obtained at a CF replacement ratio of 10%, reaching 139.2 MPa.

The above results suggest that although CF can significantly accelerate early hydration and promote the rapid development of early strength, excessive CF incorporation may adversely affect the later-age performance of the material. One possible reason is that excessively rapid hydration promotes the early precipitation of hydration products on the surfaces of cement particles, resulting in the formation of a relatively dense layer. This layer may hinder the further penetration of water into unhydrated particles and consequently reduce the degree of later hydration. In addition, rapid hydration may lead to heterogeneous microstructure development and the formation of relatively coarse crystalline products. The associated heat release during early hydration may also contribute to pore structure coarsening and internal microdefects. These factors may collectively weaken the later-age mechanical properties of the material.

Comprehensive analysis of the experimental results indicates that a CF replacement ratio of 10% was beneficial for achieving the highest later-age strength. However, when the CF replacement ratio was further increased, the adverse effects on later-age performance gradually became more pronounced. In this study, a CF replacement ratio of 30% was selected for subsequent investigations because it provided a more balanced overall performance between ultra-early strength development and later-age mechanical properties. At this replacement ratio, the grout exhibited a compressive strength of 55.3 MPa at 1 h and 125.3 MPa at 28 d, indicating excellent ultra-early strength while maintaining relatively high later-age strength.

3.2. Effect of Retarder Dosage on Workability and Mechanical Properties

As shown in Figure 3a, the flowability of the grout gradually increased from 312 mm to 328 mm as the retarder dosage increased from 0.02% to 0.18%. This behavior may be related to the adsorption of retarder molecules on the surfaces of cement particles, which can delay early hydration and reduce particle flocculation, thereby improving the initial rheological properties of the grout. As illustrated in Figure 3b, the setting time increased significantly with increasing retarder dosage, rising from 20 min to 77 min, corresponding to an increase of approximately 285%. This result indicates that the retarder exhibited a pronounced effect on prolonging the setting process. One possible mechanism is that the retarder forms complexes with Ca²⁺ ions in the pore solution, thereby reducing the supersaturation of the liquid phase and delaying the precipitation of hydration products. In addition, the adsorption of retarder molecules on cement particle surfaces may inhibit the early nucleation and growth of ettringite (AFt) and C–S–H gel. These effects contributed to prolonging the workable time of the grout, which is beneficial for construction operations, particularly in large-volume or complex structural applications.

Figure 3c,d show the influence of retarder dosage on the mechanical properties of the grout. The results indicate that the retarder exhibited a significant inhibitory effect on early strength development. At an age of 1 h, both flexural strength and compressive strength gradually decreased with increasing retarder dosage. When the dosage reached 0.18%, almost no measurable strength was observed, indicating that early hydration was substantially delayed and that a stable load-bearing structure had not yet formed. At 3 h, both flexural and compressive strengths still decreased with increasing retarder dosage. However, the inhibitory effect became less pronounced compared with that observed at 1 h. Overall, the retarder prolonged the workable time of the grout by delaying the hydration process, although this effect was accompanied by a reduction in early-age strength. From 1 d to 28 d, both flexural strength and compressive strength gradually increased with increasing retarder dosage. The improvement was more pronounced for flexural strength. These results suggest that an appropriate retarder dosage may contribute to the improvement in later-age toughness and structural integrity of the material.

From a microstructural perspective, the incorporation of the retarder prolongs the induction period of cement hydration, which may allow a more uniform distribution of water within the grout system. As a result, the subsequent hydration process may proceed more gradually and uniformly throughout the matrix. This behavior may promote the formation of finer and more evenly distributed hydration products, which can contribute to pore refinement and the filling of capillary pores. In addition, a more uniform hydration process may help reduce local stress concentrations and microcrack formation associated with nonuniform hydration development. These effects may collectively contribute to the improvement in the later-age mechanical properties and durability of the material.

