Abstract
High-performance cementitious composites (HPCC) are prominently featured with high tensile ductility and toughness. Slag has been widely used in HPCC; however, HPCC with high volumes of slag has low matrix strength and limited development of micro-structure at early-age. These limitations can be mitigated by incorporating nano-particles (e.g., nano-silica) in the binder. The purpose of this study was to develop nano-modified HPCC with high ductility and matrix quality. A new form of basalt fibers termed basalt fiber pellets (BFP)—basalt fiber strands encapsulated by a polymeric resin—were used at different dosages (2.5% and 4.5% by volume), and in a hybrid system with PVA fibers (1% by volume) to develop in these composites. All composites incorporated a binder consisting of 50% general use cement and 50% slag with the addition of 6% nano-silica. The composites were tested in relation to compressive strength and flexural performance. All the nano-modified composites showed improved performance, especially at early-age, despite the high volume of slag incorporated in the binder. While the compressive strength of the mixtures was reduced with increasing the dosage of BFP, addition of 1% PVA fibers to BFP (hybrid system) enhanced the compressive strength of the composites. In the same context, the flexural performance of the composites comprising hybrid fibers was also improved in relation to flexural strength, post-cracking behavior, residual strength and toughness. Therefore, these composites have a promising potential for infrastructure applications requiring improved strength and ductility.
Numerous studies have been conducted to resolve the technical limitations (e.g., cracking, low ductility/toughness) associated with conventional concrete, resulting in the development of high-performance fiber reinforced cementitious composites. For instance, engineered cementitious composites, which are based on the principles of micromechanics, comprise controlled volumes of fibers (e.g., polyvinyl alcohol [PVA]) and have high toughness/ductility with pseudo-strain-hardening behavior ( 1 ).
Different types of fibers have been used to improve the plastic and hardened properties of cement-based mixtures, such as steel, natural and synthetic fibers, through different mechanisms such as reduction of bleeding and controlling the growth of cracks ( 2 ). In addition, the incorporation of fibers in cementitious composites can limit the width and intensity of cracks in the cementitious matrix, depending on the type, dosage, properties and geometry of the fibers used. Many studies have been conducted in the area of hybrid fibers as their performance exceeds that of single fibers, in relation to synergistic/complementary effects ( 2 – 4 ) based on factors such as fiber dimensions and function ( 2 , 5 , 6 ). For instance, one type of fibers in the hybrid system may control the growth of cracks in the cementitious matrix, thus increasing the ultimate strength, while the other type enhances the strain limit at failure in the post-cracking zone. Smaller fibers may provide effective restraint to micro-cracks and delay their initiation and growth to unstable macro-cracks, while larger fibers impede the growth and propagation of macro-cracks to improve the ductility of the composite ( 7 ). In addition, the strength and toughness retention properties of composites with age are enhanced by the presence of durable fibers resistant to surrounding exposure conditions ( 8 , 9 ) .
Since 2010, basalt fibers have emerged in the concrete industry, owing to their mechanical properties and low cost compared with other available alternatives such as glass fibers. Basalt fibers have high tensile strength (3,000–4,000 MPa) and reasonable modulus of elasticity (93–110 GPa) as well as high thermal and corrosion resistance ( 10 ). Furthermore, basalt fibers showed good bonding characteristics with cementitious matrices up to 28 days ( 10 – 12 ). However, it was reported that this bond was diminished within 90 days, because of the reaction between the silicate component in basalt fibers and alkaline pore solution, resulting in the loss of strength and durability of the reinforced composites ( 13 ). This vulnerability to alkaline media can be mitigated by encapsulating the fibers in polymeric resin, for example, vinyl ester or polyamide, resulting in a new form of basalt fibers termed basalt fiber pellets.
