Ultra-high-strength geopolymer concrete influenced by mixture proportion and steel fiber: flowability, compressive strength, and flexural performance

Ultra-high-performance geopolymer concrete is a low-carbon and environmentally friendly cementitious composite formed through alkali activation of industrial by-products and incorporation of ultrafine fillers and fiber reinforcement, exhibiting high compressive strength and excellent durability. In this study, ultra-high-strength geopolymer concretes (UHSGM) were prepared using ground granulated blast furnace slag (GGBFS), fly ash (FA), and silica fume (SF) as binders. Three GGBFS-to-FA mass ratios, namely 4:1, 1:1, and 1:4, were adopted. The SF content was varied at 0%, 5%, 10%, 20%, and 30% by mass of the precursor. In addition, three types of steel fiber were incorporated at a constant volume fraction of 2%. Two curing conditions, namely standard curing and steam curing, were adopted. The influences of mixture proportions, with or without silica fume and steel fiber, on the flowability, compressive strength, and flexural performance of UHSGM were systematically evaluated. The results indicated that an increase in GGBFS content led to a reduction in flowability but an enhancement in the compressive strength of UHSGM. The influence of SF on flowability and compressive strength was strongly dependent on the GGBFS-to-FA ratio. The highest compressive strength of the geopolymer reached 157.0 MPa at a GGBFS-to-FA ratio of 4:1 with 5% SF, which is comparable to that of ultra-high-performance concrete. An increase in GGBFS content and the incorporation of steel fibers enhanced the flexural strength of UHSGM, whereas high-temperature curing led to a reduction in flexural strength. An increase in SF had a negative effect on the deflection capacity and toughness of UHSGM, regardless of curing conditions. Further investigation is required to optimize the overall performance of UHSGM.

1 Introduction

Ultra-high performance concrete (UHPC) is a cement-based material characterized by compressive strength exceeding 120 MPa, high tensile ductility, superior toughness, and excellent durability. It is manufactured using a low water-to-binder ratio, a high cement content, supplementary materials such as silica fume (SF) and quartz powder, superplasticizers, steel fibers, and thermal curing (Shi et al., 2015; Wang et al., 2015; Fan et al., 2023; Liu et al., 2024). However, the extensive use of Portland cement (PC) and the requirement for high-temperature curing restrict the large-scale production and application of UHPC because of its high cost, high energy consumption, and significant carbon emissions (Yu et al., 2014). Compared with ordinary Portland cement concrete, the PC content in UHPC is increased by approximately three times. It is well known that PC production is resource-intensive and accounts for about 5%–7% of global carbon emissions, which compromises the environmental sustainability of construction materials, particularly UHPC (Shen et al., 2015; He et al., 2019). Therefore, to promote the cost-effectiveness and environmental friendliness of UHPC for broader applications, further investigations are required.

Recently, alkali-activated materials (AAMs), including geopolymers, have been widely regarded as promising low-carbon clinker-free binders for replacing Portland cement (PC), thereby enhancing the sustainability of construction materials (Provis, 2018; Sun et al., 2022). These materials are produced by activating aluminosilicate sources such as ground blast furnace slag (GGBFS) (Bhojaraju et al., 2023), fly ash (FA) (Al-Shmaisani et al., 2022), metakaolin (Geu et al., 2026), and waste glass (Heriyanto et al., 2018) using alkaline activators, typically sodium hydroxide (NaOH) and sodium silicate (Guo et al., 2019). Compared with PC, the preparation process of alkali-activated binders is more environmentally friendly and cost-effective. Previous studies have reported that the unit energy consumption and carbon emissions of alkali-activated binders are approximately 60% lower than those associated with PC production (Chen et al., 2023). Moreover, AAMs prepared with GGBFS and FA exhibit excellent overall performance, particularly in terms of mechanical properties, and demonstrate strong potential for producing ultra-high-strength materials comparable to UHPC. Feng et al. (2024) investigated AAM based on FA and GGBFS and found that the addition of cellulose nanocrystals significantly improved mechanical performance, especially flexural strength. They reported that incorporating 0.3% cellulose nanocrystals increased the 28-day compressive strength by about 17%–19% and the flexural strength by 50%–61%, which was attributed to enhanced alkali-activation reactions, a denser microstructure, and nano-reinforcing effects. Zhao et al. (2021) investigated FA–GGBFS-based AAM incorporating partial replacement with steel slag and reported that an optimal steel slag content of approximately 15% provided a favorable balance between strength and ductility. The corresponding mixture exhibited high compressive strength (approximately 93 MPa), enhanced tensile strain capacity (around 3.9%), and stable multiple cracking behavior. These improvements were mainly attributed to modifications in the fiber–matrix interfacial bonding, rather than to pronounced changes in the matrix microstructure. Wang et al. (2024) studied fiber-reinforced alkali-activated composites and demonstrated that their mechanical performance is predominantly governed by the matrix composition and fiber–matrix interactions. Their results indicated that an optimized matrix improved compressive strength and stiffness, while suitable interfacial bonding facilitated effective crack bridging and markedly enhanced tensile strength and strain capacity.

