Mechanical properties of steel fiber concrete modified with nano-TiC and nano-SiO₂

Abstract

The incorporation of admixtures into concrete is a common method to improve its mechanical properties, and this study investigates the effects of nano-SiO2 (NS) and nano-TiC (NT) on steel fiber concrete (SFRC). In this paper, 1%, 2%, and 3% of NS, and 0.5%, 1.0%, and 1.5% of NT were individually mixed into SFRC, and their effects on the compressive, split tensile, and flexural strengths of SFRC at 7 and 28 days were investigated. The effect of mixed NS and NT on the mechanical properties of SFRC was further investigated, as was the ability of modified concrete to resist freeze-thaw damage. The modified SFRC was analyzed using a scanning electron microscope (SEM). The EDS was employed to characterize the elements. The phase composition of the modified concrete was examined using XRD. The distribution of the microporous structure in the concrete was analyzed using the mercury intrusion method. An analysis of variance was conducted using SPSS to examine the interaction between NT and NS. The experimental results showed that single doping with NT and NS had little effect on the 7-day compressive, split tensile, and flexural strengths of SFRC, but improved the 28-day compressive, split tensile, and flexural strengths of SFRC. The mechanical enhancement effect of single blending of NS in SFRC was better than that of single blending of NT, where the optimal addition of NS was 2% and that of NT was 1%. Mixing NS and NT can improve the mechanical strength of SFRC. The optimal dosage is 2% NS and 0.5% NT. Their enhancement rate of compressive strength of SFRC is -8.92%~17.48%, the enhancement rate of splitting tensile strength is -3.85%~13.46%, and the enhancement rate of flexural strength is -6.67%~12.00%. The addition of NT and NS to SFRC can improve concrete’s resistance to freeze-thaw erosion. The volcanic ash effect of NS, combined with the filling and nucleation effects of NT carbide, promotes greater consumption of Ca(OH)₂, improving the microstructure of SFRC and increasing its density. NS has a significant impact on mechanical properties throughout the entire SFRC process. Based on the results of the mercury porosimetry test, the internal pore structure of the NS2NT0.5 group was optimized accordingly.

Introduction

Concrete has a higher compressive strength and can withstand large loads, making it an ideal material for constructing large structures such as high-rise buildings, bridges, and roads¹. In addition, concrete has good durability due to its various admixtures, which can resist the effects of severe weather, natural disasters, and other environmental factors, thereby extending the service life of buildings^2–4. Therefore, concrete occupies a pivotal position in the field of building structures. However, concrete materials exhibit poor resistance to splitting and are prone to cracking under concentrated longitudinal stresses⁵. Reinforcement bars are often added to concrete in engineering, but this practice inevitably increases material costs⁶.

To improve its cracking resistance, scholars have incorporated steel fibers into concrete^7,8. The addition of steel fibers can significantly improve the splitting tensile strength of concrete, especially with longer fibers, which exhibit a more pronounced effect⁹. Steel fibers can be classified into straight, bent hook, and wave shapes. Abbass et al.¹⁰ noted that steel fibers can impede the expansion of microcracks in concrete and enhance its ductility and toughness after initial cracks form, thereby improving its mechanical strength. Yang et al.¹¹ noted that when the concrete volume fraction exceeds 2%, the enhancement of its compressive strength and modulus of elasticity by steel fibers begins to slow. Zeybek et al.¹² improved the mechanical properties of concrete by incorporating recycled steel fibers from scrap tires, thereby further reducing the cost of steel-fiber concrete. Han et al.¹³ reported that adding steel fibers to concrete improves its resistance to freeze-thaw erosion. Double-ended hook-shaped steel fibers were selected for this experiment.

Incorporating appropriate amounts of nanomaterials into concrete can also improve its service performance and resistance to freeze-thaw erosion, mainly through the microaggregate effect¹⁴. The microaggregate impact refers to the phenomenon in which the addition of fine particulate aggregates can significantly improve concrete’s mechanical and durability properties. Nano-SiO₂ (NS) has come into researchers’ field of view because it is abundant in nature and readily accessible. Samsung et al.¹⁵ stated that an appropriate amount of NS can densify the concrete microstructure, thereby improving its mechanical properties and durability. Saleh et al.¹⁶ added NS to concrete, thereby reducing its specific heat capacity and improving its thermal insulation and compressive strength. Liu et al.¹⁷ stated that an appropriate amount of NS and recycled concrete aggregate can improve the durability performance of recycled concrete. Oh et al.¹⁸ used NS to partially replace silica fume to enhance the mechanical strength of ultra-high-performance concrete.

Theory of Composite Materials Mechanics provides a theoretical basis for the combination of fibers and nanomaterials in concrete, enabling the effects of steel fibers and NS on concrete to be traced¹⁹. A number of these scholars have studied the effects of steel fibers and NS on concrete^20–22. Peng et al.²³ pointed out that steel fibers can improve the impact resistance and durability of concrete containing NS. Keshavarzian et al.²⁴ used Taguchi’s method and analysis of variance to investigate the effects of NS and steel fibers on concrete, achieving high-strength concrete. Altalib et al.²⁵ investigated the effect of NS and steel fibers on the size and compressive strength of concrete specimens. The results indicate high compressive strengths for small-sized specimens and propose a conversion relationship between the compressive strengths of non-standard and standard specimens. The durability of concrete could be enhanced through various admixtures, and the mechanisms of action of different admixtures vary significantly²⁶. For example, materials such as superplasticizers often improve durability by reducing self-shrinkage through internal curing and regulating the internal moisture state. In contrast, general nanomaterials primarily rely on the micro-aggregate filling effect and hydration nucleation to refine the pore structure, thereby improving the concrete’s microstructure. For instance, NS primarily optimizes the matrix microstructure and improves the interface transition zone through its filling effect, pozzolanic activity, and nucleation effect, thereby effectively enhancing the mechanical strength of concrete. Nano-TiC (NT), on the other hand, relies more on the “micro-aggregate effect” to improve the performance of the cement matrix.

