The research team assembled a database of 396 tensile testing records for sub-sized Stainless Steel 316 specimens, drawn from an extensive review of peer-reviewed literature and scientific databases. Each record was described by 45 input features spanning material composition, manufacturing and heat treatment history, irradiation conditions, specimen geometry, and testing parameters such as temperature and strain rate. The team set out to predict four key tensile properties: ultimate tensile strength, yield strength, total elongation, and uniform elongation. After cleaning to remove records with missing values, the usable datasets shrank further—to just 209 points for ultimate tensile strength, 204 for yield strength, 190 for total elongation, and a mere 86 for uniform elongation. For a field where safety margins depend on reliable property predictions, those numbers are dauntingly small.

The imbalance in the data was as troubling as its sparsity. Yield strength values, for example, showed a distinctly bimodal distribution with a conspicuous gap between roughly 450 and 500 megapascals. The low-strength cluster corresponded to solution-annealed specimens treated above 1000 degrees Celsius, while the high-strength cluster came from cold-worked and stress-relieved material. This kind of clustering means a model can be trained heavily on one regime and fail badly in another, particularly in the underrepresented regions of the feature space that matter most for understanding how specimen size affects measured properties. Size effects themselves complicate matters: when specimen dimensions approach microstructural length scales such as grain size or precipitate spacing, deformation mechanisms shift from bulk-dominated to surface-mediated behavior, and conventional models of deformation and fracture can lose validity.

To attack the data problem, the researchers turned to two very different augmentation strategies. The first was a Generative Adversarial Network, or GAN, in which a generator network produces synthetic feature vectors from random noise while a discriminator network learns to distinguish real from fake samples, forcing the generator to create increasingly realistic data. The team trained the GAN for 10,000 epochs using the Adam optimizer, with fully connected layers, batch normalization, and LeakyReLU activations. The second method, SMOGN—the Synthetic Minority Oversampling Technique for Regression with Gaussian Noise—works more directly: it interpolates between existing samples and their nearest neighbors, then perturbs the results with Gaussian noise, targeting the sparsest regions of the target variable’s distribution. Synthetic compositions were constrained to physically realistic ranges, with elemental fractions normalized to sum to 100 percent. Both methods were applied at augmentation ratios of 20, 40, 60, 80, and 100 percent additional synthetic data.

Nine machine learning algorithms were then evaluated in a remarkably thorough experimental design: 396 individual model configurations, each assessed with five-fold cross-validation using the Pearson correlation coefficient, the coefficient of determination, and root-mean-square error. The algorithms spanned the modern machine learning spectrum, from Random Forest and Extreme Gradient Boosting (XGBoost) to decision trees, k-nearest neighbors, support vector machines, Gaussian process regression, neural networks, and two transformer architectures—one treating each feature as a token, and a Materials-Informed Transformer that grouped features by physics-based categories such as composition, processing, and irradiation conditions.

The results delivered a clear verdict: tree-based ensemble methods dominated. Random Forest achieved a Pearson correlation of 0.987 for yield strength with a root-mean-square error of 32.6 megapascals, while XGBoost reached 0.989 for uniform elongation with an error of just 3.414 percent—extraordinary accuracy given that the uniform elongation dataset contained only 86 points. The researchers took care to verify that this performance reflected genuine learning rather than statistical luck: a naive baseline that simply predicts the training-set mean produced a root-mean-square error of 21.714 percent and a negative coefficient of determination, and permutation importance analysis showed that the most influential features were heat treatment condition and test temperature, both of which have well-established physical relevance to ductility. Support vector machines fared worst, with yield strength correlation dropping to 0.892 and errors 27 percent higher than Random Forest, underscoring how poorly linear decision boundaries capture the coupled, non-linear relationships in materials data.

Perhaps the most intriguing finding was that data augmentation is strongly property-dependent. For strength-related properties, which are governed by discrete and competing mechanisms such as solid-solution strengthening, precipitation, and cold work, GAN-generated samples consistently helped. XGBoost improved by about 1 percent for ultimate tensile strength at 60 percent GAN augmentation, and k-nearest neighbors—normally expected to be hypersensitive to synthetic data—gained 1.4 percent at full GAN augmentation, apparently because the generator produced chemically realistic intermediate compositions that improved local neighborhood quality. For elongation properties, which follow smoother, more continuous geometric scaling trends, SMOGN’s interpolative approach worked better. XGBoost peaked at 80 percent SMOGN augmentation for total elongation, and a decision tree jumped 1.7 percent for uniform elongation at 60 percent SMOGN. Meanwhile, some combinations backfired badly: SMOGN degraded Random Forest’s ultimate tensile strength predictions by over 8 percent, and support vector machines deteriorated under both augmentation types across the board.

The study also delivered a cautionary tale for the physics-informed machine learning community. The team built Physics-Informed Neural Networks embedding two classical empirical relationships—Barba’s Law and the Bertella-Oliver formula—as constraints on total elongation prediction. Intuitively, adding physics should help a data-starved model. Instead, both PINN variants underperformed a standard neural network by roughly 10 percent in correlation, with root-mean-square errors rising from 9.87 percent to 15.4 and 16.0 percent. The authors attribute this to a mismatch between the rigid, globally applied empirical laws and a heterogeneous dataset spanning multiple alloys and processing conditions. When the imposed physics is wrong for parts of the data, it actively hurts. The lesson is that physics constraints must be flexible and data-adaptive—learning material-dependent parameters or using soft constraints—rather than hard-coded relationships.

Transformer architectures told a more nuanced story. The Simple Tabular Transformer ranked third for yield strength at 0.979 correlation but fell to eighth place for ultimate tensile strength, a 14 percent performance gap revealing its sensitivity to underlying data structure. The Materials-Informed Transformer, with its physics-based feature grouping, consistently underperformed its simpler cousin by 2 to 15 percent, and ablation experiments ruled out grouping granularity as the culprit—suggesting the prescribed attention structure conflicted with data-driven patterns. Interestingly, the MIT showed the largest relative gains from augmentation, improving 4.4 percent for total elongation under SMOGN, hinting that physics-based architectures may simply need more training diversity to learn useful attention patterns.