On the other hand, although solid wastes can potentially be used to produce high-strength construction materials, as noted above, studies on the mixture proportioning of ultra-high strength geopolymer materials, such as concrete and concretes (UHSGM), remain limited, particularly with respect to the incorporation of SF. It is well established that SF plays a significant role in enhancing the microstructure and fiber–matrix bonding of UHPC (Luan et al., 2022; He et al., 2024). However, its effects on the workability and mechanical properties of AAM are still inconclusive. Lin et al. (2024) demonstrated that the synergistic use of steel slag (SS), GGBS, and FA optimizes the performance of AAM, where the inclusion of SS improves workability by delaying initial setting, while the GGBS-rich ternary system achieves superior mechanical properties, including high compressive strength (up to 110 MPa) and exceptional tensile strain capacity (5%–7%). Kong and Kurumisawa (2023) conducted a systematic evaluation of mix-design parameters governing the workability of AAM by synthesizing data from published studies and developing a machine-learning-based predictive model. Their findings indicated that precursor characteristics and activator composition indirectly affect the practical performance of AAM through coupled effects on fresh-state behavior. While recent studies have investigated the general mix design of geopolymer composites (Liu et al., 2020; Liu et al., 2025), the coupling effect between the GGBFS-to-FA ratio and the saturation threshold of SF remains underexplored. Unlike conventional UHPC, where high SF dosage is standard, the optimal SF content in alkali-activated systems is hypothesized to be strictly dependent on the precursor chemistry. Therefore, this study aims to elucidate this interdependent mechanism and establish a direct benchmark against OPC-based UHPC.

In this study, three GGBFS-to-FA mass ratios of 4:1, 1:1, and 1:4 were adopted with different SF contents. The SF contents were set to 0%, 5%, 10%, 20%, and 30%, respectively. Three types of steel fibers, including straight and hooked-end fibers with different sizes and geometries, were incorporated. Two curing conditions, namely 28 days standard curing and steam curing, were adopted. The flowability, compressive strength, and flexural properties, including flexural strength, load–deflection behavior, and toughness, were evaluated. This study aims to provide a preliminary assessment of the effectiveness of mixture proportion design and steel fiber reinforcement in UHSGM, with comparison to UHPC.

2 Experimental program

2.1 Materials

The geopolymer matrix was synthesized via the alkaline activation of GGBFS, FA, and SF. For comparison, a reference UHPC was prepared using Portland cement (P.I. 42.5). The chemical compositions of the raw materials are summarized in Table 1. The alkaline activator was prepared by mixing sodium silicate solution (8.3% Na₂O, 26.5% SiO₂, and 65.2% H₂O) and industrial-grade NaOH pellets (98% ± 1% purity) to achieve a target silicate modulus (Ms, molar ratio of SiO₂/Na₂O) of 2.0. The activator was prepared 24 h before casting to ensure the complete dissolution of NaOH and to allow the solution to cool to room temperature. To ensure batch-to-batch consistency, the solution was prepared in bulk, sealed to prevent evaporation and carbonation, and maintained at a constant liquid-to-binder ratio (L/B) of 0.30 for all mixtures. Natural Xiang River sand, with a maximum particle size of 2.36 mm, served as the fine aggregate. To ensure workability at a low W/B ratio of 0.18, a polycarboxylate-based SP was incorporated.

Table 1

Table 1. Chemical composition of GGBFS, FA, cement and SF (%).

Three types of steel fibers with a consistent length of 13 mm and tensile strengths exceeding 2,500 MPa were employed: two straight fibers with diameters of 0.12 mm (Fiber Z) and 0.22 mm (Fiber S), and one hooked-end fiber with a diameter of 0.22 mm (Fiber D). Their geometric configurations are illustrated in Figure 1.

Figure 1

Figure 1. Dimensions of steel fibers.

2.2 Mix design of UHSGM and UHPC

Based on previous studies (Li et al., 2018), a Na₂O content of 7% and a silicate Ms of 2.0 were adopted. The effective water-to-binder (W/B) ratio was controlled at 0.335, accounting for both the added water and the water present in the sodium silicate solution. Additionally, the fine aggregates were used in a saturated surface dry (SSD) state to prevent moisture absorption. To systematically investigate the influence of precursor composition on strength development, three GGBFS-to-FA mass ratios (GGBFS/FA) were employed: 4:1 (Group A), 1:1 (Group B), and 1:4 (Group C). Additionally, the dosage of SF was varied at 0%, 5%, 10%, 20%, and 30% by weight. This range was adopted based on previous studies (Yu et al., 2014; Shi et al., 2015), which suggest that the optimal replacement level typically falls within this spectrum. A constant dosage of 2 vol.% for all three types of steel fibers was maintained, and the sand-to-binder ratio was fixed at 1.0. Furthermore, the alkaline activator dosage and composition were kept constant across all mixtures to isolate the effects of the binder variations, despite the increased water demand associated with higher SF contents.

Consistent with a prior study (Wu et al., 2017), a reference UHPC was prepared as a benchmark. The binder phase consisted of 45% cement, 30% GGBFS, and 25% SF by mass, with a W/B ratio of 0.18 and a SP dosage of 2% by weight of the binder. The detailed mix proportions for both the UHSGM and the UHPC are summarized in Table 2.

Table 2

Table 2. Mixture proportions of UHSGM and UHPC.

2.3 Preparation and curing of samples

For the UHSGM, the binders and sand were initially dry-mixed for 3 min at a low speed. Subsequently, the alkaline activators were introduced gradually and mixed for 3 min at a low speed, followed by 1 min of high-speed mixing to ensure homogeneity. The steel fibers were then carefully incorporated using a 5 mm sieve to prevent agglomeration and mixed for an additional 5 min at a low speed.

The mixing protocol for the reference UHPC followed the methodology established in previous research (Wu et al., 2017). The dry constituents, including cement, GGBFS, SF, SP, and sand, were initially blended for 3 min at a low speed. Subsequently, water was gradually introduced and mixed for 5 min at a low speed, followed by 1 min of high-speed mixing to achieve a fluid state. Finally, the steel fibers were carefully dispersed into the mixture through a 5 mm sieve to ensure uniform distribution.