Currently, there are relatively few studies on the incorporation of NT into concrete, leaving room for further research. Yang et al.^27,28 showed that NT exerted a filling effect in concrete, significantly improving the compactness of the cement matrix. However, no further research has been conducted on the effect of NT on concrete freeze-thaw resistance. Wen et al.²⁹ showed that 0.5% nano-CaCO₃ with 1% NT, when incorporated into steel fiber concrete, can enhance its mechanical properties and durability by improving hydration reactions and the pore structure. However, no further studies have been conducted on the relationship between the relative elastic modulus and mass loss of NT concrete. Tian et al.³⁰ indicate that an appropriate amount of NT can enhance the energy-absorbing properties of magnesium sulfate cement, effectively reducing material brittleness. NT is a nanomaterial with high hardness, electrical conductivity, and thermal conductivity, and has been widely used in various fields. Based on previous studies, appropriate amounts of NS and steel fibers can improve concrete strength (Table 1).

Table 1.

Comparison table of articles.

Author	Term of duration	Research content	Conclusion
Yang et al.27	2024	Basalt fiber (BF), secondary fly ash (FA), and NT composite-modified concrete were optimized for performance using RSM and analyzed microscopically using SEM.	The optimal dosages of NT, BF, and FA are 0.88%, 0.24% and 5.49% respectively. The RSM model has good accuracy. Appropriate admixtures can enhance the compactness and mechanical properties of concrete.
Yang et al.28	2024	BF-NS-NT was used to prepare concrete, and the slump and mechanical properties were tested. RSM was employed to optimize and validate the model.	Appropriate addition can enhance the mechanical properties of concrete. The optimal addition amounts are NT 0.85%, BF 0.11%, and NS 1.94%. The RSM model’s error ranges from 9.03% to 9.82%.
Wen et al.29	2025	Calcium carbonate NC and NT were used to partially replace cement in the modification of SFRC to investigate its mechanical properties and durability. Microscopic analysis was conducted using SEM-EDS.	The optimal mix ratio is 0.5% NC and 1% NT. NC-NT promotes hydration and optimizes the pore structure.
Tian et al.30	2025	1% to 5% micron/nanometer-sized TiC-modified magnesium oxide cement, testing multiple properties and evaluating the optimal ratio.	TiC reduces the brittleness and water resistance of cement, while carbonation weakens the flexural strength of cement.

In the field of sustainable building materials, nanoscale additives have attracted significant attention for their ability to significantly improve the microstructure and macroscopic performance of concrete. Yet, current research mostly focuses on nanoparticles such as NS and NC. NT, as a typical carbide, has unique physicochemical properties that provide new possibilities for optimizing the performance of concrete: on the one hand, NT has a nano-sized particle size with a high specific surface area, which fills the pores of the cement matrix and regulates the pore size distribution, forming the “nanoscale effect”. On the other hand, the unsaturated Ti sites on the surface of NT can adsorb Ca²⁺ generated during the early stage of cement hydration via electrostatic interactions, thereby promoting the orderly growth of C-S-H gel²⁹. Currently, some research has been conducted on the performance enhancement of cementitious materials incorporating NT fibers. However, when NS, NT, and steel fibers are added to concrete, their interfacial enhancement effects, optimal dosage ranges, and environmental impact on concrete have yet to be systematically investigated.

Based on this, this paper incorporates NT, NS, and steel fibers into concrete (Fig. 1). The effects of NT and NS on the properties of steel fiber concrete (SFRC) and its mechanical properties under combined exposure to these nanomaterials were explored. The changing patterns of compressive, split tensile, and flexural strengths of concrete were analyzed. The ability of modified SFRC to resist freeze-thaw erosion was further investigated. A scanning electron microscope (SEM) was used to observe the state of concrete after crushing, providing scholars with a theoretical basis and reference value for subsequent studies on high-strength concrete.

Fig. 1.

Modified concrete production.

Materials and methods

Materials

P.O. 42.5 Concrete was selected as the primary cementitious material for this experiment, and its main components were measured by X-ray fluorescence spectroscopy (Fig. 2). Its main properties are shown in Table 2. In addition, crushed stone is used as the coarse aggregate of concrete, and its properties are shown in Table 3. Medium sand serves as the fine aggregate for concrete, with its main properties presented in Table 4. Furthermore, a highly efficient polyhydroxy acid system was used as the water-reducing agent in this experiment, achieving a water-reducing rate of ≥ 25%.

Fig. 2.

Material: XRF image of cement.

Table 2.

Cement performance indicators.

Type of cement	Stability	Incipient condensation time/min		Compressive strength/MPa		Flexural strength/MPa
		Condensation	Condense	3d	28d	3d	28d
P.O 42.5	Eligible	280	335	24.2	46.3	5.0	6.7

Table 3.

Coarse aggregate performance indicators.

Coarse aggregate	Crushing index /%	Apparent density (kg·m − 3)	Particle size/mm	Packing density/(kg·m − 3)	Mud content/%
Crushed stone	5.4	2771	10 ~ 31	1360	0.7

Table 4.

Physical properties of sand.

Type of sand	Fineness modulus	Apparent density (kg·m − 3)	Particle size (mm)	Packing density (kg·m − 3)	Moisture content (%)	Mud content (%)
Medium	2.4	2594	≤ 3	1606	3.2	1.5

Building upon these materials, NT, NS, and steel fibers, used as modifiers in the concrete mix, were included in this experiment. The samples are shown in Fig. 3. The performance indexes of the end-hooked steel fibers are provided in Table 5. NS is a white, powder-like solid with particle sizes of 10–50 nm and a purity of 99%. NT is a black solid powder with an average particle size of 40 nm and a purity of 99%.

Fig. 3.

Materials: a NT; b NS; c Steel Fiber.

Table 5.

Performance indicators of SF.

Geometry	Tensile strength/MPa	Lengths/mm	Equivalent diameter/mm	Aspect ratio
End hook type	1000	35	1	75

Production and maintenance of modified concrete

The mixing ratios of the concrete examples are detailed in Table 6, with a water-cement ratio of 0.4. The experiment was conducted in two phases, each divided into seven groups. Seven experimental groups of steel fiber-reinforced concrete were created using NT and NS as cement substitutes, respectively. The second phase entailed the production of concrete specimens employing NT and NS composites as substitutes for cement, and due to nanoparticles’ tendency to agglomerate, resulting in inevitably increased surface energy and a significant specific surface area. Therefore, ultrasonic dispersion should be used to retreat the nanomaterials. It is more effective for enhancing the dispersion of the nanopowder materials used in the experiment. In this experiment, NT and NS were combined with water and ultrasonically dispersed for 10 min, followed by hand mixing for 2 min. Immediately add the dispersed solution to the stirred aggregate.