3.3. Effect of w/b on Workability and Mechanical Properties

As shown in Figure 4a, the flowability of the grout increased significantly from 293 mm to 347 mm as the water-to-binder ratio (w/b) increased from 0.18 to 0.22, corresponding to an increase of 18.4%. An approximately linear growth trend was observed. This behavior is mainly associated with the increase in free water content within the system. The additional free water reduces internal friction between particles and increases particle spacing, thereby improving the rheological properties of the grout. Figure 4b presents the variation in setting time with w/b ratio. The results show that the setting time increased slightly with increasing w/b, although the overall variation remained relatively small. This result indicates that the w/b ratio had a limited influence on the hydration rate of the system within the investigated range. One possible reason is that the increase in free water diluted the ionic concentration in the pore solution, thereby slightly delaying the time required for the system to reach the critical supersaturation state and slowing the precipitation of hydration products.

In terms of mechanical properties, both flexural strength and compressive strength at 1 h, 3 h, 1 d, and 3 d gradually decreased with increasing w/b ratio, as shown in Figure 4c,d. For example, the compressive strength at 1 h decreased from 83.7 MPa at a w/b ratio of 0.18 to 66.4 MPa at a w/b ratio of 0.22. Similarly, the compressive strength at 3 h decreased from 102.7 MPa to 85.6 MPa. The reduction in early-age strength may be related to the decrease in effective solid content per unit volume at higher w/b ratios. In addition, excess free water may evaporate or gradually transform into capillary pores during hardening, thereby reducing the compactness of the hardened matrix. Regarding later-age performance, the compressive strengths at 7 d and 28 d also gradually decreased with increasing w/b ratio. However, the flexural strength at 28 d exhibited a slight increasing trend, rising from 20.4 MPa at a w/b ratio of 0.18 to 22.4 MPa at a w/b ratio of 0.22. This result suggests a certain improvement in material toughness. Compressive strength is generally sensitive to the porosity and pore structure characteristics of cement-based materials. Increasing the w/b ratio may increase total porosity and promote pore coarsening, particularly the formation of capillary pores, thereby weakening the load-bearing capacity of the material. In contrast, flexural strength is influenced not only by porosity but also by crack propagation resistance. An appropriate increase in w/b ratio may facilitate later hydration reactions, particularly the secondary reaction of SF, and contribute to the formation of a more uniform gel structure. In addition, fine pores introduced by the higher w/b ratio may help alleviate local stress concentration at crack tips and enhance energy dissipation during loading. These effects may contribute to the improvement in material toughness to a certain extent.

Comprehensive analysis of the experimental results indicates that a lower w/b ratio (0.18) was beneficial for improving the mechanical strength of the material. However, excessively low w/b may reduce the uniformity of the matrix and increase the possibility of localized stress concentration. In contrast, a moderate increase in w/b ratio (e.g., 0.22) may introduce additional pores but may also contribute to the improvement in material toughness. Considering the balance among flowability, mechanical performance, and workability, a w/b ratio of 0.19 was selected as the optimal value in this study.

Under the optimal parameter combination, namely a CF replacement ratio of 30%, a retarder dosage of 0.02%, and a w/b ratio of 0.19, the grout exhibited excellent overall performance. The initial flowability reached 312 mm, and the setting time was 20 min. The compressive strength reached 81.4 MPa at 1 h, 99.2 MPa at 3 h, and 121.3 MPa at 28 d. This mixture proportion provided good flowability and an appropriate workable time while maintaining outstanding ultra-early strength development, indicating its potential suitability for ultra-early high-strength repair applications.

3.4. Volume Stability and Durability

As shown in Figure 5, the prepared ultra-early high-strength grout exhibited good volumetric stability from 1 d to 28 d. Under water-curing conditions, as illustrated in Figure 5a, the volumetric change in the material showed a trend of initial shrinkage followed by slight expansion. The maximum shrinkage occurred at 1 d, with a shrinkage rate of 0.014%. Subsequently, the material gradually transitioned to a slight expansion state, and the shrinkage rate reached −0.002% at 28 d. Under air-curing conditions, as shown in Figure 5b, the shrinkage rate at 28 d was 0.027%, which was lower than that commonly reported for conventional cement-based materials. This result indicates good volumetric stability of the grout. This behavior may be related to the influence of CF on the early hydration process. CF promoted the rapid formation of ettringite (AFt) during the early hydration stage. The formation of AFt was accompanied by a certain expansion effect, and the needle-like or rod-like AFt crystals may contribute to pore filling and matrix densification. These effects may partially compensate for the shrinkage generated during cement hydration, thereby contributing to the improvement in volumetric stability.