The freshly mixed pastes were cast into steel molds with dimensions of 40 × 40 × 160 mm, in accordance with the procedures outlined in the Chinese Standard GB/T 17,671. The specimens were then pre-cured in the molds for 24 h at an ambient temperature of 20 °C and a relative humidity (RH) of 65%. To prevent moisture evaporation, the mold surfaces were tightly covered with a plastic film. Following the pre-curing period, the specimens were demolded and subjected to two distinct curing regimes:

1. Steam curing: cured in steam at 80 °C for 24 h.

2. Standard curing: cured in the chamber at ambient temperature of 20 °C and RH > 90% for 27 days (28 days in total)

2.4 Experimental methods

2.4.1 Flowability

The flowability of the mixtures was evaluated in accordance with the Chinese Standard GB/T 2419-2005. The fresh mixture was cast into a truncated mini-cone mold and tamped firmly to ensure compaction. Upon rapid removal of the mold, the flow table was activated to jolt the plate 25 times within a specified duration. Subsequently, the spread diameters were measured using a ruler in two perpendicular directions. The average of these two measurements was recorded as the flow value.

2.4.2 Compressive strength

Subsequent to the three-point flexural strength test, the six resulting prism halves, with dimensions of 40 × 40 × 40 mm, were utilized to determine the compressive strength. Following the protocols established in previous studies (Wu et al., 2018), a constant loading rate of 2.4 kN/s was applied. The results were reported as the mean value obtained from the six specimens to ensure statistical consistency.

2.4.3 Flexural behavior

The three-point flexural strength was evaluated using a universal testing machine with a maximum capacity of 200 kN. The supporting span was fixed at 100 mm. To accurately capture the mid-span deflection, two linear variable differential transformers (LVDTs) were mounted at the center of the specimen. The test was performed under displacement control at a constant loading rate of 0.2 mm/min.

The flexural toughness was determined by calculating the area under the flexural load-deflection curve up to a deflection limit of 5 mm. The results were reported as the mean values ±standard deviation (SD) obtained from three specimens to ensure statistical reliability. Furthermore, error bars representing the standard deviation are included in all graphical figures to visualize the dispersion and statistical significance of the data trends.

3 Results and discussion

To facilitate the discussion, the results of flowability, compressive strength, and flexural properties are summarized in Table 3.

Table 3

Table 3. The results of flowability, compressive strength and flexural properties of UHSGM and UHPC.

3.1 Flowability

3.1.1 Effect of mixture proportion

Figure 2 illustrates the influence of GGBFS/FA ratios and SF content on the flowability of geopolymer concretes. In the absence of SF, an increase in FA content led to an enhancement in flowability. Specifically, the flowability of mixture A0 (GGBFS/FA = 4:1) was 188 mm; as the ratio decreased to 1:1 and 1:4, the flowability increased by 16 mm and 21 mm, respectively. Furthermore, the incorporation of SF significantly modified the fresh properties of the geopolymers. At a GGBFS/FA ratio of 4:1, the flowability initially increased from 188 mm to a peak of 260 mm at 20% SF replacement, before declining to 235 mm at 30% SF content. Similar trends were observed for the mixtures with a 1:1 ratio, where flowability peaked at 265 mm with 10% SF content and subsequently decreased to 246 mm with further SF addition. In contrast, the effect of SF was less pronounced at a GGBFS/FA ratio of 1:4. The maximum flowability in this group was 234 mm (at 5% SF), representing a modest increase of 15 mm compared to the reference.

Figure 2

Figure 2. Flowability of UHSGM. (a) Effect of mixture proportion (b) Effect of adding steel fibers.

A lubricating effect, often referred to as the “ball-bearing” effect, is facilitated by the spherical morphology and glassy surface of FA particles, leading to a reduction in both water demand and internal friction. Conversely, a slight elevation in yield stress is induced by a high FA content, thereby resulting in decreased fluidity. It has been noted in previous research (Wu et al., 2018) that the influence of FA on the workability of alkali-activated slag/FA systems is intricate; while a reduction in plastic viscosity can be induced by the ball-bearing effect, the overall rheological balance remains complex. Regarding the addition of SF, an enhancement in fluidity is observed at low replacement levels, which is likely attributable to the spherical shape of SF particles providing additional lubrication. However, as the SF content is further increased, a higher water requirement is necessitated by the significantly larger specific surface area of SF, thus leading to reduced flowability at a constant W/B ratio.

3.1.2 Effect of steel fiber reinforcement

Figure 2B illustrates the effect of various steel fibers on the flowability of UHSGM mixtures. A reduction in flowability was induced by the incorporation of steel fibers, which is consistent with findings reported in previous studies (Alrefaei and Dai, 2018). However, these discrepancies were less pronounced, particularly when the SF content remained below 10%. For instance, a flowability of 228 mm was recorded for mixture ZA2, representing a marginal decrease of 2 mm compared to the fiber-free mixture A2. Furthermore, the variation trends in flowability for mixtures with steel fibers followed those observed in fiber-free mixtures. This consistency implies that utilizing SF is a feasible strategy for enhancing the workability of fiber-reinforced geopolymer composites.

The flowability of UHSGM was influenced by the total number of steel fibers. The incorporation of Fiber Z, with a smaller diameter of 0.12 mm, resulted in the lowest flowability of 182 mm. In contrast, the flow diameters increased by 4 mm and 2 mm when Fiber S and Fiber D (both with a larger diameter of 0.22 mm) were used, respectively. This is attributed to the fact that at a fixed volume fraction of fibers, a smaller fiber diameter yields a larger quantity of fibers in the mixture, thereby exhibiting a more significant effect on reducing the flowability.

3.1.3 Comparison between UHSGM and UHPC

As shown in Figure 2, the flowability of the UHPC without steel fibers was 198 mm. A reduction of 15 mm in flowability was observed upon the addition of steel fibers. Interestingly, the flowability of the UHSGM was found to be higher than that of the UHPC when SF was incorporated. For instance, a flowability of 230 mm was recorded for sample ZA3, which was 25% higher than that of sample ZUHPC. These results indicate that the flowability of UHSGM is comparable to that of UHPC, demonstrating that construction requirements can be satisfied using UHSGM.