Table 6.

Modified concrete mix ratios.

	Number	Water (kg/m3)	River sand (kg/m3)	Gravel (kg/m3)	Cement (kg/m3)	Steel fiber (kg/m3)	NT (%)	NS (%)	Water-reducing agent (%)
Part1	CG	140.8	676	1202	348.48	40	0	0	1.0
	NS1	140.8	676	1202	344.96	40	0	1.0	1.0
	NS2	140.8	676	1202	341.44	40	0	2.0	1.0
	NS3	140.8	676	1202	337.92	40	0	3.0	1.0
	NT0.5	140.8	676	1202	346.72	40	0.5	0	1.0
	NT1	140.8	676	1202	344.96	40	1.0	0	1.0
	NT1.5	140.8	676	1202	343.2	40	1.5	0	1.0
Part2	NS1NT0.5	140.8	676	1202	343.2	40	0.5	1.0	1.0
	NS1NT1	140.8	676	1202	341.44	40	1.0	1.0	1.0
	NS1NT1.5	140.8	676	1202	339.68	40	1.5	1.0	1.0
	NS2NT0.5	140.8	676	1202	339.68	40	0.5	2.0	1.0
	NS2NT1	140.8	676	1202	337.92	40	1.0	2.0	1.0
	NS2NT1.5	140.8	676	1202	336.16	40	1.5	2.0	1.0
	NS3NT0.5	140.8	676	1202	336.16	40	0.5	3.0	1.0
	NS3NT1	140.8	676	1202	334.4	40	1.0	3.0	1.0
	NS3NT1.5	140.8	676	1202	332.64	40	1.5	3.0	1.0

NT, NS, and water-reducing agents represent the proportion of cement mass, replacing a portion of it. Water-reducing agents are solid white powders.

A forced mixer with a wet-mixing mechanism was employed. The coarse aggregate, fine aggregate, and cement were initially dry mixed for two minutes, after which a combined solution of water, NT, NS, and mixing water was added. Steel fibers were added during mixing. To ensure an even distribution of steel fibers, it was mixed for an additional 3 to 5 min. After mixing, it was placed in the mold. Following demolding, the specimens were preserved for a designated period in a controlled curing chamber at 20 ± 2 °C and 95% relative humidity³¹. Considering the cost of nanomaterials and findings from related literature, the final dosage ranges were determined as 1–3% for NS and 0.5–1.5% for NT^27–30.

Experimental mechanical properties testing instrument

The mechanical property tests of the specimens were carried out at 7 days and 28 days of curing age, respectively. According to the technical specification of the Chinese national standard GB/T 50081-2019³¹, compressive and split tensile strength tests were conducted on standard cubic specimens measuring 150 mm×150 mm×150 mm, and flexural strength tests were performed on standard prismatic specimens measuring 150 mm×150 mm×600 mm. The test equipment was a CSS-WAW1000 universal testing machine produced by Jinan Zhongluchang Testing Machine Manufacturing Co. Among them, the compressive strength test followed a continuous uniform loading rate of 50 N/s; the split tensile strength test was carried out by fixing the specimen with a positioning steel bracket at a constant loading rate of 50 N/s; and the flexural strength test was carried out by applying a continuous uniform stress to the specimen with a loading rate of 30 N/s. Three parallel specimens were set up for each test group, and the final test results were taken as the arithmetic average of the three specimens’ test values to ensure the reliability and accuracy of the test data.

Test block frost resistance and modulus of elasticity testing

The mass loss rate indicates that cementitious materials undergo a transition from a compact state to a loose, dislodged condition during freeze-thaw cycles³². The main factor contributing to mass loss is the growth of fissures due to the freezing and expansion of moisture within the specimen, which undermines the material’s integrity. The Chinese national standard GB/T50082-2009 states that a mass loss of 5% indicates insufficient frost resistance of the cementitious material³³. The ability of a material to regain its elasticity following exposure to a specific stress level is represented by its relative dynamic modulus of elasticity. This metric is essential for assessing concrete’s frost resistance, as it partially indicates the extent of damage from freeze-thaw cycles to concrete performance. This test employs a prismatic standard specimen measuring 100 × 100 × 400 mm to perform the freezing test using the fast-freezing method. After the 24-day curing period for the specimen, submerge it in water at 20 ± 2 °C for 4 days. The specimens were extracted after 28 days, their surfaces were cleaned with a moist towel, and their external dimensions were recorded. The initial weight and modulus of elasticity of the specimens were assessed using the resonance method. After testing, the specimens were placed in a specimen box, and water was added to the upper 5 mm of each specimen. The duration of each freeze-thaw cycle ranged from 2 to 4 h. Upon completion of 25 freeze-thaw cycles, the freeze-thaw tests were finalized, surface water was removed, the specimens were weighed, and the weight loss rate was calculated. The dynamic elastic modulus of the rectangular specimens was evaluated and calculated.

Specimen micro-physical inspection

After the compressive strength test, irregular fragments were removed from the specimens and employed as test samples. The fragments were submerged in anhydrous ethanol for 24 h and then dried in an oven until a stable weight was achieved before microscopic analysis. Before the test began, each sample was coated with gold under vacuum to enhance conductivity and improve image quality. The scanning electron microscope (SEM) was used to obtain micrographs of the microstructure of several fragment groupings.

XRD analysis of concrete

X-ray diffraction (XRD) testing is a key method for analyzing the phase composition of concrete and its hydration products. Based on diffraction principles, it can accurately identify major crystalline phases, including cement clinker minerals, calcium hydroxide, ettrite, and calcium silicate hydrate gel, thereby clearly characterizing the hydration process and the evolution of microstructural phases in concrete. During testing, hardened concrete specimens are ground to a specific particle size. After X-ray irradiation, diffraction patterns are obtained and compared with standard charts to perform qualitative and semi-quantitative phase analysis, providing reliable microscopic phase evidence for analyzing the mechanisms underlying the evolution of concrete properties.