In terms of durability, the electric flux at 28 d was only 12 C, indicating extremely low ionic permeability and excellent resistance to chloride ion penetration. A low electric flux value suggests that the penetration of harmful media, such as chloride ions and sulfate ions, was effectively restricted. This characteristic may contribute to improved durability under chloride-rich environments, sulfate exposure, and freeze–thaw conditions. From a microstructural perspective, the favorable durability performance may be associated with the formation of a relatively dense internal microstructure. Hydration products such as AFt and C–S–H gel were uniformly distributed within the matrix and may contribute to pore filling and the reduction in pore connectivity, thereby limiting the transport pathways of harmful ions. This dense microstructure may improve the impermeability of the material and contribute to its long-term durability performance.

3.5. Microstructural Analysis

Figure 6 presents the XRD patterns of the grout without CF (0%) and with a CF replacement ratio of 30% at different ages (1 h, 1 d, and 28 d). For the specimen without CF, the main hydration products at 1 d included ettringite (AFt), calcium hydroxide (Ca(OH)₂), and C–S–H gel. Diffraction peaks corresponding to partially unhydrated calcium silicate phases were still observed, indicating that hydration was not yet complete at this stage. As the curing age increased to 28 d, the diffraction peak intensities of unhydrated calcium silicate phases and Ca(OH)₂ gradually decreased. Meanwhile, the characteristic peaks of AFt became significantly weaker, and the hydration products were mainly dominated by C–S–H gel. These results suggest that the hydration reaction continued to progress with curing age and that the internal structure of the system gradually evolved toward a more stable and denser state.

In contrast, the hydration behavior of the system changed significantly after the incorporation of a CF replacement ratio of 30%. A strong characteristic diffraction peak of ettringite (AFt) was already observed at 1 h, indicating that CF promoted the rapid formation of AFt during the early hydration stage. In addition, a small amount of residual gypsum was still detected. At 1 d, the diffraction peaks of unhydrated mineral phases further decreased, while the characteristic peaks of AFt and C–S–H became more pronounced. At 28 d, most unhydrated mineral phases were consumed. The diffraction peaks of Ca(OH)₂ became significantly weaker or nearly disappeared, whereas the characteristic peaks of AFt and C–S–H further increased. Meanwhile, weak gypsum peaks could still be observed. These results suggest that CF promoted the formation of AFt and accelerated the hydration of silicate minerals, thereby increasing the formation of C–S–H gel. In addition, the persistence of AFt peaks at later ages may indicate that CF influenced the transformation behavior of AFt to monosulfate (AFm) to some extent. To further investigate the microstructural evolution of the grout, SEM and EDS analyses under different conditions are presented in Figure 7. For the specimen without CF, the microstructure at 1 d exhibited a typical sponge-like or flocculent C–S–H gel morphology, as shown in Figure 7a. No obvious needle-like or rod-like crystals were observed. The EDS results showed that the analyzed region mainly contained Ca, Si, and O elements, together with small amounts of Al and S. These observations suggest that C–S–H gel was the dominant hydration product in the analyzed region, while AFt may have existed in a finely dispersed form. At 28 d, the microstructure gradually evolved into a denser and more continuous gel network. A large amount of C–S–H gel filled the pores, resulting in improved compactness of the matrix, as illustrated in Figure 7b.