3.2 Compressive strength

3.2.1 Effect of mixture proportion

Figure 3 illustrates the effects of the GGBFS/FA ratio and SF content on the compressive strength of geopolymer concretes under steam curing. The improvement in compressive strength was associated with the increase in GGBFS content, which is consistent with a previous study (Ding et al., 2018). In the absence of SF, a maximum compressive strength of 98.8 MPa was obtained at a GGBFS/FA ratio of 4:1. Conversely, the compressive strength decreased as the GGBFS content was reduced or the FA content was increased. When the GGBFS/FA ratio was adjusted to 1:1, a reduction of 21.4 MPa in compressive strength was recorded. The minimum compressive strength of 39.0 MPa was observed at a GGBFS/FA ratio of 1:4.

Figure 3

Figure 3. Compressive strength of geopolymer concretes with different mixture proportion under steam curing.

The compressive strength development of geopolymer concretes was significantly influenced by the incorporation of SF. The maximum compressive strength of 115.9 MPa was recorded for sample A2. For concretes with a GGBFS/FA ratio of 4:1, the compressive strength was enhanced by 12.0% and 17.3% with the addition of 5% and 10% SF, respectively. However, a sharp decline in compressive strength was observed with a further increase in SF content; at a 20% SF dosage, the compressive strength decreased by 30.2% compared to that of sample A2. Subsequently, a slight increment in compressive strength was noted when the SF content reached 30%. In contrast, no positive effect was observed for concretes with GGBFS/FA ratios of 1:1 and 1:4, as their compressive strengths decreased upon the addition of SF.

Significant improvements in the strength of geopolymers have been attributed in previous research to the high reactivity and CaO content of GGBFS (Li et al., 2017), a trend consistent with the observations in this study. Consistent with established findings in the literature (Luan et al., 2022; He et al., 2024), it is inferred that the proper addition of SF (≤10%) contributes to pore refinement and the densification of the interfacial transition zone (ITZ), thereby enhancing the matrix strength. However, a distinct distinction from conventional UHPC is identified here. While a high dosage of SF (20%–30%) is typically required in UHPC for pozzolanic reaction and packing density, a distinct saturation threshold was observed in the UHSGM. It is indicated that the contribution of SF is non-linear and regressive beyond a 10% dosage for the GGBFS-rich matrix. This suggests that in alkali-activated systems, excess SF may act as an inert filler or hinder the geopolymerization network formation, contrasting with its role in hydrated cement systems.

3.2.2 Efficiency of steel fiber reinforcement and SF

Figure 4 presents effect of steel fiber reinforcement on compressive strengths of UHSGM under different curing conditions. With the addition of steel fiber, the compressive strength significantly improved, compared to the plain concretes, as shown in Figure 4a. By comparing Figures 4a,b, it can be seen that with the steel fiber reinforcement, the compressive strengths of samples cured at ambient temperature were also satisfactory. Among three types of steel fiber, Fiber Z was most effective on strengthening the concretes, followed by Fiber D and S, respectively. For instance, when the SF was 5%, the highest compressive strength of 157.0 MPa was obtained under the steam curing condition, which was increased by 41.8% and higher than 2.9%, in comparison with the sample A1 and SA0, respectively. This is probably attributed to larger amount of Fiber Z among others in the sample, as discussed in Section 3.1.2.

Figure 4

Figure 4. Compressive strength of UHSGM under different curing conditions. (a) Steam curing. (b) Standard curing.

Moreover, SF played a limited role in the compressive strength development in this study. A distinct trend was observed in the compressive strength of steel fiber-reinforced UHSGM compared to that of the plain UHSGM. As the SF content increased, the reinforcement efficiency of the steel fibers was found to diminish. For instance, the incorporation of steel fibers into sample A2 resulted in a 13.4% increase in compressive strength. As illustrated in Figure 4, a sharp decline in compressive strength occurred once the SF dosage exceeded 5%. Regardless of the curing conditions, the improvement in compressive strength remained marginal when the SF content was 5% or lower. For example, for samples cured at ambient temperature, the addition of 5% SF led to a negligible increase of only 0.3% compared to the SF-free counterparts. This phenomenon might be attributed to the observation that the continuous addition of SF in the presence of fibers led to an increase in air voids, thereby weakening the bond strength between the matrix and the fibers.

In summary, irrespective of the curing regimes, a significant improvement in compressive strength was observed through steel fiber reinforcement. Nevertheless, the contributory role of SF remained limited in this context.

3.2.3 Comparison between UHSGM and UHPC

The compressive strength results for the UHPCs are illustrated in Figure 5. The average compressive strength of the UHPC sample cured at ambient temperature was 111.4 MPa, which was slightly higher than that of sample A1. Similarly, the compressive strengths of sample ZUHPC under steam and standard curing conditions were recorded at 153.9 MPa and 130.8 MPa, respectively, which were comparable to those of the fiber-reinforced UHSGMs. These findings indicate that the strength requirements of UHSGMs are met across all curing conditions, demonstrating their comparability to UHPCs. Additionally, as these results were obtained at a relatively high W/B ratio of 0.32, a significant potential for further strength enhancement in UHSGMs is implied through the reduction of water dosage.

Figure 5

Figure 5. Compressive strengths of UHSGM and UHPC under different curing conditions.

3.3 Flexural property

3.3.1 Flexural strength

3.3.1.1 Effect of mixture proportion

As illustrated in Figure 6a, the flexural strengths were strongly dependent on the mixture proportions. Consistent with previous findings (Ding et al., 2018) and the observed compressive strength trends, an increase in GGBFS content contributed to an improvement in the flexural strength of the UHSGMs. In addition to the GGBFS/FA ratio, the variation in flexural strength was influenced by the SF content. Unlike the compressive strength trends, particularly for samples with a GGBFS/FA ratio of 4:1, the flexural strength initially decreased and subsequently increased upon the incorporation of SF. When the SF dosage exceeded 20%, the flexural strengths were found to be higher than those of the reference samples, with the exception of the group with a GGBFS/FA ratio of 4:1.