Analysis of concrete mercury extraction testing

The mercury intrusion method (MIP) is a commonly used technique for analyzing the pore structure of concrete. By applying varying pressures to force mercury into the specimen’s pores, key parameters such as total porosity, pore-size distribution, critical pore size, and pore tortuosity can be quantitatively determined. It enables a systematic characterization of the quantity and distribution of capillaries, gel pores, and macropores within concrete. When combined with macro-scale performance data, such as freeze-thaw and mechanical properties, it reveals the mechanisms underlying the evolution of material durability and mechanical performance at the pore-scale, thereby effectively enhancing the evaluation system for concrete microstructure and durability.

Results

Compressive failure state of concrete

The failure states of the CG and NS2NT0.5 concrete specimens are shown in Fig. 4. Under uniaxial ultimate compression, the specimens in both the CG and NS2NT0.5 groups remained largely intact. The CG group exhibited some ductile failure. However, the specimen split into two parts along a vertical crack, and the steel fibers played a significant role in “bridging” and “suspension.” This prevented the concrete specimens from shattering and disintegrating, allowing them to retain their general shape even after failure.

Fig. 4.

Compressive failure modes of concrete: a CG; b NS2NT0.5.

In contrast, the NS2NT0.5 group exhibited superior ductile failure; the specimens did not separate at the ultimate state. Instead, they maintained higher structural integrity, with only minor surface cracks and localized spalling, demonstrating that the nanomaterials improved the steel fiber concrete’s resistance to deformation. The differences in failure modes reflect the distinct reinforcement mechanisms of steel fibers and nanomaterials. The CG group relied on steel fibers to restrict crack opening and propagation. In the NS2NT0.5 group, however, the incorporation of NS and NT not only optimized the matrix density but also significantly enhanced interfacial bond strength, greatly delaying the rate at which macroscopic cracks propagated through the specimen.

Effect of single-doped NS and NT on SFRC

The effect of a single admixture of NS and NT on the mechanical strength of SFRC is shown in Fig. 5. The experimental results show that the addition of appropriate amounts of NS and NT can improve the 28-day mechanical strength of SFRC, but the enhancement of 7-day concrete mechanical strength is not obvious. The reason may be that the initial hydration degree of the cement is relatively low, the cement matrix structure has not fully reacted, and the pore state is unstable. The pore-filling effect and interface-strengthening effect of NS and NT cannot be effectively exerted. And the bond strength between steel fibers and the cement matrix is not fully developed in the early stages, further limiting the effect of nanomaterials on early strength. With increasing dosage, the compressive, split tensile, and flexural strengths of SFRC showed a tendency to increase, then decrease. For SFRC, a single dose of NS is more effective than a single dose of NT, and the optimal addition amount of NS should be 2%, while that of NT should be 1%.

Fig. 5.

Mechanical strength of concrete at 7 and 28 days: a compressive strength; b split tensile strength; c flexural strength.

Based on the experimental results in Fig. 5, the maximum enhancement rate of NS on 28-day SFRC compressive strength was 5.24%, the enhancement rate of splitting tensile strength was 0%~9.62%, and the enhancement rate of flexural strength was − 2.67%~5.33%. The enhancement rate of NT on 28-day SFRC compressive strength was 2.10%~5.24%, the enhancement rate of splitting tensile strength was − 5.77%~1.92%, and the enhancement rate of flexural strength was − 5.77%~1.92%. The main reason for the improvement of the mechanical strength of SFRC by NS and NT is that NS and NT play a “dense material effect” inside the concrete, which fills the internal pores of the concrete and makes the concrete more compact^34–36.

However, excessive amounts of NS and NT can reduce the concrete’s mechanical strength. The main reason is that nanomaterials have a large specific surface area and surface energy due to their tiny particle size. There are strong interactions between particles, such as van der Waals and Coulomb forces, which make these nanoparticles readily attracted to each other and form agglomerates during concrete mixing. These agglomerates affect concrete hydration, thereby reducing its strength³⁷.

Effect of mixing NS and NT on mechanical properties of SFRC

Figure 6 illustrates the impact of NS and NT on the mechanical strength of SFRC. The experimental findings indicate that incorporating suitable amounts of NS and NT enhances the mechanical strength of SFRC at both 7 and 28 days. Compared with the experimental results in Fig. 6, the effect of NS and NT is more pronounced in improving concrete strength. As the dosage increased, the impact on the compressive, split tensile, and flexural strengths of SFRC at 28 days showed an initial increase, followed by a subsequent decrease. The ideal increments for compressive, split tensile, and flexural strengths were 2% of NS and 0.5% of NT.

Fig. 6.

Mechanical strength of concrete at 7 and 28 days: a compressive strength; b split tensile strength; c flexural strength.

Based on the experimental results in Fig. 6, the enhancement rates of mixed NS and NT on 28-day SFRC compressive strength ranged from − 8.92% to 17.48%, split tensile strength from − 3.85% to 17.31%, and flexural strength from − 6.67% to 12.00%. Compared to mixing NS and NT alone in SFRC, mixing NS and NT exhibited higher mechanical strength. This is mainly due to NS’s volcanic ash effect on concrete, which produces more C-S-H gel and improves concrete’s mechanical strength³⁸. And with the incorporation of NT, it further compensates the internal defects of concrete and makes the stress distribution inside the concrete more uniform³⁹. However, excessive NS and NT still reduce the mechanical strength of concrete.

Analysis of mass loss rate of concrete

Figure 7 depicts the experimental results of mass reduction in SFRC incorporating NS and NT after a 28-day curing period under standard conditions, followed by freeze-thaw cycles. The mass loss rate of concrete escalates with the frequency of freeze-thaw cycles. This process occurs during the freeze-thaw cycle, as free water within the concrete’s microscopic pores generates fatigue stress due to the combined effects of expansion pressure and osmotic pressure. Fatigue stress induces microcracks in concrete, leading to surface-layer spalling that further exacerbates mass loss⁴⁰.

Fig. 7.

Mass loss rate after freeze-thaw cycles: a single blending of NS and NT; b mixed blending of NS and NT.

Figure 7a shows that moderate incorporation of NS can markedly reduce the mass loss rate of SFRC. Throughout the entire freeze-thaw cycle, the mass loss rate of NS2 samples was much lower than that of the control CG. Using the correct amount of NS to replace cement can significantly reduce Ca(OH)₂ production. Furthermore, NS exhibits a filling effect, increasing the matrix density⁴¹. The mass loss rate of NS3 specimens continuously surpassed that of the control CG, while the mass loss rate of NS2 exceeded that of NT1 during 50 to 150 freeze-thaw cycles. This primarily results from the formation of a dense matrix by a substantial quantity of NS, which inhibits the effective dissipation of internal free-water expansion pressure, leading to surface spalling of the specimen^42,43.