After the incorporation of a CF replacement ratio of 30%, the early microstructure of the material showed significant differences, as illustrated in Figure 7c. At an age of 1 h, a large number of rod-like or columnar AFt crystals were observed. The crystal lengths ranged from approximately 0.5 to 2 μm. These crystals intersected and interconnected to form an initial spatial framework. The EDS results showed that the analyzed region mainly contained Ca, Al, and S elements, which is consistent with the chemical composition of AFt. The rapid formation of this crystalline framework may contribute to the ultra-early strength development of the material. At 1 d, as shown in Figure 7d, AFt crystals still acted as the primary skeletal structure of the system. A large amount of C–S–H gel with fibrous and flake-like morphology was observed on the surfaces of AFt crystals and within the surrounding pores, gradually forming a crystal–gel composite microstructure. The AFt crystals may provide nucleation sites for the deposition and growth of C–S–H gel, thereby contributing to microstructural densification. At 28 d, as illustrated in Figure 7e, the system further evolved into a relatively dense gel network structure. Some AFt crystals appeared to be encapsulated by C–S–H gel and remained distributed within the matrix. The EDS analysis showed that, in addition to Ca, Si, and O elements, Al and S elements were still detected, which may indicate the continued presence of AFt within the system. At this stage, the continuous formation of C–S–H gel and progressive pore filling contributed to the interconnection of the crystalline and gel phases, resulting in a denser microstructure.

Comprehensive analysis of the XRD and SEM/EDS results suggests that the incorporation of CF significantly influenced the hydration behavior and microstructural evolution of the system. During the early hydration stage, CF promoted the rapid formation of AFt, which may contribute to the development of an initial skeletal framework. At later ages, C–S–H gel was gradually generated on the surfaces of AFt crystals and within surrounding pores, contributing to pore filling and microstructural densification. The interaction between AFt crystals and C–S–H gel may play an important role in the development of ultra-early strength, improved compactness, and favorable volumetric stability of the material. In addition, the observed microstructural characteristics were generally consistent with the macroscopic mechanical properties and durability results.

4. Conclusions

This study systematically investigated the effects of CF replacement ratio, retarder dosage, and water-to-binder ratio (w/b) on the flowability, setting behavior, mechanical properties, durability, and microstructural characteristics of ultra-early high-strength cement-based grouts. The main conclusions are summarized as follows.

(1)

CF had a limited influence on grout flowability but significantly shortened the setting time and enhanced early-age strength development. Increasing the CF replacement ratio markedly improved the 1 h and 3 h compressive strengths. However, excessive CF incorporation adversely affected later-age strength. A CF replacement ratio of 30% provided a balanced performance between ultra-early strength and later-age mechanical properties.

(2)

The retarder significantly prolonged the setting time and improved grout workability. However, increasing the retarder dosage reduced early-age strength development. In contrast, appropriate retarder incorporation contributed to the improvement in later-age mechanical performance and toughness.

(3)

Increasing the w/b ratio significantly improved flowability but gradually reduced compressive strength at all curing ages. Meanwhile, the 28 d flexural strength showed a slight increase, indicating improved toughness. Considering the balance among flowability, strength, and workability, a w/b ratio of 0.19 was selected as the optimal value.

(4)

The optimal mixture proportion consisted of a CF replacement ratio of 30%, a retarder dosage of 0.02%, and a w/b ratio of 0.19. Under this condition, the grout exhibited a flowability of 312 mm, a compressive strength of 81.4 MPa at 1 h, and a compressive strength of 121.3 MPa at 28 d, demonstrating favorable overall performance.

(5)

The prepared grout exhibited good volumetric stability and durability. Under air-curing conditions, the 28 d shrinkage rate was 0.027%, while the electric flux at 28 d was only 12 C, indicating low ionic permeability and good resistance to chloride ion penetration.

(6)

XRD and SEM/EDS analyses showed that CF promoted the rapid formation of AFt during the early hydration stage. The interaction between AFt and C–S–H gel contributed to the development of a denser microstructure, which may be associated with the improved early-age strength and durability performance of the grout.

Overall, the results demonstrate that the combined regulation of CF replacement ratio, retarder dosage, and w/b ratio can effectively balance setting behavior and strength development in ultra-early high-strength cement-based grouts.