Figure 6

Figure 6. Flexural strengths of UHSGM and UHPC. (a) Effect of mixture proportion. (b) Effect of steel fiber reinforcement. (c) Effect of steel fiber type. (d) Effect of curing condition.

3.3.1.2 Effect of steel fiber reinforcement

As illustrated in Figure 6b, the incorporation of steel fibers resulted in a significant improvement in flexural strength, which is consistent with previous studies (Kim et al., 2015). For instance, the flexural strength of sample ZA0 under steam curing was recorded at 14.2 MPa, representing a 163% increase compared to that of sample A0. Similarly, for samples with a GGBFS/FA ratio of 1:4, an increase of 107% in flexural strength was observed. This enhancement is attributed to the inhibition of crack initiation and propagation by the bridging effect of the fibers, thereby improving the overall strength of the composite. However, at high silica fume contents, the failure mode shifts. The significant reduction in flowability leads to increased entrapped air and the formation of agglomerates. Analysis suggests that these processing defects dominate the flexural degradation by creating weak points at the fiber–matrix interface. This directly leads to reduced fiber pullout work, as the effective bonding area is compromised by voids, overriding the effects of pure matrix embrittlement.

In addition, as shown in Figure 6c, the type of steel fiber exerted a significant influence on the flexural strength, regardless of the curing conditions. The reinforcing effect of Fiber Z was found to be the most notable, followed by Fibers D and S, respectively. As discussed in Section 3.1.2, this superiority is attributed to the higher quantity of individual filaments provided by Fiber Z compared to the other two fiber types. On the other hand, the flexural strengths of sample DA0 under steam and standard curing were 39.5% and 53.1% higher, respectively, than those of sample SA0. This indicates that at an equivalent diameter and length, hooked-end fibers exhibit a superior mechanical response compared to straight fibers, which is attributed to the enhanced mechanical anchorage provided by the hooked ends.

3.3.1.3 Effect of curing condition

Figure 6d illustrates the flexural strengths of UHSGM and UHPC under various curing conditions. For the UHSGMs, the flexural strengths under steam curing were found to be slightly lower than those under standard curing. This trend does not correspond with the compressive strength results, an observation that is consistent with a previous study (Huseien et al., 2016). The alteration in the geopolymer microstructure during steam curing, compared to that cured at ambient temperature, is thought to result in increased brittleness. Furthermore, the formation of a denser geopolymer matrix at higher temperatures can induce slight debonding between the matrix and the steel fibers, which is attributed to the contraction of the matrix.

3.3.2 Flexural load-deflection curve and toughness

Figure 7 illustrates the effects of mixture proportions and steel fibers on the average flexural load-deflection curves of UHSGM and UHPC under different curing conditions. It is clearly observed that the flexural load initially increases linearly, followed by a nonlinear ascent until the peak load is reached. Upon further loading, the load-carrying capacity decreases gradually and nonlinearly. The descending portions of the curves exhibit distinct characteristics, which are influenced by various experimental parameters. Consequently, the toughness of the specimens is also strongly affected by these factors.

Figure 7

Figure 7. Flexural load-deflection curves. (a) Effect of GGBFS/FA ratio. (b) Effect of silica fume under steam curing. (c) Effect of silica fume under standard curing. (d) Comparison between UHSGM UHPC. (e) Effect of steel fiber under steam curing. (f) Effect of steel fiber under standard curing.

3.3.2.1 Effect of mixture proportion

The increase in GGBFS content not only enhanced the ultimate flexural load but also significantly impacted the mid-span deflection, as illustrated in Figure 7a. The maximum deflection of 0.85 mm was obtained at a GGBFS/FA ratio of 4:1. As this ratio decreased to 1:1 and 1:4, the recorded deflections were reduced to 0.50 mm and 0.63 mm, respectively. Furthermore, an increase in GGBFS content was found to exert a significant and positive influence on the toughness. Specifically, the toughness of the sample with a GGBFS/FA ratio of 4:1 was 28.5% and 512.6% greater than those of the samples with GGBFS/FA ratios of 1:1 and 1:4, respectively.

As illustrated in Figures 7b,c, the addition of SF resulted in a decrease in deflection. In Figure 7b, compared to the reference sample ZA0, reductions in deflection of 0.21 mm, 0.46 mm, 0.26 mm, and 0.18 mm were recorded for SF contents of 5%, 10%, 20%, and 30%, respectively. In addition to deflection, the toughness was found to be dependent on the SF content. Under steam curing, the toughness initially decreased and subsequently exhibited a slight increase as the SF dosage was raised. Notably, the descending portion of the curves exhibited a serrated shape, a feature that became more pronounced at higher SF contents.

3.3.2.2 Effect of steel fiber type

Figure 7e,f illustrate the influence of steel fiber type on the flexural load-deflection curves of UHSGMs under various curing conditions. It can be observed that irrespective of the curing regime, the highest reinforcement efficiency was achieved using Fiber Z (straight, d = 0.12 mm, l = 13 mm), followed by Fiber D (hooked-end) and Fiber S (straight, d = 0.22 mm, l = 13 mm). For instance, under standard curing, the deflection recorded for the sample containing Fiber Z was 0.66 mm, which was 50% and 24% higher than those obtained with Fiber S and Fiber D, respectively. Additionally, the toughness of sample ZA0 was 89.4% and 78.0% greater than that of samples SA0 and DA0, respectively. Consistent with previous research (Bhutta et al., 2017), samples reinforced with hooked-end fibers exhibited a superior flexural response compared to those utilizing straight fibers of the same diameter and length. Furthermore, a more jagged descending branch was observed for samples containing hooked-end fibers, particularly after steam curing, which is attributed to the presence of mechanical anchorage.