The practical implications extend well beyond Stainless Steel 316. Nuclear material qualification traditionally demands large specimens and extensive irradiation campaigns that can stretch over years, and miniaturized testing is one of the most promising routes to accelerating the screening of candidate alloys for advanced reactors. By demonstrating that carefully chosen algorithm-augmentation pairings can extract reliable predictions from a few hundred records, the study offers a template for data-starved materials science more broadly: match the augmentation strategy to the physics of the property being predicted, favor ensemble methods for stability, and treat rigid physics constraints with skepticism when the data spans diverse regimes. The team’s framework, validated through cross-validation stability analysis showing XGBoost’s standard deviation as low as 0.003 for uniform elongation, points toward faster, cheaper, and safer qualification of the materials that next-generation nuclear reactors will depend on.

Subject of Research: Machine learning and data augmentation for predicting tensile properties of sub-sized nuclear structural materials

Article Title: Predicting tensile properties of sub-sized nuclear materials using data augmentation strategies on imbalanced and sparse datasets

Article References: Smagulova, A., Vakanski, A., Merickel, J. W., Lee, I., Purushotham, C., Tang, Y., Cerpa, A. E., Xu, F., & Song, R. (2026). Predicting tensile properties of sub-sized nuclear materials using data augmentation strategies on imbalanced and sparse datasets. Frontiers in Manufacturing Technology, 6, Article 1811951. https://doi.org/10.3389/fmtec.2026.1811951

Image Credits: AI Generated

DOI: 10.3389/fmtec.2026.1811951

Keywords: machine learning, data augmentation, nuclear materials, tensile properties, Stainless Steel 316, GAN, SMOGN, Random Forest, XGBoost, sub-sized specimens, physics-informed neural networks, materials informatics

The core problem the researchers set out to solve is one of data scarcity. Machine learning thrives on large, consistent datasets, but additively manufactured austenitic stainless steels, the family that includes workhorses such as 304L and 316L, have only been studied intensively for roughly a decade. Every published experiment reports different process parameters, different powder chemistries, different build orientations and different testing conditions, which makes the mapping from composition to mechanical properties noisy and fragmented. Casting, by contrast, has more than a century of carefully documented composition-property relationships. The insight behind the new work is that although casting and additive manufacturing produce very different microstructures, the fundamental thermodynamics that govern whether an alloy solidifies as stable austenite, forms ferrite, or sits near a phase boundary are shared across processes. If the model could learn those underlying phase-stability relationships from casting data, the relative ordering of compositions might transfer to printing even when the absolute strength numbers do not.

To make that transfer possible, the team needed descriptors that capture metallurgical meaning rather than raw elemental percentages. They turned to two of the oldest and most trusted quantities in stainless steel metallurgy: the chromium equivalent and the nickel equivalent. These equivalents, familiar to generations of engineers through the Schaeffler diagram used to predict weld microstructures, condense the effects of ferrite-stabilizing elements such as chromium, molybdenum and silicon, and austenite-stabilizing elements such as nickel, manganese, carbon and nitrogen, into two intuitive axes. By expressing each alloy’s composition in terms of these empirical descriptors, the researchers gave the machine learning algorithm a physically informed coordinate system in which alloys with similar phase behavior cluster together, regardless of whether their properties were measured in a foundry or beside a laser powder bed.

On top of this descriptor choice, the authors introduced a divide-and-conquer strategy for model construction. Rather than forcing a single algorithm to describe the entire composition space, including the treacherous regions near phase boundaries where small chemical changes cause dramatic microstructural shifts, the divide-and-conquer model partitions the data into sub-regions defined by phase stability and trains separate predictors within each partition. The study evaluated this framework against three targets: ultimate tensile strength, elongation, and a combined tensile property defined as the product of strength and elongation, a single figure of merit that rewards alloys that are both strong and ductile rather than merely one or the other. The results were impressive. The divide-and-conquer model achieved coefficients of determination of 0.944 for ultimate tensile strength, 0.812 for elongation, and 0.899 for the combined property, and its mean absolute percentage errors were lower than those of ten commonly used machine learning models tested for comparison.

But a model that only predicts well on the data it was trained on is of limited use for designing new alloys. The real test came when the researchers coupled their model to a genetic algorithm, an optimization technique inspired by biological evolution that iteratively breeds candidate compositions toward better predicted performance. This inverse design loop allows the researchers to start from a desired property profile and work backwards to a chemical recipe, rather than laboriously testing one composition after another. The genetic algorithm searches the composition space through the lens of the divide-and-conquer model, proposing candidate austenitic stainless steels whose predicted strength-ductility combinations exceed those of conventional grades. The team then took the most promising candidates out of the computer and into the laboratory, fabricating them by additive manufacturing and measuring their actual tensile behavior.

Here the study delivers its most important and nuanced finding. When the printed alloys were tested, their absolute strength and ductility values did not match the model’s numerical predictions exactly, a systematic deviation the authors attribute to the profound process differences between slow, equilibrium-like casting solidification and the rapid, cyclic thermal history of layer-by-layer printing. Casting produces coarse, slowly formed grains, while additive manufacturing generates fine, rapidly solidified structures with unique sub-grain features, dislocation densities and segregation patterns that can dramatically alter mechanical response. Yet when the researchers compared the relative rankings of different compositions, the agreement was essentially perfect. The order in which the model predicted the compositions would perform, from best to worst, matched the order observed in the printed specimens. For the predicted improvement in ultimate tensile strength between two specific compositions, the relative error was as low as 0.81 percent, a level of agreement that would have been unthinkable without the phase-stability-based partitioning scheme.

