Tiny Fibers, Big Defense: How PP and PVA Fibers Shield Concrete From Freeze–Thaw Destruction

When water seeps into concrete and freezes, it expands with relentless force, prying open microscopic pores and stitching together networks of cracks that can quietly dismantle bridges, dams, and pavements from the inside out. In cold and seasonally frozen regions, this freeze–thaw assault is one of the most damaging durability problems in civil engineering. A new study published in Case Studies in Construction Materials offers one of the most complete cross-scale pictures yet of how two common polymer fibers—polypropylene (PP) and polyvinyl alcohol (PVA)—slow this destruction, tracking the damage from surface scaling all the way down to the fiber–matrix interface and into a three-dimensional computer model of a cracking cube of mortar.

The research team, led by Hui Huang and Cong Zhang, prepared three mixtures under an identical binder–aggregate–water framework: a plain mortar control, a PP-fiber-reinforced composite, and a PVA-fiber-reinforced composite. Both fiber types were 6 millimeters long and added at a volume fraction of 1.0 percent, so that fiber dosage would not confound the comparison. The fibers themselves, however, are chemically and mechanically very different. The PP fiber, with a diameter of 20 micrometers, has a tensile strength of 800 megapascals and an elastic modulus of 5 gigapascals, and its surface is hydrophobic. The finer PVA fiber, at 14 micrometers, is roughly twice as strong at 1,800 megapascals and nearly three times stiffer at 14 gigapascals, and its hydrophilic chemistry bonds more readily with cement hydration products. Those differences turned out to matter enormously.

The specimens, 70.7-millimeter cubes, were cured for 28 days and then subjected to a rapid water-freezing and water-thawing regime in which temperatures swung between −18 °C and 5 °C every four hours, with the samples fully immersed. Measurements were taken after 0, 25, 50, 75, 100, 125, and 150 cycles. Before exposure, the plain mortar was actually the strongest material, at 65.03 megapascals of compressive strength, compared with 61.92 megapascals for the PVA composite and 54.61 megapascals for the PP composite. But strength at the start is not the same as strength that lasts.

After 150 freeze–thaw cycles, the ranking had reversed in the way that matters for durability. The fiber-reinforced composites retained 82.29 percent (PP) and 84.50 percent (PVA) of their initial compressive strength, while the unreinforced mortar retained only 66.45 percent. Mass loss told a parallel story: the mortar reached 5 percent mass loss by 125 cycles through continuous surface scaling, whereas both fiber mixtures degraded more slowly in a three-stage pattern—slow loss in the first 50 cycles, acceleration between 50 and 100 cycles as cracks connected, and a slowdown thereafter. Even after 150 cycles, the fibers still reduced surface material loss by more than 59 percent relative to plain mortar, with PVA consistently outperforming PP.

Perhaps the most striking finding was that stiffness, not strength, is the more sensitive fingerprint of freeze–thaw damage. The apparent elastic modulus, derived from the slope of the compressive stress–strain curve between 10 and 30 percent of peak stress, fell by 32.9 percent for PP-FRCC and 29.2 percent for PVA-FRCC over 150 cycles—nearly double the corresponding strength losses. Because the modulus reflects the load-transfer continuity of the internal microstructure, it registers the accumulation of pores, microcracks, and debonded interfaces long before catastrophic strength loss appears. For engineers assessing aging infrastructure, this suggests that stiffness-based or non-destructive measurements could serve as an early-warning system for frost damage.

Scanning electron microscopy revealed why the two fibers behaved differently. Using calibrated image analysis with two semi-quantitative descriptors—a visible defect area ratio and an interfacial deterioration ratio—the researchers quantified what they saw in the micrographs. The visible defect area ratio climbed from 1.8 to 8.2 percent in mortar, from lower starting values to 6.8 percent in PP-FRCC and 5.6 percent in PVA-FRCC. More telling was the interface: the fraction of visibly deteriorated fiber–matrix boundary rose from 3.0 to 21.2 percent for PP-FRCC but only from 2.7 to 17.4 percent for PVA-FRCC. The hydrophobic PP surface, chemically stable and inexpensive, gradually separated from the matrix under repeated hydraulic and crystallization pressures, while the hydrophilic PVA fiber maintained more continuous contact with surrounding hydration products.