Numerous studies have investigated the partial replacement of cement with supplementary cementitious materials (SCMs), for example, slag, to improve the properties of the cement-based mixtures, as they react with calcium hydroxide and produce secondary calcium silicate hydrate (C-S-H) ( 14 ). Nevertheless, slow reactivity SCMs delay the hardening and strength development of cement-based materials, but these limitations can be alleviated by the incorporation of nano-silica in the cementitious binder ( 15 – 17 ). Nano-silica is a synthetic product of porous and nearly spherical particles. It positively affects the micro-structure of the cementitious materials, and thus their strength and durability are significantly improved. It can speed up cement hydration at early ages through providing nucleation sites for early hydration products (nucleation effect). In relation to the chemical effect (pozzolanic reaction), nano-silica reacts with calcium hydroxide to form C-S-H gel, resulting in remarkable development of the micro-structure ( 17 , 18 ). Moreover, its ultrafine nature makes it act as a filler (physical effect) blocking the very fine pores in the cementitious paste, resulting in decreasing the permeability and density of the cementitious matrix ( 16 , 18 , 19 ).
Ongoing research at the University of Manitoba has been conducted to investigate the ability of utilizing the novel form of basalt fibers, basalt fiber pellets (BFP), as a single fiber system in sustainable nano-modified cementitious composites at different dosages (2.5%, 4.5% and 6.9% by volume), comprising 50% cement, 50% slag and 6% nano-silica as an additive. The developed composites showed promising results in relation to mechanical and durability performance ( 16 , 17 ). However, the developed composites experienced considerable reduction in the first-peak flexural strength that reached up to 30%, especially with the higher dosages of BFP (4.5% and 6.9%). On the other hand, no data is available on the incorporation of BFP (macro-fibers) in cementitious composites comprising other types of fibers (e.g., micro-fibers) forming a hybrid system to investigate its effects on the mechanical performance of the composites. Therefore, the primary objective of this study was to develop nano-modified cementitious composites comprising a high volume of slag, reinforced with single and hybrid systems of BFP (macro-fibers) and PVA (micro-fibers). The compressive strength and flexural performance of these composites were investigated in this study, and the findings were corroborated by microscopy analysis.
Methods
Materials
General use cement (GU) and slag meeting CSA-A3001 ( 20 ) requirements were used as the main components of the binder (Table 1). In addition, a commercial nano-silica solution (NS) was added to the binder; this solution comprises 50% SiO2 particles dispersed in an aqueous solution. The mean particle size of NS is 35 nm (D50), which represents 50% in the cumulative distribution. The NS specific surface, viscosity, density and pH values are 80 m2/g, 8 cP, 1.1 g/cm3, and 9.5, respectively. Locally available fine aggregate, with a continuous gradation of 0–600 μm and fineness modulus of 2.9, was used in the mixtures. The absorption and specific gravity of the fine aggregate is 1.5% and 2.6, respectively. A high-range water-reducing admixture, poly-carboxylic acid-based complying with ASTM C494 Type F ( 21 ), was added to improve the workability of the mixtures. Macro-BFP and micro-PVA fibers were used to reinforce mixtures. BFP of 36 mm length (Figure 1a and Table 2) were added to mixtures at different dosages of 2.5% and 4.5% by volume, equivalent to basalt fiber volumes of 1% and 2%. The BFP are made of 16 μm basalt roving encapsulated by polyamide resin, and the fiber component represents 60% of the pellet by mass. The PVA length is 12 mm (Figure 1b and Table 2), and it was used at a fixed dosage of 1% by volume.
Chemical and Physical Properties of General Use (GU) Cement and Slag
Physical and Mechanical Properties of Basalt Fiber Pellets (BFP) and Polyvinyl Alcohol (PVA) Fibers

Reinforcing fibers: (a) basalt fiber pellets and (b) polyvinyl alcohol micro-fibers.