This distinction between absolute and relative prediction is more than a statistical curiosity; it is the conceptual heart of the paper. In alloy development, engineers often care less about predicting an exact tensile strength to the nearest megapascal than about knowing which of a hundred candidate chemistries deserves the expense of a printing trial. The new work demonstrates that the relative composition-property relationships encoded in chromium and nickel equivalents possess genuine cross-process transferability. The physics of phase stability, it turns out, is a kind of universal language spoken by both the foundry crucible and the laser powder bed, and a model that learns that language from abundant casting data can say meaningful things about alloys it has never seen printed. This effectively converts a century of casting knowledge into a screening tool for a technology that is barely out of its adolescence.

The practical implications are considerable. Austenitic stainless steels are among the most widely printed structural metals, appearing in aerospace brackets, marine components, chemical processing equipment, nuclear applications and biomedical implants. Each new application pushes toward compositions tuned for specific balances of strength, ductility, corrosion resistance, printability and cost, and researchers have explored everything from in-situ alloying and nanoparticle additions to grain-boundary engineering to squeeze more performance from these alloys. Traditionally, exploring that composition space meant printing and testing dozens or hundreds of variants, an expensive and slow process given the cost of spherical metal powders and machine time. The new approach enables efficient preliminary screening using nothing more than compositional information and prior casting data, allowing researchers to discard poor candidates computationally and reserve laboratory resources for the handful of compositions most likely to succeed. Trend prediction, in which the model indicates how strength or ductility will shift as an element is added or removed, further guides the direction of experimental campaigns.

The study also speaks to a broader movement in materials science toward hybrid approaches that blend data-driven methods with physical understanding. Purely black-box machine learning models can interpolate well within their training data but often fail catastrophically when extrapolating to new processing routes or composition regimes, and their predictions can be difficult for metallurgists to trust or interpret. By anchoring the model in empirical descriptors with clear physical meaning and by structuring the learning problem around phase stability, the researchers preserved interpretability while retaining the efficiency of modern algorithms. The strategy echoes and extends recent successes in which machine learning has accelerated the discovery of high-entropy alloys, titanium alloys and aluminum alloys, but the cross-process element is distinctive: it explicitly addresses the situation every emerging manufacturing technology faces, where the data needed for good models do not exist yet and must be borrowed, intelligently, from adjacent fields.

There remain, of course, limits to what the approach can do. The systematic deviations in absolute property predictions mean that final qualification of any new alloy will still require real additive manufacturing experiments, and the model’s transferability rests on the assumption that phase stability remains the dominant organizing principle across processes, an assumption that may weaken for compositions near critical boundaries or for properties dominated by process-specific defects such as porosity and residual stress. The authors are candid that their method provides a pathway for preliminary screening and trend prediction rather than a replacement for experimental validation. Even so, the demonstration that a relative error of 0.81 percent is achievable for cross-process strength predictions marks a genuine advance. For a field racing to design alloys tailored to the extreme thermal conditions of 3D printing, the message is liberating: the data you need may already exist, sitting in decades of casting handbooks, waiting for the right physics-informed algorithm to translate them into the language of the laser.

Subject of Research: Machine learning-driven composition design of austenitic stainless steels for additive manufacturing using cross-process casting data and empirical phase-stability descriptors

Article Title: Composition design of austenitic stainless steels for additive manufacturing: a cross-process data and empirical knowledge-driven approach

Article References: Chen, C., Li, B., Tang, M., & Chen, H. (2026). Composition design of austenitic stainless steels for additive manufacturing: a cross-process data and empirical knowledge-driven approach. Journal of Materials Science. https://doi.org/10.1007/s10853-026-13666-z

Image Credits: AI Generated

DOI: 10.1007/s10853-026-13666-z

Keywords: additive manufacturing, austenitic stainless steel, machine learning, alloy design, chromium equivalent, nickel equivalent, divide-and-conquer model, genetic algorithm, tensile properties, phase stability, casting data, materials science

The material in question is 9Cr oxide dispersion strengthened steel, or 9Cr-ODS steel, widely regarded as a leading candidate for structural components in fusion blankets and for fuel cladding in advanced reactors. Its secret weapon is a dense dispersion of oxide particles only a few nanometres across, in this case of the complex oxide YTiTaO6, embedded in a ferritic-martensitic steel matrix containing roughly nine percent chromium. These particles pin dislocations, block grain boundary migration and resist coarsening under irradiation, which is precisely why the steel retains useful strength at temperatures where conventional reactor steels soften dramatically. But the particles are only effective while they remain small, numerous and finely dispersed.

Manufacturing large ODS components requires thermomechanical processing, in which the alloy is heated and rolled to consolidate and shape it. The trouble is that the high temperatures involved, particularly rolling above Ac3, the temperature at which the steel becomes fully austenitic, can trigger recrystallization of the matrix grains and, crucially, coarsening of the nano-oxides. Earlier work by other groups had shown oxide particle growth in 9Cr-ODS steels at temperatures above about 1473 kelvin, and the field has long sought processing routes that allow industrial-scale shaping without sacrificing the nanostructure. The Hefei team set out to compare, systematically, two representative rolling routes and to track what happens to the grains, the carbides and the nano-oxides in each case.

As a baseline, the researchers first characterized a tempered sample of the steel, which displayed a heterogeneous mixed microstructure with relatively coarse carbides sitting along grain boundaries and an average grain size of 0.80 micrometres. This starting point reflects the typical state of tempered martensitic ODS steel after standard heat treatment. From there, the team applied two contrasting thermomechanical routes: one in which rolling was performed above Ac3, fully transforming the structure to austenite, and another known as intercritical rolling, carried out within the two-phase region between Ac1 and Ac3, where austenite and ferrite coexist. They then examined the resulting matrix grains, carbide distributions and oxide particles, and measured tensile properties at 650 degrees Celsius.