3.3.2.3 Effect of curing condition

Based on Figures 7b,c,e,f, it can be observed that the curing conditions significantly influenced the deflections and toughness of the samples. Under steam curing, a steeper descending branch was observed in the load-deflection curves. Compared to those under standard curing, the deflections of samples ZA0 and ZA1 increased by 0.18 mm and 0.16 mm, respectively, when steam curing was applied. Conversely, elevating the curing temperature resulted in a decrease in the deflection of sample ZA2, where the midspan displacement was reduced by 0.73 mm. This interesting phenomenon warrants further investigation in future research.

Notably, a negative synergy between high SF content and steam curing regarding ductility was revealed. While steam curing generally accelerates strength gain, it was found to induce excessive brittleness in SF-modified matrices, leading to a deterioration in energy absorption capacity. This indicates that the high-temperature curing regime typically favored for UHPC is not universally applicable to silica-rich UHSGM, where a balanced curing approach is required to maintain interfacial bonding integrity.

3.3.3 Comparison between UHSGM and UHPC

As depicted in Figure 7d, he flexural load-deflection curves of the UHSGMs are comparable to those of the UHPCs. Under steam curing, a steeper descending branch was observed for the UHPC compared to the UHSGM. Regardless of the curing conditions, the deflections recorded for the UHSGMs were higher than those of the UHPCs, implying that the UHSGMs exhibit slightly higher ductility than the UHPCs at similar compressive strength levels (130 MPa and 150 MPa). Furthermore, the discrepancy in deflection for sample ZA0 under different curing conditions was found to be more significant than that observed for the UHPCs. This indicates that the midspan displacement of the UHSGMs is slightly more sensitive to curing temperatures than that of the UHPCs.

For the toughness, as illustrated in Table 3, under steam curing, the toughness of UHSGM was very close to that of UHPC. However, after 28 days of standard curing, there was a significant difference in the toughness between UHSGM and UHPC. Not like UHSGM, standard curing had positive effect on strengthening the toughness of UHPC when the GGBFS content was at a high level. On the other hand, the geopolymerization was promoted when the curing temperature was elevated, leading to the improvement in mechanical properties of matrix.

4 Conclusion and suggestion

4.1 Conclusion

This study investigated the effects of mixture proportions—specifically SF content and various steel fiber types—on the flowability, compressive strength, and flexural properties of geopolymer concretes under steam curing and 28 days of standard curing. Comparisons were also conducted with UHPC of the equivalent strength grade. The steel fibers utilized in this investigation included a fine straight fiber (d = 0.12 mm, l = 13 mm), a thicker straight fiber (d = 0.22 mm, l = 13 mm), and a hooked-end fiber (d = 0.22 mm, l = 13 mm), all at a fixed volume fraction of 2%. It is demonstrated that this research highlights the feasibility of utilizing geopolymers synthesized from industrial by-products to manufacture ultra-high-strength concretes under various curing regimes. Based on the experimental results, the following conclusions can be drawn:

1. The flowability of fresh geopolymer mixtures was found to decreased as the GGBFS content increased. While an enhancement in flowability was induced by the addition of SF, the optimal dosage was found to be significantly associated with the GGBFS/FA ratio.

2. A reduction in flowability was induced by the incorporation of the three types of steel fibers. The most pronounced decrease in the flowability of fresh geopolymer mixtures was observed with the addition of Fiber Z, followed by Fiber S and Fiber D, respectively. Furthermore, the flowability of the UHSGM reinforced with these fibers was found to be highly comparable to that of the UHPC.

3. The compressive strength of the geopolymer concretes was found to increase with rising GGBFS content. The optimal dosage of SF was significantly influenced by the GGBFS/FA ratio. A maximum compressive strength of 115.9 MPa for the plain geopolymer was achieved at a GGBFS/FA ratio of 4:1 with the incorporation of 10% SF.

4. The incorporation of steel fibers and the application of steam curing contributed to a significant enhancement in the compressive strength of the geopolymer concretes. A maximum compressive strength of 157.0 MPa was achieved with the addition of Fiber Z and 5% SF. Notably, even without SF, the compressive strength exceeded 150 MPa at a GGBFS/FA ratio of 4:1. Furthermore, the compressive strength of the UHSGMs was found to be comparable to that of the UHPCs, irrespective of the curing conditions.

5. An improvement in the flexural strength of the geopolymers was observed with increasing GGBFS content; however, the addition of SF did not exert a consistent effect. While steel fiber reinforcement significantly enhanced the flexural strength, steam curing led to a slight reduction in strength compared to 28 days of standard curing. A maximum flexural strength of 14.5 MPa was achieved for the steel fiber-reinforced geopolymer concretes under standard curing conditions.

6. The deflection and toughness were enhanced by increasing the GGBFS content. Conversely, the addition of SF induced a reduction in deflection and led to a more serrated descending branch in the load-deflection curves. The effectiveness of elevated curing temperatures on toughness was found to be marginal and was highly dependent on the SF content.

7. The highest reinforcement efficiency regarding deflection and toughness was achieved with Fiber Z, which is attributed to the higher fiber count per unit volume. For fibers of equivalent diameter and length, the flexural properties were significantly enhanced by the mechanical anchorage effect of Fiber D (hooked-end fiber).

8. The flexural properties of the UHSGMs were found to be comparable to those of the UHPCs, irrespective of the curing conditions. Similarly, the toughness of the UHSGMs was recorded to be close to that of the UHPCs. Furthermore, the UHSGMs exhibited slightly superior ductility compared to the UHPCs at similar strength levels.

