The team set out to answer a deceptively simple question: when a plastic item begins to break down in the environment, does its biodegradable pedigree actually reduce the risk that it will shed microscopic fragments? The answer matters because biodegradation is a process that requires specific conditions, particular microbes, adequate temperature, sufficient moisture and time. In the open environment, whether at sea, on a beach or in a soil, those conditions are rarely met. Before a bioplastic ever fully mineralises, it may spend weeks, months or years in a limbo state, physically fragmenting into smaller and smaller pieces even as its chemistry slowly degrades.

To probe that limbo, the researchers compared five polymers: polylactic acid (PLA), polyhydroxybutyrate (PHB) and thermoplastic starch (TPS), all biodegradable bioplastics, alongside low-density polyethylene (LDPE) and polypropylene (PP), two of the most widely used conventional plastics. They exposed the materials to two distinct degradation pathways designed to mimic what plastics encounter in nature. The first was accelerated ultraviolet ageing in saline water, simulating the combined assault of sunlight and seawater that floating or beached debris endures. The second was mechanical abrasion with sand, reproducing the grinding action of waves tumbling debris across sediment in the swash zone of a shoreline.

Under the UV ageing regime, the bioplastics clearly suffered the most. PLA and PHB showed the strongest degradation of the four polymers tested in that pathway, exhibiting greater mass loss, more pronounced surface damage, larger colour changes and more severe deterioration of their mechanical properties than either LDPE or PP. This is chemically unsurprising: the ester bonds in PLA and PHB are susceptible to hydrolysis, and UV radiation accelerates oxidation reactions that scission polymer chains, embrittling the material. Polyethylene and polypropylene, with their backbone of stable carbon-carbon bonds, weather more slowly, which is precisely why they persist so notoriously in the ocean.

But faster degradation does not automatically mean a cleaner outcome. The study’s central and most provocative finding is that the biodegradable plastics evaluated did not necessarily present a lower risk of microplastic release than the conventional plastics tested under identical conditions. In fact, when degradation conditions and exposure time were matched, some of the biodegradable polymers generated even larger amounts of particulate material, indicating a higher potential for microplastic generation. The very reactivity that makes a bioplastic susceptible to biological breakdown also makes it fragile under physical weathering, and a brittle, cracking sheet of plastic sheds fragments far more readily than a tough, flexible one.

The abrasion experiments reinforced that picture from a different angle. When the samples were ground against sand, PHB and thermoplastic starch showed the highest mass loss, while LDPE and PP proved markedly more resistant. Mechanical fragmentation is now recognised as a major route by which plastics break apart in coastal environments, and the Alicante results suggest that softer, more hydrophilic biopolymers may be particularly vulnerable to it. TPS, derived from starch, and PHB, produced by bacterial fermentation, simply lack the toughness that decades of polymer engineering have built into commodity polyolefins.

Across both experimental pathways, one trend held without exception: microplastic generation increased with degradation time for every material examined. No polymer, biodegradable or conventional, was immune to fragmentation. What varied was the rate and the quantity, and those differences depended strongly on both the polymer type and the degradation route. A material that behaves well under sunlight may fare poorly under abrasion, and vice versa, which complicates any simple ranking of plastics by environmental friendliness.

The technical implications reach well beyond the laboratory. Microplastics, particles smaller than five millimetres, are now found from the deep sea to Arctic fjords, in agricultural soils, in garden compost and in human tissues. A growing body of literature documents that biodegradable microplastics are themselves contaminants: they can adsorb and transport antibiotics and other pollutants, affect soil ecosystems and plant growth, and persist longer than their labels suggest. Recent studies have even shown that municipal biowaste treatment plants release residues of biodegradable plastics with a putative persistence potential that may exceed expectations. Meanwhile, the bioplastics market continues to expand rapidly, driven by packaging demand and regulatory pressure, meaning that ever larger volumes of these materials are entering waste streams and open environments.

The Alicante team assessed degradation through a battery of complementary measurements, tracking mass loss, surface morphology, colour variation and mechanical properties, and quantifying the microplastic particles released. This multi-parameter approach matters because no single metric captures the full picture of how a polymer weathers. A sample might lose little mass while its surface crumbles into particles, or change colour dramatically while retaining its structural integrity. By combining these indicators with direct microplastic counts, the researchers could connect the chemistry of degradation to the physical reality of fragmentation, offering one of the more complete head-to-head comparisons of bioplastic and conventional plastic weathering published to date.