The results were striking. Rolling above Ac3 induced partial recrystallization of the grains, and inside the recrystallized regions the nano-oxides agglomerated and coarsened noticeably. Even more telluric for the material’s future performance, the coarsened oxides lost the specific crystallographic orientation relationship they normally share with the surrounding steel matrix. That orientation relationship matters: it reflects a degree of crystallographic coherence between particle and matrix that is associated with interfacial stability. When coarsening breaks that relationship, the particles behave more like incoherent inclusions, and prior studies of irradiated ODS alloys have linked incoherent interfaces to poorer dispersoid stability at elevated temperature.

Intercritical rolling told a very different story. By keeping the deformation within the Ac1 to Ac3 window, the process effectively restrained both grain growth and nano-oxide coarsening. The fine YTiTaO6 particles retained their characteristic orientation relationship with the matrix, indicating that their coherent or semi-coherent interfaces survived the processing intact. In other words, the two-phase region acts as a thermal sanctuary for the nanostructure: hot enough to allow the plastic deformation needed for shaping, but not so hot or so transformative that the particles dissolve, agglomerate or lose their crystallographic anchoring to the surrounding iron lattice.

The mechanical consequences were measurable. The optimally intercritically processed steel achieved a higher grain boundary density and produced finer carbides, a microstructural combination the authors associate with improved tensile behaviour. At 650 degrees Celsius, a temperature squarely relevant to reactor operating conditions, the intercritically processed material reached an ultimate tensile strength of 473.4 megapascals with a total elongation of 33.1 percent. Compared with the conventionally high-temperature rolled counterpart, that represents higher strength at essentially comparable ductility, a combination that is notoriously difficult to achieve in high-temperature structural alloys, where strength gains usually come at the expense of the ability to deform without cracking.

The study’s significance extends beyond a single alloy recipe. By demonstrating that nano-oxide stability in 9Cr-ODS steel is temperature-dependent and can be regulated through the choice of processing window, the work clarifies the mechanism by which thermomechanical processing either preserves or degrades the strengthening dispersion. It also provides practical guidance for industrial fabrication of large-size ODS nuclear components, which is one of the field’s most persistent bottlenecks. Producing ODS steels at laboratory scale is well established; producing them as sizeable, reliable components with uniform nanostructure remains a genuine manufacturing challenge for fusion and fast reactor programmes worldwide.

For the broader materials community, the findings reinforce a growing appreciation that the interface between a nanoscale precipitate and its host matrix is as important as the precipitate itself. Coherent interfaces slow diffusion, suppress coarsening and help particles survive both thermal excursions and radiation damage. The Hefei team’s demonstration that a carefully chosen intercritical rolling window can preserve those interfaces through industrial deformation suggests a path toward ODS steels that are not only superb on paper but manufacturable at scale, bringing the dream of durable fusion blanket and advanced reactor cladding materials one practical step closer to reality.

Subject of Research: Microstructural evolution and nano-oxide stability in 9Cr-ODS steel during thermomechanical processing

Article Title: Microstructural evolution of 9Cr-ODS steel during thermomechanical processing

Article References: Zhu, G., Huang, T., Hai, C., & Huang, Q. (2026). Microstructural evolution of 9Cr-ODS steel during thermomechanical processing. Journal of Materials Science. https://doi.org/10.1007/s10853-026-13866-7

Image Credits: AI Generated

DOI: 10.1007/s10853-026-13866-7

Keywords: ODS steel, nano-oxides, thermomechanical processing, intercritical rolling, fusion materials, Generation IV reactors, YTiTaO6, grain refinement, tensile properties, carbides, recrystallization, nuclear structural materials

The researchers took C7 zirconium alloy, a material engineered for the punishing environment inside light water reactors, and deliberately charged it with hydrogen to a concentration of 800 parts per million. That figure is not arbitrary. Hydrogen accumulates in cladding over years of service through the slow corrosion reaction between the zirconium surface and the surrounding coolant, and 800 parts per million represents a level at which hydride damage becomes a serious engineering concern. Once the hydrogen was locked in, the team annealed samples at three temperatures: 470, 490, and 510 degrees Celsius, each for three hours, and then interrogated the resulting microstructures and mechanical properties.

The diagnostic toolkit combined scanning electron microscopy in backscattered electron mode, which reveals hydride morphology through compositional contrast, with electron backscatter diffraction, which maps the crystallographic state of the surrounding zirconium matrix. This pairing matters because hydride behavior cannot be understood in isolation. The mechanical response of a hydrided alloy depends on where the hydrides sit, how large they are, whether they form continuous networks, and what the matrix itself looks like after annealing, including how much of it has recrystallized into new, strain-free grains versus how much retains the deformed substructure of the original material.

At 470 degrees Celsius, the lowest annealing temperature, the hydrides remained fine and slender, appearing as plate-like and strip-like precipitates distributed in bands but without the kind of segregation that concentrates damage along single weak paths. This is the microstructural equivalent of a well-dispersed reinforcement: many small obstacles, none of them dominant. The zirconium matrix at this temperature retained much of its deformed character, since 470 degrees is only modestly above the temperatures at which recovery and recrystallization begin in these alloys.

Raising the temperature to 490 degrees Celsius produced a dramatic and, at first glance, counterintuitive change. Rather than dissolving harmlessly, the hydrides coarsened and aggregated markedly, forming continuous or semi-continuous layers draped along grain boundaries and interconnected networks threading through the grain interiors. Electron backscatter diffraction showed that this condition coincided with a lower recrystallized fraction, meaning the matrix still carried substantial stored deformation energy. The combination is metallurgically unfortunate: hydrides concentrated along boundaries create easy fracture paths, while an incompletely recrystallized matrix limits the material’s capacity for uniform plastic flow. Continuous boundary hydride networks are among the most damaging configurations a hydrided zirconium alloy can adopt, because cracks can propagate along these brittle plates with minimal energy absorption.