4.2 Suggestion

In summary, the mixture proportions of UHSGM require further optimization based on the findings of this study. Future research should be directed toward investigating the mechanical performance and long-term durability of UHSGM. In particular, microstructural investigations (e.g., SEM, XRD, and pore structure analysis) are necessary to quantitatively elucidate the chemical evolution of the binding phase and the topological nature of the fiber–matrix interface. Furthermore, specific attention will be given to verifying the micro-mechanisms underlying the flexural strength reduction under steam curing, such as potential interfacial debonding induced by thermal mismatch. Considering the inherent variability of industrial by-products, future work will also assess the sensitivity of the optimal mixtures to fluctuations in precursor reactivity (FA and GGBFS) to establish robust performance bounds for practical quality control.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

HH: Writing – original draft, Writing – review and editing. LY: Writing – review and editing, Funding acquisition, Resources. WJ: Writing – review and editing, Data curation. YC: Writing – review and editing, Methodology. XH: Writing – review and editing, Investigation.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research is financially supported by the Major Industrial Technology Innovation Projects (CYY-HT2023-JSJJ-0032), the Central Government–Guided Local Scientific and Technological Development Funds of the Guangxi Zhuang Autonomous Region (Achievement Transfer and Transformation category, No. ZY24212027), the Guangxi Science and Technology Major Program (No. AA24206012), and the Youth Fund of the Guangxi Natural Science Foundation (No. 2025GXNSFBA069112).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Al-Shmaisani, S., Kalina, R. D., Ferron, R. D., and Juenger, M. C. G. (2022). Assessment of blended coal source fly ashes and blended fly ashes. Constr. Build. Mater. 342, 127918. doi:10.1016/j.conbuildmat.2022.127918

CrossRef Full Text | Google Scholar

Alrefaei, Y., and Dai, J.-G. (2018). Tensile behavior and microstructure of hybrid fiber ambient cured one-part engineered geopolymer composites. Constr. Build. Mater. 184, 419–431. doi:10.1016/j.conbuildmat.2018.07.012

CrossRef Full Text | Google Scholar

Bhojaraju, C., Mousavi, S. S., and Ouellet-Plamondon, C. M. (2023). Influence of GGBFS on corrosion resistance of cementitious composites containing graphene and graphene oxide. Cem. Concr. Compos. 135, 104836. doi:10.1016/j.cemconcomp.2022.104836

CrossRef Full Text | Google Scholar

Bhutta, A., Borges, P. H. R., Zanotti, C., Farooq, M., and Banthia, N. (2017). Flexural behavior of geopolymer composites reinforced with steel and polypropylene macro fibers. Cem. Concr. Compos. 80, 31–40. doi:10.1016/j.cemconcomp.2016.11.014

CrossRef Full Text | Google Scholar

Chen, H., Qin, R., Chow, C. L., and Lau, D. (2023). Recycling thermoset plastic waste for manufacturing green cement mortar. Cem. Concr. Compos. 137, 104922. doi:10.1016/j.cemconcomp.2022.104922

CrossRef Full Text | Google Scholar

Ding, Y., Shi, C.-J., and Li, N. (2018). Fracture properties of slag/fly ash-based geopolymer concrete cured in ambient temperature. Constr. Build. Mater. 190, 787–795. doi:10.1016/j.conbuildmat.2018.09.138

CrossRef Full Text | Google Scholar

Fan, D., Zhu, J., Fan, M., Lu, J.-X., Chu, S. H., Dong, E., et al. (2023). Intelligent design and manufacturing of ultra-high performance concrete (UHPC)-A review. Constr. Build. Mater. 385, 131495. doi:10.1016/j.conbuildmat.2023.131495

CrossRef Full Text | Google Scholar

Feng, H., Bilal, I., Sun, Z., Guo, A., Yu, Z., Du, Y., et al. (2024). Mechanical and shrinkage properties of cellulose nanocrystal modified alkali-activated fly ash/slag pastes. Cem. Concr. Compos. 154, 105753. doi:10.1016/j.cemconcomp.2024.105753

CrossRef Full Text | Google Scholar

Geu, M. J., Zhuge, Y., Ma, X., and Pham, T. M. (2026). Systematic review of physical, mechanical and durability performances of metakaolin concrete. Appl. Clay Sci. 279, 108048. doi:10.1016/j.clay.2025.108048

CrossRef Full Text | Google Scholar

Guo, S., Zhang, Y., Wang, K., Bu, Y., Wang, C., Ma, C., et al. (2019). Delaying the hydration of Portland cement by sodium silicate: setting time and retarding mechanism. Constr. Build. Mater. 205, 543–548. doi:10.1016/j.conbuildmat.2019.02.041

CrossRef Full Text | Google Scholar

He, Z., Zhu, X., Wang, J., Mu, M., and Wang, Y. (2019). Comparison of CO2 emissions from OPC and recycled cement production. Constr. Build. Mater. 211, 965–973. doi:10.1016/j.conbuildmat.2019.03.289

CrossRef Full Text | Google Scholar

He, Z., Jiang, Y., Shi, J., Qin, J., Liu, D., Yalcinkaya, C., et al. (2024). Effect of silica fume on the performance of high-early-strength UHPC prepared with magnesium ammonium phosphate cement. Case Stud. Constr. Mater. 20, e03351. doi:10.1016/j.cscm.2024.e03351

CrossRef Full Text | Google Scholar

Heriyanto, G., Pahlevani, F., and Sahajwalla, V. (2018). From waste glass to building materials - an innovative sustainable solution for waste glass. J. Clean. Prod. 191, 192–206. doi:10.1016/j.jclepro.2018.04.214

CrossRef Full Text | Google Scholar

Huseien, G. F., Mirza, J., Ismail, M., and Hussin, M. W. (2016). Influence of different curing temperatures and alkali activators on properties of GBFS geopolymer mortars containing fly ash and palm-oil fuel ash. Constr. Build. Mater. 125, 1229–1240. doi:10.1016/j.conbuildmat.2016.08.153

CrossRef Full Text | Google Scholar

Kim, S.-W., Jang, S.-J., Kang, D.-H., Ahn, K.-L., and Yun, H.-D. (2015). Mechanical properties and eco-efficiency of steel fiber reinforced alkali-activated slag concrete. Materials 8, 7309–7321. doi:10.3390/ma8115383