Proportions and Mixing Procedures
Six mixtures were prepared at water to binder ratio (w/b) of 0.3, using 50% GU cement and 50% slag by mass of the base binder (700 kg/m3). A single dosage of nano-silica, 6% (42 kg/m3 SiO2), was added to five mixtures (i.e., ternary binder content of 742 kg/m3). This dosage was used to enhance the hardened properties of composites as shown in previous studies at the University of Manitoba ( 15 , 17 ). Four mixtures were reinforced with single dosages of macro-BFP (2.5% and 4.5% by volume) and micro-PVA (1% by volume) fibers, while the other two mixtures were reinforced with hybrid fibers including 2.5% or 4.5% BFP combined with 1% PVA. The proportions of the six mixtures are listed in Table 3. For the mixture IDs, the letters G, N, B and V refer to slag replacement, nano-silica, BFP and PVA respectively, while the numbers indicate the dosage of nano-silica and fibers. The constituent materials were mixed in a concrete pan mixer with a speed of 60 rpm. The mixing process comprised mixing the dry constituents followed by the addition of the required water, admixtures and nano-silica while constantly mixing until homogeneity of the mixture was achieved. Subsequently, fibers were added and the ingredients were mixed for about 10 min to achieve uniform distribution of fibers.
Mixture Proportions per Cubic Meter
Note: BFP = basalt fiber pellets; PVA = polyvinyl alcohol. Adjusted amount of mixing water considering the water content of nano-silica (aqueous solution with 50% solid content of SiO2 [provided by the manufacturer]).
Tests
The hardened properties of the composites were assessed by determining their compressive strength and flexural behavior. To determine the compressive strength of the mixtures, triplicate cylinders (100 × 200 mm) were tested at 3 and 28 days according to ASTM C39 ( 22 ). The flexural strength and post-cracking behavior of the mixtures at 28 days were determined according to ASTM C1609 ( 23 ), by testing triplicate prisms (100 × 100 × 350 mm) in four-point bending. A closed-loop, servo-controlled testing machine was used to apply the load, where the loading rate was controlled by the measured net mid-span deflection of the beam. Moreover, the residual strength was determined from the post-cracking curve at deflections of L/600 and L/150, and the flexural toughness was calculated as the total area under the load-deflection curve (P-δ). Scanning electron microscopy (SEM) assisted with JED-2300 energy-dispersive X-ray spectroscopy (EDX) was used to corroborate the results of flexural and compressive tests on fracture pieces comprising BFP and PVA after 28 days. Qualification of chemical elements was performed by EDX using ZAF method (Z is the atomic number correction relating to stopping power of the element, A is the absorption correction and F is the fluorescence correction) standard less analysis at voltage 15 kV, probe current 1 nA and acquisition time 90 s. Samples were coated with a fine layer of carbon before analysis to improve their conductivity for the SEM imaging.
Results and Discussion
Compressive Strength
Table 4 lists the average compressive strength for the composites at different ages (3 and 28 days). The results indicated that the compressive strength for all mixtures comprising 6% of nano-silica are greater than 50 MPa at early- and later-age, despite the high dosage of slag (50%) in the binder. For instance, the compressive strength for mixture G-6N-2.5B, comprising 6% nano-silica, was increased by 29% and 16%, respectively, compared with that of mixture G-2.5B (without nano-silica) at 3 and 28 days, as listed in Table 4. This could be attributed to the contribution of nano-silica to the improvement of the hardened properties of cement-based materials through multiple mechanisms. These include catalyzing the hydration process by creating nucleation sites for precipitation of hydration products ( 18 , 19 ), accelerating the pozzolanic reactivity ( 14 , 18 ), filling effect and water absorption within the high surface area of nano-silica agglomerates, resulting in reduction of w/b in the paste ( 18 ).