At 510 degrees Celsius, the highest temperature studied, the picture shifted again. The hydride distribution became more discontinuous and homogeneous, and after cooling the samples displayed acicular, needle-shaped hydrides. The authors interpret this as evidence of partial hydride dissolution during the anneal itself, followed by re-precipitation as the dissolved hydrogen came out of solution during cooling. In effect, the 510-degree treatment broke up the coarse boundary networks formed at intermediate temperatures, redistributed the hydrogen more evenly, and then re-precipitated it as finer, more dispersed needles rather than continuous films. This dissolution-and-reprecipitation cycle is a classic lever in physical metallurgy: by cycling a second phase into and out of solution, one can reset its size and spatial distribution.

The mechanical consequences were measured with circumferential tensile tests, an appropriate geometry for cladding tubes whose most service-relevant loading is hoop stress around the tube circumference. At room temperature, the 510-degree annealed condition delivered the standout result: an elongation to failure of 32.42 percent, the highest among the annealed conditions, accompanied by a ductile-dominated mixed fracture surface. In a material carrying 800 parts per million hydrogen, elongation above 30 percent is a meaningful achievement, and the ductile character of the fracture indicates that cracks were meeting substantial resistance rather than cleaving along brittle hydride paths.

The story changed at 350 degrees Celsius, a temperature relevant to reactor operating conditions. There, the ultimate tensile strength was only slightly affected by annealing temperature, but ductility peaked in an unexpected place: the 490-degree condition, the same one that produced the coarse boundary hydride networks, yielded the highest elongation of 41.31 percent. The authors attribute this to the combined effects of the retained substructured matrix, the particular hydride distribution, and the enhanced plastic deformation that zirconium alloys exhibit at elevated temperature. At 350 degrees, the matrix has enough thermal activation for dislocations to move and for local stress concentrations at hydrides to be relieved by matrix flow, so the embrittling penalty of the coarser hydrides is partially offset by the strengthening and work-hardening character of the substructure. It is a reminder that hydride embrittlement is not a fixed property but a temperature-dependent competition between brittle crack initiation and ductile stress relaxation.

Taken together, the results identify 510 degrees Celsius for three hours as the most favorable annealing condition among those examined, providing the best balance of room-temperature ductility, microstructural homogeneity, and strength in hydrogen-charged C7 alloy. The practical implication is that heat treatment schedules for zirconium cladding components, including the strips used in nuclear fuel positioning frameworks, which the authors’ funding acknowledgements suggest motivated the work, can be tuned not merely to relieve stress or recrystallize the matrix but specifically to manage hydride architecture. A treatment chosen at the wrong temperature, such as 490 degrees in this study, could inadvertently concentrate hydrides along grain boundaries and degrade room-temperature performance even as it appears to soften and homogenize the material.

For the nuclear industry, the study adds quantitative detail to a long-running effort to understand and mitigate hydrogen-induced damage in zirconium alloys, a problem documented across decades of research on Zircaloy cladding corrosion, hydride precipitation kinetics, hydride reorientation under stress, and the abrupt ductile-to-brittle transitions that delta hydrides can trigger. What distinguishes this work is its systematic coupling of three annealing temperatures to both hydride morphology and tensile response at two temperatures, showing that the optimal treatment depends on the service temperature of interest. As reactors age and fuel duty cycles intensify, that kind of temperature-specific guidance becomes increasingly valuable for keeping hydrogen’s silent sabotage in check.

Subject of Research: Thermal evolution of zirconium hydrides and its effect on the mechanical properties of hydrogen-charged zirconium alloys

Article Title: Thermal evolution of hydrides and subsequent mechanical response in hydrogen-charged zirconium alloys

Article References: Gao, Z., Bi, Q., Liu, Y., Cao, T., Yang, J., Yu, Y., Gao, Z., & Bai, G. (2026). Thermal evolution of hydrides and subsequent mechanical response in hydrogen-charged zirconium alloys. Journal of Materials Science. https://doi.org/10.1007/s10853-026-13800-x

Image Credits: AI Generated

DOI: 10.1007/s10853-026-13800-x

Keywords: zirconium alloys, zirconium hydrides, hydrogen embrittlement, nuclear fuel cladding, annealing, microstructure, tensile properties, ductility, recrystallization, grain boundaries, Journal of Materials Science, materials science

The research team, led by Urwa Arshad and corresponding author Sobia Naseem, tackled one of the central obstacles in plastics recycling: the fact that recycled polymers typically suffer degraded properties compared with virgin material, which limits them to low-value applications. Rather than accepting this downgrading, the group set out to upgrade recycled nonwoven polypropylene by introducing tiny amounts of carbon-based nanofillers, materials renowned for their exceptional stiffness, thermal resilience and aspect ratios. The goal was to determine whether the reinforcement effects documented for virgin polypropylene could be reproduced in a genuinely post-consumer feedstock, and to map precisely how filler type and loading govern the outcome.

Methodologically, the study stands out for its dual-step processing route. Instead of feeding nanofillers directly into a melt mixer, where hydrophobic carbon particles tend to clump into micron-scale agglomerates, the researchers first dispersed carbon nanotubes and graphene nanoplatelets in an aqueous medium assisted by ultrasonication. High-frequency sound waves generate intense cavitation bubbles that break apart nanotube bundles and exfoliate platelets before they ever meet the polymer. This pre-dispersion was then combined with the polypropylene in a twin-screw extruder, where controlled shear and residence time distributed the fillers through the molten matrix. The team tested loadings of 0.25, 0.5 and 1 weight percent for each filler, a deliberately narrow window designed to locate the tipping point between useful reinforcement and counterproductive agglomeration.