The results in Fig. 7b show that the NS2NT0.5 group exhibits excellent freeze-thaw resistance during 0-150 freeze-thaw cycles, and its mass loss rate is always lower than that of the baseline group CG, the single-doped NS2 group, and the single-doped NT1 group. Moreover, NS2NT0.5 showed a negative growth trend at 25 freeze-thaw cycles. This phenomenon may be attributed to the high water-cement ratio of the specimens, resulting in a relatively loose structure that allowed them to absorb external water during the freeze-thaw cycle, leading to mass increase⁴⁴. In contrast, the mass loss rates of both NS3NT1 and NS3NT1.5 specimens were higher than that of the benchmark group CG, indicating that when the dosage of NS exceeds 3% and that of NT exceeds 1%, their synergistic effect will have a negative effect on the SFRC, leading to an increase in the deterioration of the internal structure of the material.

The test findings indicate that incorporating NT can decrease the mass loss rate of concrete and markedly enhance the durability of concrete specimens. Over 0 to 150 freeze-thaw cycles, the mass loss rates of NT0.5, NT1, and NT1.5 were lower than those of the CG group. Under identical quality settings, NT concrete is more efficacious than SFRC concrete with NS alone in mitigating mass loss resulting from freeze-thaw damage.

Relative dynamic elastic modulus analysis of concrete

Figure 8 illustrates the change in the relative dynamic elastic modulus of SFRC (NS and NT) subjected to varying numbers of freeze-thaw cycles following 28 days of conventional curing. The elastic modulus of modified SFRC concrete decreases with increasing numbers of freeze-thaw cycles. The principal cause is the repeated freezing and expansion of free water within the concrete in a freeze-thaw climate, which exacerbates existing cracks and generates new microcracks, ultimately resulting in the deterioration of the material’s mechanical qualities^45,46.

Fig. 8.

Relative dynamic elastic modulus of concrete after freeze-thaw cycles: a single mix of NS and NT; b mixed mix of NS and NT.

The results in Fig. 8a show that after 0 ~ 150 freeze-thaw cycles, the relative dynamic elastic modulus of both NS2 and NT1 groups is higher than that of the reference group CG. With increasing freeze-thaw cycles, the attenuation in each group continues to increase. This is mainly because the filling effect of NS and NT optimizes the concrete pore structure, and the uniformly distributed steel fibers inhibit crack development and improve the concrete’s resistance to freeze-thaw erosion^47,48. It is noteworthy that the NS2 group maintained the highest relative dynamic modulus of elasticity after 150 cycles, at 90.123%. While the NS3 group had the lowest with 88.987%. The results in Fig. 8b indicate that the relative kinetic elastic modulus of the NS2NT0.5 group was always higher than the other groups in 0 ~ 150 freeze-thaw cycles, which suggests that the synergistic effect of an appropriate amount of NS and NT has a delaying effect on the freeze-thaw damage of SFRC⁴⁹.

Figure 9 shows the relationship between the mass loss rate and the relative elastic modulus of modified concrete, indicating a negative correlation. As the number of freeze-thaw cycles increases, the mass loss rate of concrete continues to rise. At the same time, the relative elastic modulus decreases simultaneously (the greater the mass loss, the larger the decrease in elastic modulus). This correlation is a core characteristic of freeze-thaw damage—cracks formed within the concrete due to water expansion during freezing cause both mass loss and a reduction in the material’s stiffness. Specifically, nanomaterial modification can mitigate the strength of this negative correlation between mass loss and elastic modulus reduction. Single-modified groups (e.g., NS2, NT1) exhibited smaller increases in mass loss rate compared to the control group, accompanied by a slower decline in relative elastic modulus. The hybrid group NS2NT0.5 demonstrated a stronger regulatory effect.

Fig. 9.

Elastic modulus and mass loss of modified concrete: a single addition of NT and NS; b blended addition of NT and NS.

After 150 freeze-thaw cycles, the mass loss rate of the CG reference group was 1.996%, with a relative dynamic modulus retention rate of 89.125%. When NS was added alone, a 1–2% content reduced mass loss by over 15% compared to CG, whereas a 3% NS content caused deterioration in freeze resistance due to agglomeration. For NT alone, a 1% content yielded the best results, reducing mass loss by 21.3% compared to CG. The optimal combination of NS and NT was NS2NT0.5 (NS 2% and NT 0.5%), with a mass loss rate of only 1.299% (34.9% lower than CG) and a relative dynamic modulus retention rate of 92.683%, demonstrating the best freeze resistance among all groups. After 150 freeze-thaw cycles, the mass loss rates (maximum 2.435%) and relative dynamic modulus retention rates (minimum 88.623%) of all groups remained well below the specification failure thresholds. The degradation patterns were clearly defined, with significant intergroup differences.

Additionally, the NS2NT0.5 group exhibited a slight negative mass loss rate of -0.032% during the initial freeze-thaw phase (25 cycles). This may be attributed to the water-saturated pretreatment in the freeze-thaw test, which fully saturated the capillary and gel pores within the concrete matrix. During the initial freeze-thaw cycles, specimens exhibited no significant surface spalling or internal cracking. These pore waters proved difficult to fully expel within a short timeframe, resulting in the specimens maintaining a higher apparent mass than their initial dry state and leading to net mass increase.

Effect of mixing NS and NT on mechanical properties of SFRC

The two-way ANOVA (Table 7; Fig. 10) clearly reveals differential regulatory patterns of NT and NS, as well as their interaction, on the mechanical properties of modified concrete at 7 and 28 days. For the core compressive strength metric, both NT and NS exerted extremely significant effects at the 7-day age, accompanied by a highly substantial interaction effect, indicating synergistic gains from dual admixture in the early stage. However, at 28 days, only NS retains a highly significant impact, while NT’s influence drops to a considerable level and their interaction disappears, reflecting that single factors dominate later-stage compressive strength development. Regarding splitting strength and flexural strength, NS exhibits a highly significant effect throughout the entire process. In contrast, NT and its interaction with NS make no substantial contribution, highlighting NS’s greater impact on concrete flexural performance.