Compressive and Flexural Test Results of the Composites
In relation to the BFP, the compressive strength of composites decreased with increased BFP dosage. For example, at 28 days, the compressive strength of G-6N-4.5B was reduced by 11.2% relative to that of G-6N-2.5B. Branston et al. ( 24 ) reported a similar trend of reduction in compressive strength with increase of the dosage of basalt fiber mini-bars (0.3%, 1% and 2% by volume) in ordinary concrete (37 MPa). This could be ascribed to the creation of additional interfacial transitional zones (ITZs) with higher dosages of BFP, which acted as weak links and stress concentrators in the matrix, thus reducing its compressive capacity.
PVA fibers had a significant effect of increasing the compressive strength of mixture specimens. For instance, the compressive strengths of G-6N-1V increased by 20% and 30% relative to G-6N-2.5B and G-6N-4.5B, respectively. Li et al. ( 25 ) indicated that the typical dosage of PVA fibers to improve the compressive strength was 2% by volume. In this study, adding only 1% PVA micro-fibers with BFP macro-fibers (hybrid system) slightly improved the compressive strength of the composites, relative to those comprising BFP only. For example, the compressive strength of G-6N-4.5B-1V at 28 days was 6% higher than that of G-6N-4.5B (Table 4). Addition of PVA fibers changed the mode of failure of specimens from sudden crushing of the matrix to gradual evolution of macro-cracks but the composite remained intact, as shown in Figure 2. This can be attributed to the effect of PVA micro-fibers that impeded/delayed initiation of micro-cracks and micro-bridged them in the matrix. In addition, the hydrophilic nature of PVA created strong bonding with cement paste ( 25 ), resulting in marked confinement of the matrix.

Mode of failure of specimens: (a) G-6N-2.5B and (b) G-6N-2.5B-1V.
Flexural Strength
Table 4 lists the flexural and residual strengths as well as toughness for all composites at 28 days, based on the representative load-deflection (P-δ) curves shown in Figures 3–5. Conforming to the compressive strength trends, it was observed that the incorporation of nano-silica in concrete led to marked improvement in the first-peak flexural strength (Figure 3). The first-peak flexural strength for mixture G-6N-2.5B, comprising 6% nano-silica, was 38% higher than that of the reference mixture G-B2.5. This improved performance can be linked to the function of nano-silica agglomerates on densifying the cementitious matrix by multiple mechanisms, as explained above. In the same context, mixture G-6N-V1 had higher first-peak flexural load than that of the other two composites reinforced with a single type of fibers (G-6N-2.5B and G-6N-4.5B), as shown in Figure 3. The increase of the first-peak flexural strength for G-6N-V1 was 37% and 55% higher than that of G-6N-2.5B and G-6N-4.5B, respectively. This might be attributed to the adsorbed water on the PVA micro-fibers creating strong chemical bonds with cementitious paste and delaying the growth and propagation of micro-cracks ( 26 , 27 ), till reaching the first-peak load, representing the flexural capacity of the matrix. G-6N-2.5B achieved higher first-peak flexural load than that of G-6N-4.5B, because of the creation of additional ITZs with higher dosages of BFP. As shown in Figure 3, after first-cracking, mixture G-6N-1V, comprising 1% PVA micro-fibers, did not show any appreciable ductility and broke in a brittle manner. This can be attributed to the smaller dosage of PVA fibers compared with the typical dosage (2% by volume) used in previous studies; as the dosage of PVA fibers in cement paste increases, it may exhibit a ductile behavior with strain softening/hardening depending on the PVA dosage ( 27 ). However, the PVA dosage was not increased in this case to avoid the clustering of fibers that may occur when PVA is incorporated with BFP in the hybrid system. Comparatively, the other two mixtures comprising BFP continued to carry the load progressively up to the deflection limit of 2 mm. After the initiation of the first-crack, a sudden decline of the load was observed, likely because of redistribution of stresses to fibers in the failure plane showing a limited strain softening behavior. However, the BFPs were able to restrain the cracks and restore the load-carrying capacity of the section till reaching a second peak, depending on the dosage of BFP. The mixtures with the higher BFP dosages, 4.5%, exhibited a more efficient deflection-hardening behavior, relative to those reinforced with 2.5% BFP. This can be attributed to this novel type of macro-fibers, which consists of strands of basalt fibers encapsulated by polyamide resin. Each pellet has a specific surface texture with tailored micro-grooves in the longitudinal direction. These grooves improved the bond between BFPs and cementitious system and enhanced the pellets/matrix interface, especially in the presence of nano-silica particles, as shown by SEM. Furthermore, the reasonable modulus of elasticity and high tensile capacity of BFP efficiently restrained growth of macro-cracks and restored the load-carrying capacity of the composites till reaching a second peak load (Figure 3).