The results reveal a strikingly consistent pattern: less is more. At low loadings of 0.25 to 0.5 weight percent, scanning electron microscopy showed uniform dispersion of both nanofillers throughout the recycled matrix, and tensile testing confirmed that mechanical performance peaked at the lowest loading of 0.25 weight percent. When the filler content was pushed to 1 weight percent, however, the particles began to cluster, creating stress-concentrating defects that degraded the composite’s tensile behavior. This concentration-dependent boundary is a well-known phenomenon in nanocomposite science, but demonstrating it clearly in a recycled, nonwoven-derived matrix provides practical guidance for anyone attempting to industrialize the process: the sweet spot lies well below the loadings that intuition might suggest.

Thermal analysis told an equally compelling story. Thermogravimetric measurements showed that the temperature of maximum degradation, Tmax, shifted upward by as much as 53 degrees Celsius with 0.25 weight percent graphene nanoplatelets, and by 42 degrees Celsius with 0.5 weight percent carbon nanotubes, relative to neat recycled polypropylene. For a material recovered from discarded protective equipment, such a dramatic improvement in thermal stability is remarkable, and it suggests that the nanofillers act as both physical barriers to volatile degradation products and as efficient heat sinks within the matrix. In practical terms, a recycled compound that withstands temperatures tens of degrees higher can tolerate more aggressive processing and serve in more demanding end uses.

Differential scanning calorimetry added another layer of insight by revealing a strong nucleation effect. The crystallization temperature of the recycled polypropylene rose from 103.23 degrees Celsius to approximately 115 degrees Celsius in the presence of the nanofillers. This shift means the polymer chains crystallize earlier and more readily as the melt cools, because the carbon surfaces provide templates on which lamellae can grow. Faster, more complete crystallization at higher temperatures translates into shorter molding cycles and potentially more uniform crystalline morphology, both of which matter enormously in industrial polymer processing. The effect also indicates that the fillers were not merely inert inclusions but active participants in shaping the material’s microstructure.

Spectroscopic evidence helped explain why the low-loading composites performed so well. Fourier-transform infrared spectroscopy confirmed strong interfacial interactions between the carbon nanofillers and the polypropylene matrix, arising from a combination of partial chemical bonding and physical adsorption. In nanocomposites, the interface is everything: load transfer from the weak polymer to the strong filler occurs across this boundary, and any gap or poor wetting undermines the reinforcement. The FTIR findings, combined with the morphological observations, indicate that the ultrasonication-assisted pre-dispersion step succeeded in delivering well-separated particles that the molten polymer could then wet and bond with effectively during twin-screw blending.

Not every property followed the same concentration trend, and this nuance is one of the study’s most useful contributions. While tensile performance declined at 1 weight percent loading, Shore D hardness increased progressively with filler content, rising by up to 13.7 percent across the range tested. This divergence makes physical sense: hardness is a surface and bulk resistance to indentation that benefits from the mere presence of rigid particles, even when those particles cluster, whereas tensile strength is exquisitely sensitive to agglomerate-induced flaws. The lesson for product designers is that the optimal formulation depends on the target property, and that a single loading cannot simultaneously maximize every mechanical characteristic.

The broader significance of the work lies in its circular-economy framing. Post-consumer personal protective equipment represents an expanding yet largely unrecycled fraction of global municipal solid waste, and nonwoven polypropylene in particular has few established recycling pathways because the fabric form complicates sorting and reprocessing. By demonstrating a scalable melt-blending route that converts this waste into functional nanocomposites, the Lahore team offers a template that builds on their earlier work on upscale recycling of nonwoven polypropylene using novel blending methods. The processing equipment involved, namely ultrasonic baths and twin-screw extruders, is standard in polymer manufacturing, which means the approach does not require exotic infrastructure to be adopted at industrial scale.

There remain, of course, questions that the study’s preview data do not fully resolve, including the economics of collecting and sanitizing PPE waste, the consistency of feedstock quality, and the long-term durability of the composites in service. The authors note that data are available on request, and the work was supported by the Pakistan Science Foundation under its Competitive Research Program. Still, the central message is clear and provocative: the mountains of discarded masks and gowns that symbolize a global health crisis need not be a permanent environmental liability. With the right nanoscale engineering, they can be reborn as thermally stable, mechanically reinforced materials, proving that even the most problematic plastic waste can be upgraded rather than merely disposed of.

Subject of Research: Upcycling post-consumer PPE polypropylene waste into carbon nanofiller-reinforced nanocomposites

Article Title: Dual-step melt blended recycled polypropylene nanocomposites: interfacial behavior and reinforcement effects of carbon nanofillers

Article References: Arshad, U., Gill, Y. Q., Ijaz, M. W., Naseem, S., & Rizwan, M. (2026). Dual-step melt blended recycled polypropylene nanocomposites: interfacial behavior and reinforcement effects of carbon nanofillers. Polymer Bulletin, 83(12), Article 656. https://doi.org/10.1007/s00289-026-06717-7

Image Credits: AI Generated

DOI: 10.1007/s00289-026-06717-7

Keywords: recycled polypropylene, carbon nanotubes, graphene nanoplatelets, nanocomposites, PPE waste recycling, melt blending, thermal stability, tensile properties, crystallization, nonwoven waste, circular economy, interfacial interactions

The motivation behind the work lies in one of the most stubborn limitations of conventional lightweight alloys. Aluminum alloys in the AA2XXX and AA7XXX families, the workhorses of the aerospace industry, rely on fine precipitates for their strength, but those precipitates dissolve or coarsen above roughly 200 degrees Celsius, causing properties to collapse. For applications such as supersonic aircraft, where skin temperatures can climb well beyond that threshold, designers have had no choice but to turn to heavier titanium or nickel-based materials, sacrificing fuel efficiency and increasing carbon dioxide emissions. Cerium-containing aluminum alloys offer a way out of this bind. When cerium is added to aluminum, it forms an Al11Ce3 intermetallic phase whose slow diffusion and low solubility in the aluminum matrix make it exceptionally resistant to coarsening, with stability demonstrated up to at least 350 degrees Celsius. Cerium is also a relatively abundant rare earth element, currently lacking high-volume applications, and is recovered as a byproduct of refining more valuable rare earths, which makes the alloy family economically attractive.