Table 7.

Two-factor significance analysis.

Categories	Source	7 days					28 days
		Sum of squares	Degrees of freedom	Mean square	F-value	P-value	Sum of squares	Degrees of freedom	Mean square	F-value	P-value
Compressive strength	Modified model	3.18.156	15	21.210	10.180	< 0.001	631.346	15	42.09	6.214	< 0.001
	Intercept	82261.8	1	82261.8	39481.716	< 0.001	174230.95	1	174230.95	25724.656	< 0.001
	NS	180.904	3	60.301	28.942	< 0.001	454.414	3	151.471	22.364	< 0.001
	NT	55.799	3	18.6	8.927	< 0.001	65.929	3	21.976	3.245	0.035
	NT and NS	81.454	9	9.05	4.344	< 0.001	111.004	9	12.334	1.821	0.103
	Error	66.673	32	2.084			216.733	32	6.773
	Total	82646.63	48				175079.03	48
	Total after adjustments	384.83	47				848.08	47
Splitting strength	Modified model	7.246	15	0.483	3.589	0.001	5.12	15	0.341	2.199	0.03
	Intercept	841.687	1	841.687	6254.025	< 0.001	1408.333	1	1408.333	9073.826	< 0.001
	NS	4.751	3	1.584	11.767	< 0.001	3.985	3	1.328	8.558	< 0.001
	NT	0.454	3	0.151	1.125	0.354	0.242	3	0.081	0.519	0.672
	NT and NS	2.041	9	0.227	1.685	0.134	0.893	9	0.099	0.64	0.755
	Error	4.307	32	0.135			4.967	32	0.155
	Total	853.21	48				1418.42	48
	Total after adjustments	11.553	47				10.087	47
Flexural strength	Modified model	2.971	15	0.198	1.512	0.159	7.723	15	0.515	3.603	0.001
	Intercept	1741.225	1	1741.225	13287.569	< 0.001	2778.563	1	2778.563	19441.843	< 0.001
	NS	1.684	3	0.561	4.284	0.012	5.188	3	1.729	12.101	< 0.001
	NT	0.471	3	0.157	1.197	0.327	0.863	3	0.288	2.014	0.132
	NT and NS	0.817	9	0.091	0.693	0.71	1.672	9	0.186	1.3	0.275
	Error	4.193	32	0.131			4.573	32	0.143
	Total	1748.39	48				2790.860	48
	Total after adjustments	7.165	47				12.297	47

P < 0.05 indicates a significant effect, P < 0.01 indicates a highly significant effect.

Fig. 10.

Significance analysis plot: a 7-day effect analysis; b 28-day effect analysis.

From an engineering application perspective, these patterns provide clear, targeted design strategies for optimizing the mix design of nano-modified steel fiber concrete. If the engineering scenario prioritizes early-stage concrete strength and compressive performance, the dosages of NT and NS should be co-regulated to leverage the synergistic gain from their early-stage dual addition fully. For projects prioritizing long-term crack resistance and flexural performance, since NT exhibits no significant regulatory effect on these properties and carries higher material costs, optimization should focus solely on NS dosage. NT dosage can be flexibly adjusted based on project cost control and workability requirements. Regarding the long-term durability of modified concrete, the freeze-thaw resistance of modified steel fiber concrete performs best when the NS dosage is 2%, and the NT dosage is 0.5%. Furthermore, the differing age-related effects of NT and NS on concrete mechanical properties suggest that differentiated dosage designs are necessary for modified concrete in various service scenarios, such as components requiring early strength or those subjected to long-term bending. This approach ensures precise fulfillment of required mechanical and durability properties while achieving optimal economic efficiency in concrete mix design.

Mercury-in-pressure test on modified concrete

Pore structure is the core microscopic basis for determining the mechanical and durability properties of concrete, and the mercury intrusion porosimetry (MIP) method is a testing technique used to characterize concrete’s porosity, pore-size distribution, and other properties. This study combines results from concrete mechanical properties and freeze-thaw durability tests, while also considering the cost-effectiveness of the test materials. Specifically, MIP testing and analysis were conducted on the fragments from the 28-day compressive strength test of the reference group CG and NS2NT0.5, as well as on specimens after 150 freeze-thaw cycles. The test results are shown in Table 8, and the mercury intrusion pore size distribution curves are shown in Fig. 11.

Table 8.

Microscopic pore size parameters of the material.

Test specimen	Specific volume (ml/g)	Porosity (%)	Average diameter (µm)	Median particle diameter (µm)	Volume fraction of pore size distribution (%)
					0 ~ 20 nm	20 nm~50 nm	50 nm~200 nm	> 200 nm
CG	0.0280	6.1660	0.01553	0.02746	45.56	13.17	7.04	34.22
NS2NT0.5	0.0309	6.9114	0.01401	0.02122	50.03	10.98	12.45	26.53
CG*	0.0808	17.1385	0.06960	2.88472	9.33	11.39	10.76	68.51
NS2NT0.5*	0.0606	13.1177	0.11795	1.74050	3.59	14.55	16.99	64.86

“*” indicates test specimens that have undergone 150 freeze-thaw cycles.

Fig. 11.

Pore size distribution: a, b CG; c, d NS2NT0.5; e, f CG*; g, h NS2NT0.5*; i Pore size distribution volume ratio.

The total porosity of NS2NT0.5 increased only slightly compared to the reference group CG, but the pore-size distribution improved. The proportion of harmful large pores (greater than 200 nm) decreased significantly from 34.22% to 26.53%. In contrast, the proportion of harmless micro-pores (0–20 nm) increased from 45.56% to 50.03%, and the median volume diameter decreased from 0.02746 μm to 0.02122 μm. These results indicate that co-blending NS and NT improved the concrete’s internal structure. Although the total pore volume increased slightly, the large pores—which act as stress concentration points—were significantly refined into harmless micropores. At the same time, the high pozzolanic activity of NS consumed Ca(OH)₂ via secondary hydration, forming a denser C-S-H gel that improved the transition zone between the aggregate and the cement paste (ITZ) and enhanced the matrix’s load-bearing capacity at the microscopic level. This is consistent with the improved mechanical and freeze-thaw properties observed in the NS2NT0.5 group during macroscopic mechanical performance testing.