Load-deflection (P-δ) curves for composites with single fibers.

Load-deflection (P-δ) curves for composites reinforced with 2.5% BFP (single fibers) and 2.5% BFP with 1% PVA fibers (hybrid fibers).

Load-deflection (P-δ) curves for composites reinforced with 4.5% BFP (single fibers) and 4.5% BFP with 1% PVA fibers (hybrid fibers).
The first-peak loads for G-6N-2.5B-1V and G-6N-2.5B were comparable, as shown in Figure 4. However, adding 1% PVA with 4.5% (G-6N-4.5B-1V) improved the flexural strength by 46% compared with that of G-6N-4.5B specimens (Table 4), offsetting the negative effect imparted by the additional ITZs. In addition, adding PVA with BFP significantly enhanced the post-cracking behavior of composites, as depicted in Figures 4 and 5. After first-cracking, the composites containing hybrid fibers had little or no drop in the load-carrying capacity, and they continued to carry significant loads up to the designated deflection limit. For example, specimens from mixture G-6N-4.5B-1V showed a pseudo-hardening behavior (Figure 5), where the average residual strength (9 MPa) was 14% higher than the first-cracking flexural strength.
In addition, the post-cracking behavior of the mixtures can be assessed by the toughness (total area under the P-δ curve). Nano-silica has a pronounced effect on improving the toughness, owing to enhancement of the microstructural features of the matrix by modifying the ITZ with fibers and deposition of the pozzolanic products in the BFP longitudinal grooves, which improved the interface between matrix and BFP. The toughness of mixture G-6N-2.5B is 46%, higher than that of mixture G-2.5B (Table 4). While reinforcing the matrix with only 1% PVA was not effective at attaining high toughness (15.5. J), the use of BFP macro-fibers led to higher toughness of the composites. The toughness values of mixtures G-6N-2.5B and G-4.5B were higher by 50% and 62%, respectively, relative to that of mixture G-6N-1V. Mixtures G-6N-2.5B-1V and G-6N-4.5B-IV, comprising the hybrid fiber system, had higher toughness values (61% and 46%, respectively) compared with those of corresponding mixtures with BFP only (G-6N-2.5B and G-6N-4.5B). This improvement is because of the synergistic effects of micro- (PVA) and macro- (BFP) fibers, as the former impeded the nucleation and coalescence of micro-cracks and the latter controlled the propagation of macro-cracks. Homogenous distribution of hybrid PVA/BFP fibers in the matrix (Figure 6) provided effective restraint to propagation of cracks and better chance to arrest and bridge micro- and macro-cracks in the failure plane, resulting in enhanced toughening beyond first-cracking.

Distribution of hybrid fibers in the failure plane: (a) G-6N-2.5B-1V and (b) G-6N-4.5B-1V.