The alloy was prepared by melting high-purity aluminum with aluminum-cerium and aluminum-magnesium master alloys at 850 degrees Celsius and casting the melt into a graphite mold. Inductively coupled plasma analysis confirmed the composition at 10.096 percent cerium and 4.676 percent magnesium by weight, with only trace iron and silicon. The researchers deliberately chose the 4 percent magnesium content based on CALPHAD thermodynamic modeling, which showed that deforming the material at or just below the solvus of a τ-Al13CeMg6 intermetallic and above the metastable β-AlMg solvus would allow the high dislocation density generated during deformation to stimulate the nucleation of fine precipitates either during processing or upon subsequent cooling. The cast rods were then hot extruded at 300 degrees Celsius to a final diameter of 9.5 millimeters, an 86 percent reduction in cross-sectional area, using conditions deliberately kept within the envelope of conventional industrial practice.

Microstructural characterization revealed exactly why the extrusion was so effective. X-ray diffraction, scanning electron microscopy, electron backscatter diffraction, differential scanning calorimetry and high-resolution transmission electron microscopy together painted a detailed picture of the transformation. In the as-cast condition, the alloy displayed the classic script-like eutectic structure of Al11Ce3 intermetallics distributed unevenly through a large-grained aluminum-magnesium matrix, with an average grain size of about 23.6 micrometers. After extrusion, the grains shrank to roughly 7.2 micrometers and the intermetallic particles were broken down and aligned into the extrusion direction, with the average dispersoid size falling from 3.57 to 1.91 micrometers. The dislocation density, calculated from kernel average misorientation measurements, roughly doubled, and a brass-type crystallographic texture emerged. Crucially, when the extruded material was annealed at 300 degrees Celsius for up to 100 hours, neither the grain size nor the dispersoid size changed to any statistically significant degree, confirming that the refined structure is thermally stable.

The mechanical property gains were striking across the board. Yield stress and tensile strength increased by factors of roughly 2.5, while elongation at failure jumped from 3.8 percent to 17.2 percent. Hardness rose from 98 HV in the cast state to 114 HV after extrusion. The simultaneous improvement in strength and ductility, a combination that is often difficult to achieve, was attributed to the uniform refinement of the intermetallic distribution, the increased dislocation density and the development of texture. Fractographic analysis told a consistent story: cast samples failed along cleavage facets in a brittle manner, while extruded samples fractured through dimpled, ductile surfaces, reflecting intense plastic deformation before failure. The extruded material also reached a Young’s modulus of about 75 gigapascals.

Thermal endurance proved to be the alloy’s signature quality. After exposure to 300 degrees Celsius for 100 hours, the as-cast alloy retained 82 percent of its room-temperature strength, and the extruded version retained 88 percent. Fracture toughness, measured for the first time in any Al-Ce-Mg alloy using the single-edge bend method of ASTM E1820, reached 23 kilojoules per square meter in the extruded state, a value comparable to some of the highest figures reported for aluminum alloys, including aluminum-lithium systems and wire-arc additively manufactured Al-Si. After the same 100-hour anneal, 97.6 percent of that toughness was preserved. Compared with the commercial high-temperature alloy AA2618, which loses about 90 percent of its room-temperature strength at 300 degrees Celsius, the ternary alloy held on to roughly 40 percent of its strength, although it fell short of the binary Al-10Ce alloy, which retains about 70 percent.

The reason for that trade-off lies in the magnesium. Transmission electron microscopy revealed fine, nanoscale magnesium-rich precipitates both within the aluminum grains and along their boundaries, too small to be detected by X-ray diffraction or EBSD. Differential scanning calorimetry showed an endothermic dissolution event with an onset at about 215 degrees Celsius, and a weak exothermic peak on cooling at roughly 360 degrees Celsius, indicating that these precipitates dissolve as the temperature climbs and re-form when it falls. The researchers estimate that the magnesium-rich precipitates account for about 55 percent of the room-temperature flow stress, which explains why strength drops rapidly in the 150 to 250 degrees Celsius range and why the extra strength of the ternary alloy over the binary Al-10Ce is essentially erased above 250 degrees Celsius. The reversible nature of the dissolution and re-precipitation, however, means the material recovers its properties on cooling, ensuring excellent property retention through thermal cycling.

The study also documented an unusual deformation behavior: negative strain rate sensitivity at room temperature. Tensile tests across strain rates from 10^-5 to 10^-2 per second revealed serrated flow curves characteristic of the Portevin-LeChatelier effect, driven by dynamic strain aging as magnesium atoms diffuse toward and interact with moving dislocations. The strain rate sensitivity parameter reached about minus 0.03 at intermediate rates, a range similar to that observed in the commercial alloy AA5182, which contains 4.5 percent magnesium. The serrations vanished at both the fastest and slowest rates and disappeared entirely at 150 degrees Celsius, confirming the diffusion-controlled origin of the phenomenon and demonstrating that sufficient magnesium remains in solid solution even after processing.

By decomposing the strength contributions of different mechanisms, the team showed that the refinement from extrusion contributes about 106 megapascals to the yield stress, split roughly equally between the increased dislocation density and the finer grains and intermetallics, while magnesium in solid solution adds another 32 megapascals in the cast state. The most impressive single contribution, however, is strain hardening: the extruded ternary alloy hardens by about 190 megapascals between yielding and failure, nearly four times the strain hardening of the equivalent binary alloy, an effect the authors link directly to the nanoscale magnesium-rich precipitates. Taken together, the results establish extrusion-processed Al-10Ce-4Mg as a serious candidate for lightweight, temperature-resistant structural applications, from supersonic airframes to powertrain components, and demonstrate that the path from laboratory curiosity to industrial adoption may run through the most conventional of metalworking processes.