After 150 freeze-thaw cycles, the pore structure of the control group (CG) deteriorated significantly. The total porosity rose sharply to 17.1385%, an increase of 178% compared to the untreated CG group. The proportion of harmful large pores (greater than 200 nm) surged to 68.51%, while the proportion of harmless micropores (0–20 nm) plummeted to 9.33%. Concurrently, the median pore diameter soared to 2.88472 μm. These results indicate that the freeze-thaw stress generated by the freezing of pore water during freeze-thaw cycles caused the formation of a large number of interconnected microcracks within the matrix, transforming the pore structure from a “microporous-dominated” dense structure to a “macroporous-dominated” loose structure.

In contrast, the degree of pore degradation in NS2NT0.5 after freeze-thaw cycles showed some improvement. Its total porosity was 13.1177%, a 23.46% decrease compared to the CG group. The total pore volume decreased to 25.00%, and the median pore diameter decreased to 1.74050 μm. At the same time, the proportion of large, harmful pores (> 200 nm) decreased to 64.86% compared with the CG group. These results indicate that the blended addition of NS and NT effectively mitigated the destructive effects of freeze-thaw stresses on the matrix, reduced the formation of harmful large pores, and enhanced the concrete’s freeze-thaw resistance at the microscopic level.

Microstructural analysis of modified concrete

The micropores, fissures, and crystalline structure within the concrete, along with the interfacial transition zone (ITZ) between aggregates, significantly influence the strength and durability of concrete (Figure 12). Among these features, the formation of the ITZ is particularly important, as it results from the synergistic interaction of multiple factors, including differences in material composition, physical effects, chemical hydration, and process conditions. The core origin of the ITZ lies in the multidimensional inhomogeneity of the interface region during concrete preparation and hardening. Specifically, the “wall effect” on aggregate surfaces leads to sparse cement particle packing, while the adsorbed water film further increases the local water-cement ratio, forming an initial pore enrichment zone. This localized high water-cement ratio at the interface, coupled with constraints on aggregate surfaces, promotes preferential orientation of hydration products (e.g., Ca(OH)₂) during growth. As a result, these products exhibit lower hydration levels than the surrounding paste matrix, exacerbating structural porosity. Furthermore, mismatches in thermal expansion coefficients and volumetric deformation between aggregates and paste generate internal stresses at the interface, which catalyze microcrack formation. The overall distribution condition and morphology of several phases in concrete, including calcium aluminate and calcium hydroxide in cement hydration products as well as unhydrated cement particles, also influence the mechanical characteristics of concrete^50–52. Building on this, Zhang et al.⁵³ asserted that incorporating NS into cement enhances the ITZ of concrete, hence augmenting its mechanical strength. In line with this, Yang et al.³⁷ reported that the inclusion of NT enhanced concrete densification and improved the ITZ of the material.

Fig. 12.

Concrete interface transition zone.

The SEM of 28-day-old concrete reveals a dense microstructure with a substantial C–S–H gel matrix and tightly interwoven multi-phase components. The C–S–H gel forms a continuous amorphous network with a dense, resilient surface. It intertwines and bonds tightly with residual hydration products, such as trace acicular calcium aluminate hydrate crystals and incompletely hydrated cement clinker particles. Together, these form the core framework of the concrete microstructure. By 28 days, the hydration process had reached a stable phase. Some early-formed calcium hydroxide (Ca(OH)₂) crystals had been consumed through secondary hydration reactions or encapsulated within the dense C–S–H gel network. This makes them difficult to discern clearly in the micrographs (Fig. 13).

Fig. 13.

28-day SEM images of concrete: a CG; b EDS spectrum of CG; c NS2; d EDS spectrum of NS2; e NT1; f EDS spectrum of NT1; g NS2NT0.5; h EDS spectrum of NS2NT0.5.

Notably, the incorporated SF appears as elongated strands in the micrograph, exhibiting strong interfacial bonding with the C–S–H gel and other hydration products⁵⁴. The gel network tightly envelops the fiber surfaces, with some gel even infiltrating minute pores on the fibers, thereby effectively enhancing the interfacial bonding strength between the fibers and the matrix and preventing interfacial delamination. Meanwhile, NS acts as an active admixture, participating in chemical reactions that generate additional C–S–H gel during hydration. Through a filling effect, NS also refines the pores at the microscopic scale, resulting in a denser, more uniform C–S–H gel network⁵⁵. In addition, NT, dispersed uniformly as nanoscale particles within the C–S–H gel matrix, partially adsorbs onto the gel surface or fills minute voids within the gel network, thereby further enhancing the microstructure’s compactness and integrity.

Figure 13 shows the SEM pictures of concrete after 28 days of crushing. Figure 13a shows micrographs of the benchmark SFRC group with steel fibers alone, from which it can be seen that the concrete as a whole is dense, but it still contains a small number of pores and cracks. The ITZ section is obvious, mainly because the surface properties of the aggregate-cement paste contact area differ from those of the cement paste proper, resulting in a unique section. Figure 13b and c show microstructure images of SFRC with 2% NS and 1% NT alone, respectively. The concrete as a whole tends to be denser, and the ITZ is improved with progressively fewer pores and fissures compared to Fig. 13a. Figure 13d shows the SFRC microstructure images for 2% NS and 1% NT; the concrete appears denser in both cases. The synergistic effect of NS and NT can significantly improve the microstructure of SFRC.

The multi-phase interwoven structure—characterized by C–S–H gel dominance, steel fiber reinforcement, and nanoparticle filling and modification—observed in the 28-day concrete micrographs significantly reduces internal porosity and connectivity. This effectively suppresses microcrack initiation and propagation, providing a robust microstructural foundation for synergistically enhancing the macro-mechanical properties of concrete.