PubMed Abstract | CrossRef Full Text | Google Scholar

Kong, Y. K., and Kurumisawa, K. (2023). Application of machine learning in predicting workability for alkali-activated materials. Case Stud. Constr. Mater. 18, e02173. doi:10.1016/j.cscm.2023.e02173

CrossRef Full Text | Google Scholar

Li, N., Shi, C., Wang, Q., Zhang, Z., and Ou, Z. (2017). Composition design and performance of alkali-activated cements. Mater. Struct. 50, 178. doi:10.1617/s11527-017-1048-0

CrossRef Full Text | Google Scholar

Li, N., Shi, C., Zhang, Z., Zhu, D., Hwang, H.-J., Zhu, Y., et al. (2018). A mixture proportioning method for the development of performance-based alkali-activated slag-based concrete. Cem. Concr. Compos. 93, 163–174. doi:10.1016/j.cemconcomp.2018.07.009

CrossRef Full Text | Google Scholar

Lin, J.-X., Liu, R.-A., Liu, L.-Y., Zhuo, K.-Y., Chen, Z.-B., and Guo, Y.-C. (2024). High-strength and high-toughness alkali-activated composite materials: optimizing mechanical properties through synergistic utilization of steel slag, ground granulated blast furnace slag, and fly ash. Constr. Build. Mater. 422, 135811. doi:10.1016/j.conbuildmat.2024.135811

CrossRef Full Text | Google Scholar

Liu, Y., Shi, C., Zhang, Z., Li, N., and Shi, D. (2020). Mechanical and fracture properties of ultra-high performance geopolymer concrete: effects of steel fiber and silica fume. Cem. Concr. Compos. 112, 103665. doi:10.1016/j.cemconcomp.2020.103665

CrossRef Full Text | Google Scholar

Liu, Y., Wang, L., Wei, Y., Sun, C., and Xu, Y. (2024). Current research status of UHPC creep properties and the corresponding applications - a review. Constr. Build. Mater. 416, 135120. doi:10.1016/j.conbuildmat.2024.135120

CrossRef Full Text | Google Scholar

Liu, Y., Ren, Y., Li, Q., and Shi, C. (2025). Influence of steel fiber hybridization on the static mechanical performance of ultra-high performance geopolymer concrete (UHPGC). Cem. Concr. Compos. 160, 106064. doi:10.1016/j.cemconcomp.2025.106064

CrossRef Full Text | Google Scholar

Luan, C., Wang, J., Gao, J., Wang, J., Du, P., Zhou, Z., et al. (2022). Changes in fractal dimension and durability of ultra-high performance concrete (UHPC) with silica fume content. Arch. Civ. Mech. Eng. 22, 123. doi:10.1007/s43452-022-00443-3

CrossRef Full Text | Google Scholar

Provis, J. L. (2018). Alkali-activated materials. Cem. Concr. Res. 114, 40–48. doi:10.1016/j.cemconres.2017.02.009

CrossRef Full Text | Google Scholar

Shen, W., Cao, L., Li, Q., Zhang, W., Wang, G., and Li, C. (2015). Quantifying CO2 emissions from China’s cement industry. Renew. Sust. Energ. Rev. 50, 1004–1012. doi:10.1016/j.rser.2015.05.031

CrossRef Full Text | Google Scholar

Shi, C., Wu, Z., Xiao, J., Wang, D., Huang, Z., and Fang, Z. (2015). A review on ultra high performance concrete: part I. Raw materials and mixture design. Constr. Build. Mater. 101, 741–751. doi:10.1016/j.conbuildmat.2015.10.088

CrossRef Full Text | Google Scholar

Sun, B., Ye, G., and de Schutter, G. (2022). A review: reaction mechanism and strength of slag and fly ash-based alkali-activated materials. Constr. Build. Mater. 326, 126843. doi:10.1016/j.conbuildmat.2022.126843

CrossRef Full Text | Google Scholar

Wang, D., Shi, C., Wu, Z., Xiao, J., Huang, Z., and Fang, Z. (2015). A review on ultra high performance concrete: part II. Hydration, microstructure and properties. Constr. Build. Mater. 96, 368–377. doi:10.1016/j.conbuildmat.2015.08.095

CrossRef Full Text | Google Scholar

Wang, X., Liu, F., Li, L., Chen, W., Cong, X., Yu, T., et al. (2024). Study on the compressive strength and reaction mechanism of alkali-activated geopolymer materials using coal gangue and ground granulated blast furnace slag. Materials 17, 3659. doi:10.3390/ma17153659

PubMed Abstract | CrossRef Full Text | Google Scholar

Wu, Z., Shi, C., and He, W. (2017). Comparative study on flexural properties of ultra-high performance concrete with supplementary cementitious materials under different curing regimes. Constr. Build. Mater. 136, 307–313. doi:10.1016/j.conbuildmat.2017.01.052

CrossRef Full Text | Google Scholar

Wu, Z., Shi, C., Khayat, K. H., and Xie, L. (2018). Effect of SCM and nano-particles on static and dynamic mechanical properties of UHPC. Constr. Build. Mater. 182, 118–125. doi:10.1016/j.conbuildmat.2018.06.126

CrossRef Full Text | Google Scholar

Yu, R., Spiesz, P., and Brouwers, H. J. H. (2014). Mix design and properties assessment of ultra-high performance fibre reinforced concrete (UHPFRC). Cem. Concr. Res. 56, 29–39. doi:10.1016/j.cemconres.2013.11.002

CrossRef Full Text | Google Scholar

Zhao, Y., Shi, T., Cao, L., Kan, L., and Wu, M. (2021). Influence of steel slag on the properties of alkali-activated fly ash and blast-furnace slag based fiber reinforced composites. Cem. Concr. Compos. 116, 103875. doi:10.1016/j.cemconcomp.2020.103875

CrossRef Full Text | Google Scholar