Those microscopic differences echoed upward into failure behavior. Under splitting tensile loading, PVA-FRCC specimens displayed primary or deflected splitting cracks in 17 of 21 observations (81.0 percent), whereas PP-FRCC showed localized or branched damage features in 15 of 21 (71.4 percent). The tensile-to-compressive strength ratio, a marker of balanced degradation, stayed remarkably stable for PVA-FRCC—fluctuating within a coefficient of variation of just 0.83 percent—while mortar’s ratio swung the most. In compression, fiber-bridged cracking kept the reinforced cubes intact even after 150 cycles, while the plain mortar failed brittly with severe spalling and a conical crushing zone at its core.

To connect these observations mechanically, the team built a three-dimensional mesoscale model in ABAQUS, representing the composite as a cementitious matrix, randomly distributed equivalent fibers, and an equivalent pore phase whose volumetric expansion stands in for freeze–thaw damage. Calibrated against the measured compressive response at each cycle stage, the model reproduced the monotonic strength decline well at early and intermediate stages, with discrepancies under about 4 percent through 75 cycles. By 150 cycles the gap widened to roughly 17 percent for both composites—a limitation the authors attribute to the simplified pore-expansion representation and the absence of explicit fiber–matrix bond-slip and pull-out. Still, the simulated damage fields were revealing: equivalent strain, stiffness degradation, and compressive damage, initially scattered around isolated weak regions, progressively coalesced into connected damage paths as freeze–thaw pre-damage accumulated, mirroring the experimentally observed shift from distributed fine cracking to localized through-cracks.

Two-way analysis of variance confirmed that material type, cycle number, and their interaction significantly shaped mass loss, compressive strength retention, and stiffness retention, with strikingly large effect sizes—partial eta squared above 0.9 for the cycle effect on mass loss. For splitting tensile strength retention, the interaction was not significant, indicating that tensile degradation followed overall material and cycle effects rather than a material-specific trajectory. Together, the statistics reinforce the central message: fiber reinforcement does not merely slow damage, it changes the pathway by which deterioration proceeds.

The practical implications reach cold-region roads, bridges, repair overlays, and any concrete that may sit water-saturated through winter. Adding just 1 percent by volume of either fiber substantially curbed surface scaling and strength loss, with PVA offering slightly better residual strength, stiffness retention, and interfacial stability under this particular mixture design, while PP delivered comparable protection at lower cost. The authors are careful to note that the comparison is bounded by their specific mix proportions, cube geometry, and exposure regime, and that dosage optimization, salt frost exposure, flexural testing, and quantitative porosity measurements remain for future work. Even so, the study delivers a rare, unified account of frost damage—from the swelling pore to the debonding fiber to the cracking cube—and a compelling case that the humble fiber–matrix interface may be the most important square micrometers in frozen concrete.

Subject of Research: Freeze–thaw damage evolution and durability of polypropylene- and polyvinyl alcohol-fiber-reinforced cementitious composites

Article Title: Freeze–thaw damage evolution of PP- and PVA-fiber-reinforced cementitious composites: Mechanical degradation, interfacial deterioration and mesoscale modeling

Article References: Huang, H., Zhang, C., Liu, X., Xu, X., Jia, B., Gu, S., & Wang, W. (2026). Freeze–thaw damage evolution of PP- and PVA-fiber-reinforced cementitious composites: Mechanical degradation, interfacial deterioration and mesoscale modeling. Case Studies in Construction Materials, 25, Article e06594. https://doi.org/10.1016/j.cscm.2026.e06594

Image Credits: AI Generated

DOI: 10.1016/j.cscm.2026.e06594

Keywords: freeze–thaw cycles, fiber-reinforced cementitious composites, polypropylene fibers, PVA fibers, concrete durability, interfacial deterioration, mesoscale modeling, elastic modulus degradation, scanning electron microscopy, compressive strength retention, cold-region infrastructure, damage localization