What should consumers and policymakers take away from these findings? The authors’ results do not suggest that bioplastics are worthless, and the study did not evaluate full biodegradation under controlled composting conditions, where materials like PLA and PHB can genuinely break down. Rather, the work highlights a critical gap between certification and reality: a product labelled compostable will only compost in a facility that provides the right conditions, and if it instead ends up in the ocean, on a roadside or in a landfill, it may fragment into microplastics just as readily as the conventional plastic it replaced, sometimes more so. As regulators weigh bans on single-use conventional plastics and incentives for bio-based alternatives, the message from this research is that end-of-life performance in the real world, not just the origin of the feedstock, must be the yardstick. Until biodegradable plastics reliably biodegrade wherever they end up, the microplastic problem may simply be changing its chemistry rather than shrinking.

Subject of Research: Microplastic release from biodegradable and conventional plastics during UV ageing and mechanical abrasion

Article Title: Microplastic formation during degradation: bioplastics vs conventional plastics

Article References: Oliver, I., Fullana, A., & Conesa, J. A. (2026). Microplastic formation during degradation: bioplastics vs conventional plastics. Environmental Science and Pollution Research, 33(30), 15618-15637. https://doi.org/10.1007/s11356-026-38204-x

Image Credits: AI Generated

DOI: 10.1007/s11356-026-38204-x

Keywords: bioplastics, microplastics, PLA, PHB, thermoplastic starch, polyethylene, polypropylene, UV degradation, mechanical abrasion, marine environment, polymer fragmentation, biodegradation

The degradation of reinforced concrete in marine settings follows a well-characterized two-phase sequence. During the initiation phase, chlorides diffuse through the concrete cover toward the embedded steel; the propagation phase begins once the chloride concentration at the rebar depth exceeds a critical threshold, triggering corrosion that produces rust products, cracking, and eventually spalling of the cover. Without intervention, this process can culminate in severe structural damage or outright failure. Traditional inspection techniques, including visual surveys and destructive or non-destructive testing, can detect corrosion, but they are costly, time-consuming, and in some cases risk misdiagnosis or further damage to the structure itself.

Modeling approaches have evolved considerably over the decades. Early studies relied on deterministic formulations of chloride transport, often using analytical solutions to Fick’s second law of diffusion. Later work introduced probabilistic methods to account for the substantial uncertainties in material properties, environmental exposure, model error, and construction quality. However, most of these probabilistic studies relied on one-dimensional chloride ingress models combined with Monte Carlo simulation or reliability methods such as FORM and SORM. That one-dimensional assumption is a serious limitation for real structural elements like columns and beams, where chlorides attack from multiple directions simultaneously. At the corners of polygonal members, chloride penetrates from two orthogonal faces, concentrating the attack precisely where the reinforcement is most vulnerable. One-dimensional models therefore tend to overestimate the time to corrosion initiation at these critical corner regions, potentially leaving structures at risk.

Two-dimensional probabilistic models have addressed the corner problem, but they have largely focused on new, single-layer concrete and ignored what happens after a repair. Repair introduces a genuinely difficult physics problem: when chloride-contaminated concrete is removed and replaced, the residual chlorides left in the original substrate do not simply stay put. They redistribute, diffusing from the old, contaminated concrete into the fresh, chloride-free repair material through the interface between the two layers. This interaction between old and new concrete, including the formation of so-called incipient anodes at the interface, has been captured by detailed deterministic finite element and finite difference models, but extending these physically rich two-dimensional repair simulations to a probabilistic setting has been computationally prohibitive. A single deterministic simulation of a 75-year maintenance history takes roughly 30 minutes on a workstation, and probabilistic assessment typically demands thousands to millions of such runs.

The research team’s solution is a surrogate modeling strategy based on polynomial chaos expansion, or PCE. Instead of running the full numerical model for every uncertainty scenario, PCE constructs a mathematical approximation of the model’s response as a series of multivariate orthogonal polynomials in the random input variables. The coefficients of this expansion, computed by least-squares minimization over a limited set of full model evaluations, map how each input and each interaction between inputs contributes to the output variability. Using least angle regression, the method retains only the polynomial terms with the largest impact, producing a sparse surrogate that captures nonlinear relationships with remarkable economy. In the case study, a surrogate built from just 200 full model evaluations achieved a leave-one-out cross-validation error of 5 times 10 to the power of minus 4, corresponding to a target accuracy of 0.999, with a polynomial order of six and 96 basis elements. Monte Carlo simulation on the surrogate then yields the probability of corrosion initiation at a tiny fraction of the original computational cost.