Microscopy Analyses
SEM assisted with EDX was used to corroborate the results of flexural and compressive tests after 28 days. For example, Figure 7a shows the dense micro-structure and ITZ between the BFP and matrix, owing to the synergistic effect of nano-silica with slag, as previously discussed. Similar trends were reported in Di Maida et al. ( 28 ) between nano-silica and polypropylene macro-fibers. According to the EDX analysis (Figure 7b), the average Ca/Si for the cement gel (1.18) had lower value compared with the conventional C-S-H generated from the hydration reaction of cement (1.7) ( 29 ). This indicated an efficient and accelerated pozzolanic reactivity within 28 days, which led to densification of the ITZ with secondary C-S-H, reflected by the better mechanical properties relative to the reference mixture without nano-silica. Figure 8a shows the BFP with a diameter of 1.8 mm, embedded in the nano-modified matrix comprising randomly distributed PVA fibers, which contributed to controlling the micro-cracks in the matrix. The pellets had longitudinal grooves, which enhanced the interface between the matrix and BFP, owing to the presence of high contact surface area for the deposition of hydration products, as depicted in Figure 8b. In addition, no degradation of BFP was observed in the micrographs, because of the protective effect of the polyamide resin against the surrounding alkaline medium.

Scanning electron microscopy for nano-modified slag composite showing: (a) interfacial transitional zones with basalt fiber pellets and (b) energy-dispersive X-ray spectrum of paste at the indicated locations (blue triangle points) (average Ca/Si = 1.18).

Scanning electron microscopy images for the interaction of basalt fiber pellets (BFP) with (a) polyvinyl alcohol fibers and (b) hydration products in the grooves.
Conclusion
Based on the materials, mixtures and tests implemented in this study, the following conclusions can be drawn:
Relative to the composite without nano-silica, addition of 6% nano-silica to the binder was effective at improving the compressive strength, flexural strength and toughness of the composites because of densification of the cementitious matrix and improved bonding with fibers.
The hydrophilic nature of PVA micro-fibers and their ability to delay the growth and propagation of micro-cracks led to increase in the compressive and flexural capacities of the matrix; however, after first-cracking, brittle failure of these specimens occurred because of the 1% dosage of PVA fibers used, which was insufficient to impede the propagation of macro-cracks.
Additional ITZs were created in the matrix by increasing the dosage of BFP, leading to reduction in compressive and flexural strengths. However, the BFP contributed to enhancing the flexural performance of the developed composites in relation to post-cracking behavior. In particular, composites comprising 4.5% BFP showed deflection-hardening behavior with significantly improved toughness and residual strength, owing to the reasonable modulus and high tensile capacity of BFP, and the enhanced bonding between these grooved pellets and nano-modified binder, as shown by SEM images.
Adding 1% PVA fibers to 2.5% and 4.5% BFP imparted synergistic effects on the composites, as the former impeded the nucleation and coalescence of micro-cracks and the latter controlled the propagation of macro-cracks, resulting in improved compressive capacity and flexural performance of the developed composites. In particular, mixture G-6N-4.5B-1V remained intact after reaching its ultimate compressive capacity and showed pseudo-hardening behavior with significantly improved residual flexural strength and toughness. Therefore, nano-modified cementitious composites incorporating hybrid BFP/PVA fibers can be used in a suite of applications requiring both high strength and ductility.
Footnotes
Acknowledgements
The first author highly appreciates support from the Price Graduate Scholarship for Women in Engineering. The second author highly appreciates funding from the Natural Sciences and Engineering Research Council of Canada and the University of Manitoba Research Grants Program. The IKO Construction Materials Testing Facility at the University of Manitoba in which these experiments were conducted has been instrumental to this research.
Author Contributions
The authors confirm contribution to the paper as follow: study conception and design: R. Elhadary and M. T. Bassuoni; data collection: R. Elhadary and M. T. Bassuoni; analysis and interpretation: R. Elhadary and M. T. Bassuoni. Both authors reviewed the results and approved the final version of the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received the following financial support for the research, authorship, and/or publication of this article. The first author received funding from the Price Graduate Scholarship for Women in Engineering. The second author received funding from the Natural Sciences and Engineering Research Council of Canada (NSERC RGPIN/4024-2014), and University of Manitoba Research Grants Program (URGP-2019/2020).