The EDS spectrum and quantitative elemental analysis results indicate (Fig. 13; Table 9) that the elemental composition of the control group (CG) is dominated by the characteristic elements O, Ca, and Si from cement hydration products. It also contains Fe introduced by steel fibers, with no detectable Ti. In the single-doped nano-silica (NS2) specimens, the mass fraction of Si increased significantly from 10.22% in the control group to 18.18%, while the mass fraction of Ca decreased from 11.09% to 5.87%. This change directly confirms the pozzolanic effect of nano-SiO₂: its reaction with the cement hydration product Ca(OH)₂ produces more C-S-H gel with a lower calcium-to-silica ratio. The energy spectrum of the single-blended nano-titanium carbide (NT1) specimen exhibited distinct Ti characteristic peaks. Quantitative analysis revealed a Ti mass fraction of 0.43%, with Si and Ca contents showing no significant deviation from the baseline group. This indicates that nano-titanium carbide primarily functions as a filler and nucleation agent, exerting limited influence on regulating the calcium-silicon ratio of hydration products. In the NS2NT0.5 specimen, both Si (18.88%) and Ti (1.04%) exhibited the highest mass fractions among the four groups, while Ca dropped to the lowest value of 3.34%. This revealed the modification mechanisms of the two nanomaterials: the volcanic ash effect of NS, with filling and nucleation effects of NT carbide, enabling more complete consumption of Ca(OH)₂ and the formation of dense hydration products. This aligns with the improved macroscopic mechanical and durability properties of the modified composite concrete.

Table 9.

Table of energy spectrum elements.

Elements	CG/%	NS2/%	NT1/%	NS2NT0.5/%
C	12.19	13.36	12.96	17.44
O	60.98	61.68	58.98	57.04
Na	0.83	0.14	0.92	0.21
Mg	0.49	0.05	0.52	0.11
Al	2.44	0.41	2.43	0.86
Si	10.22	18.18	10.24	18.88
S	0.61	0.11	0.34	0.19
K	0.32	0.08	0.32	0.31
Ca	11.09	5.87	11.57	3.34
Ti	-	-	0.43	1.04
Fe	0.83	0.11	0.89	0.58

Figure 14 shows the XRD diffraction patterns of the control group (CG), NT1, NS2, and NS2NT0.5 modified concrete. Characteristic phases such as quartz, sodium feldspar, calcite, and hematite were clearly detected in all four groups of specimens. Compared with the CG group, the intensities of the characteristic peaks for Ca(OH)₂ and calcite in the NT1 group were slightly reduced. This is because NT provides more nucleation sites for cement hydration through the “nucleation effect,” while simultaneously exerting a micro-filling effect to optimize the paste’s microstructure, thereby promoting cement hydration to a certain extent. In the NS2 group, the peak intensities of Ca(OH)₂ and calcite were significantly lower than those in the NT1 group. This is attributed to the pozzolanic effect of NS, which efficiently consumes the Ca(OH)₂ generated during cement hydration, thereby forming a denser C-S-H gel and improving the concrete’s microstructure. The NS2NT0.5 group exhibited the lowest Ca(OH)₂ and calcite peak intensities among the four groups, reflecting the synergistic effect of NS’s pozzolanic effect and NT’s nucleation and filling effects, which further accelerated cement hydration and optimized the structure of the paste-aggregate interface transition zone.

Fig. 14.

XRD analysis of concrete.

Conclusions

In this paper, NS and NT were blended into SFRC, and their effects on the compressive, split tensile, and flexural strengths of SFRC at 7 and 29 days were investigated. The microstructure of concrete was also analyzed, and the main conclusions are as follows.

1. The single mixing of NS and NT has little effect on the 7-day mechanical strength of SFRC, but it improves the 28-day mechanical strength to some extent. For SFRC, the effect of single mixing NS is better than that of single mixing NT. The optimal additions of NS and NT should be 2% and 1%, respectively. The highest improvement rate of NS on 28-day SFRC compressive strength is 5.24%, the improvement rate of split tensile strength is 0%-9.62%, and the improvement rate of flexural strength is -2.67%-5.33%. The improvement rate of NT on 28-day SFRC compressive strength is -2.67%-5.33%. The improvement rate of NS on 28-day SFRC compressive strength is -2.67%-5.33%. The improvement rate of NS on 28-day SFRC compressive strength is 2%-9.62%. compressive strength was 2.10%~5.24%, split tensile strength was − 5.77%~1.92%, and flexural strength was − 8.00%~1.33%.

2. The mixing of NS and NT affected the mechanical strength of SFRC at 7 and 28 days. The maximum increase in compressive strength of 7-day SFRC by mixing NS and NT was 9.30%, the maximum increase in split tensile strength was 4.25%, and the maximum increase in flexural strength was 4.83%. The enhancement rates for 28-day SFRC compressive strength ranged from − 8.92% to 17.48%, split tensile strength from − 3.85% to 13.46%, and flexural strength from − 6.67% to 12.00%.

3. Adding NT and NS to SFRC can improve concrete’s resistance to freeze-thaw erosion. In the case of a single mixing of NT and NS, the optimum mixing amount of NS is 2% and that of NT is 1%. The optimum admixture of mixed NT and NS is also 2% NS and 0.5% NT.

4. A two-way ANOVA indicates that NT and NS exhibit a synergistic effect on compressive strength at the 7-day stage. After 28 days, NS alone becomes dominant, and only NS significantly influences splitting and flexural strength.

5. Mixing NS and NT in concrete can improve its microstructure, thereby increasing its mechanical strength.

At present, the basic mechanical properties and resistance to freeze-thaw erosion of NS, NT, and SFRC have been investigated in this paper, but there is still much room for research. In the future, the effects of NS and NT on SFRC impact resistance can be explored to provide a theoretical basis and research value for the study of concrete.

Author contributions

Conceptualization, X.H. and X.Y.; methodology, X.H., T.B. and B.C.; software, X.Y.; validation, X.H. and Z.W.; formal analysis, X.H., Z.W. and K.L.; investigation, X.Y.; resources, Z.W., B.C. and Z.W.; data curation, X.Y. and Z.W.; writing—original draft preparation, Z.H., X.Y., and K.L.; writing—review and editing, X.H. and Z.W.; visualization, X.Y. and Z.W.; supervision, Z.W.; project administration, X.H. and Z.W.; funding acquisition, Z.W. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Heilongjiang Provincial Key R&D Program Guidance Category Scientific Research Projects (GZ20220138), Scientific Research Program of the Department of Ecology and Environment of Heilongjiang Province (HST2022GF004).

Data availability

The available figures presented in this research are obtainable upon contact with the corresponding authors. Being open to the scientific public, these digital data will be part of our continuing investigation.

Declarations

Competing interests

The authors declare no competing interests.