The underlying physics model, implemented in the open-source software FreeFem++, solves the two-dimensional chloride diffusion equation with a spatially and temporally varying apparent diffusion coefficient. This coefficient accounts for chloride binding, modeled with a Langmuir isotherm, as well as the effects of temperature, concrete aging, and humidity on transport. The repair model divides the column cross-section into two domains: the newly placed, chloride-free repair material and the original aged concrete retaining its residual chlorides. The repair geometry itself is thoughtfully designed, with a circular boundary drawn from the column center to the side reinforcement, so that concrete removal extends deepest near the corner rebars, where contamination is worst, while sparing uncontaminated interior concrete. This circular replacement strategy is both cost-effective and sustainable, minimizing the volume of material that must be removed and replaced.

The illustrative case study examines a 0.4 by 0.4 meter square column in an extreme coastal splash and tidal environment, with a 65-millimeter design cover depth specified according to Eurocode 2. Nine input parameters were treated as random variables, including surface chloride concentration, the critical chloride threshold for corrosion initiation, reference and repair diffusion coefficients, activation energy, the aging factor, and cover thicknesses, each assigned distributions drawn from prior experimental and probabilistic literature. Two repair materials were compared: S1, a conventional concrete similar to the original, and S2, a wet-mix shotcrete with a denser microstructure and lower chloride diffusivity. Repairs were assumed to be triggered when the probability of corrosion initiation reached 10 percent, a threshold consistent with the fib Model Code’s deterioration limit state and close to the Eurocode serviceability reliability target.

The results carry practical weight. With repairs every 10 years using either material, the probability of corrosion initiation stayed around 6.6 percent, safely below the 10 percent threshold, confirming the effectiveness of the circular replacement strategy even under uncertainty. But the higher-quality S2 material showed a clear advantage: after the first repair cycle, the corrosion initiation probability dropped from 6.6 percent to 3.6 percent, reflecting the dense shotcrete’s superior resistance to chloride ingress. When the framework was used to optimize the maintenance schedule rather than simply test a fixed one, the differences became economic. For the conventional S1 material, the optimal plan called for six repairs over a 75-year service life, at years 12, 24, 36, 48, 60, and 72. The superior S2 material stretched the intervals to 15 years, requiring only five repairs at years 12, 27, 42, 57, and 72. Fewer repair campaigns mean lower direct costs and a smaller environmental footprint from materials, demolition waste, and construction traffic.

Equally valuable is the framework’s global sensitivity analysis, made possible by post-processing the polynomial chaos coefficients into Sobol indices. Before repair, concrete cover thickness dominated, with a first-order Sobol index exceeding 60 percent of the output variance, a finding amplified by the two-dimensional corner geometry where the cover is the only barrier against attack from two directions. Surface chloride concentration ranked second, followed by the reference diffusion coefficient and the aging factor, while the activation energy for diffusion proved negligible under the study’s relatively stable temperature conditions. After repair, the hierarchy shifted: the diffusion coefficient of the repair material became the second most influential parameter, underscoring that material selection for repairs is a durability decision, not merely a structural one. The amount of concrete replaced, which determines the renewed cover thickness, took over the protective role previously played by the original cover. Notably, differences between first-order and total Sobol indices remained below 10 percent, indicating that the input parameters act largely independently rather than through strong interactions.

The authors are careful to frame the resulting 12-to-15-year repair intervals as illustrative outputs of an idealized preventive strategy rather than universal recommendations, and they emphasize that field observations in aggressive marine environments have documented corrosion-induced deterioration after only 20 to 25 years of exposure, sometimes earlier. The framework’s real contribution is methodological: a computationally tractable way to make proactive, reliability-based maintenance decisions for structures whose deterioration is inherently two-dimensional and whose repairs create new, coupled transport domains. The researchers identify clear next steps, including integrating empirical data on parameter correlations, characterizing geometric uncertainties after repair, and coupling the probabilistic model with life-cycle cost analysis and inspection data. As coastal infrastructure ages and climate change intensifies marine exposure, tools that convert uncertainty from an obstacle into a design variable may prove essential for keeping the world’s concrete assets safe, serviceable, and affordable to maintain.

Subject of Research: Probabilistic optimization of maintenance for chloride-corroded reinforced concrete structures in marine environments

Article Title: A probabilistic framework for optimizing maintenance of reinforced concrete structures in marine environments

Article References: Truong, Q.-C., El Soueidy, C.-P., & Bastidas-Arteaga, E. (2026). A probabilistic framework for optimizing maintenance of reinforced concrete structures in marine environments. Results in Engineering, 32, Article 113164. https://doi.org/10.1016/j.rineng.2026.113164

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113164

Keywords: reinforced concrete, chloride-induced corrosion, marine environment, polynomial chaos expansion, probabilistic modeling, maintenance optimization, surrogate modeling, sensitivity analysis, concrete repair, structural reliability, chloride ingress, service life prediction