Subject of Research: Thermomechanical extrusion processing of a ternary Al-Ce-Mg alloy to enhance strength, toughness and thermal stability.

Article Title: Effect of thermomechanical processing on mechanical properties and the microstructure of ternary Al-Ce-Mg alloy

Article References: Singh, G., Odhiambo, H. W., Tasneem, M. H. B., Ouyang, G., Soare, M. A., Kang, M.-C., Zhou, L., Cui, J., Napolitano, R. E., & Picu, C. R. (2026). Effect of thermomechanical processing on mechanical properties and the microstructure of ternary Al-Ce-Mg alloy. Journal of Materials Science: Metallurgy, 1(1), Article 11. https://doi.org/10.1007/s44492-026-00011-3

Image Credits: AI Generated

DOI: 10.1007/s44492-026-00011-3

Keywords: Al-Ce-Mg alloy, extrusion, thermomechanical processing, Al11Ce3 intermetallic, fracture toughness, high-temperature strength, Mg-rich precipitates, microstructure, dynamic strain aging, thermal stability, lightweight alloys, tensile properties

PVDF is a semicrystalline fluoropolymer prized for its chemical resistance, thermal robustness, mechanical durability, and innate hydrophobicity. It is also polymorphic, crystallizing in phases that include the nonpolar alpha form and the piezoelectric beta phase, a combination that makes it attractive for filtration membranes, sensors, energy harvesters, and water-oil separation systems. The Brazilian group, led by Gabriel da Cruz Dias of the Universidade Estadual de Maringá together with colleagues at UNESP and UNIFUNEC, asked a question that most prior studies had avoided: what happens when clay content climbs far beyond the low loadings, typically at or below 10 percent by weight, that researchers usually dare to use?

The answer required an unconventional fabrication route. Electrospinning, the dominant method for producing polymer nanofibers, relies on high electric fields, delivers modest production rates, and is difficult to scale. Solution Blow Spinning instead uses a pressurized gas jet to draw fibers from a polymer solution, offering higher throughput, simpler equipment, and lower cost. In the experiments, PVDF was dissolved in N,N-dimethylformamide at a concentration of 30 percent weight per volume, and montmorillonite K10 clay was dispersed in the solvent beforehand at levels of 3, 5, 10, 20, and 30 percent relative to the polymer mass. The mixture was fed through the nozzle at 76 microliters per minute under an air pressure of 140 kilopascals, with a working distance of 21 centimeters, yielding self-supported fibrous mats whose average fiber diameters stayed below 200 nanometers.

Scanning electron microscopy revealed that pure PVDF fibers were uniform, smooth, and cylindrical, with only occasional bead defects caused by jet instability and incomplete solvent evaporation. As clay content rose, imperfections multiplied. At 30 percent loading, clay clusters became prominent, hanging from the fibers like stones tied to a string and, according to the authors’ vivid analogy, pulling the loaded jets toward the collector faster than the polymer threads could follow. The consequence was measurable: with a constant 5 milliliters of solution for every production run, composite thickness fell only slightly up to 10 percent clay but dropped considerably at 20 and 30 percent, signaling that the process was losing efficiency as aggregates sabotaged fiber deposition.

Perhaps counterintuitively, the viscosity of the spinning solutions decreased as more clay was added. The explanation lies in sample preparation: because the clay was first dispersed in DMF, higher clay concentrations demanded additional solvent, and this dilution effect outweighed the thickening expected from suspended solid particles. Thinner solutions allowed polymer chains to disentangle and stretch more easily under the constant air pressure, producing finer fibers at high clay contents. Up to 10 percent clay, where solvent content remained constant, the trend reversed, with diameters rising slightly as viscosity increased—behavior consistent with earlier electrospinning and blow-spinning studies of clay-laden polymer solutions.

Elemental mapping by energy-dispersive X-ray spectroscopy confirmed that silicon, oxygen, and aluminum from the clay were distributed fairly homogeneously through the fibrous networks, with localized clustering at the two highest loadings. X-ray diffraction showed that both the alpha and beta crystalline phases of PVDF coexisted in every mat and, crucially, that clay addition did not alter this balance. The researchers attribute the phase formation to the blow-spinning process itself, in which rapid solvent evaporation favors the alpha phase while simultaneous fiber stretching promotes the beta phase, and they note that room-temperature processing may explain why clay did not act as the beta-phase nucleation catalyst reported in studies using heated routes. Differential scanning calorimetry reinforced the structural picture, showing a single melting peak near 176 degrees Celsius for all compositions, with a shoulder attributed to melting of the interfacial region between crystalline and amorphous lamellae.

Thermogravimetric analysis, however, delivered a caution. Pure PVDF remained stable to roughly 420 degrees Celsius before losing about 75 percent of its mass through dehydrofluorination, the reaction that releases hydrogen fluoride and leaves behind unsaturated carbon species. The composites degraded in two steps, and an initial degradation peak intensified with clay content, evidence that dispersed rather than exfoliated clay particles, and particularly the metal ions within them, catalyzed chain scission. Without the barrier effect that well-exfoliated clay layers would provide, higher filler contents reduced the composite’s thermal stability rather than enhancing it.

Mechanical testing traced a clear arc from reinforcement to collapse. Pure PVDF membranes were highly ductile, stretching about 58 percent before breaking with an elastic modulus near 13 megapascals, behavior typical of fibrous mats in which fibers align and reorient under load. Adding up to 5 percent clay increased both stiffness and tensile strength, because rigid particles acted as bridging points that transferred stress between neighboring fibers. Beyond roughly 5 to 10 percent, however, performance deteriorated sharply: elongation at break fell steadily with filler content, and at 30 percent clay the tensile strength had dropped by about 54 percent relative to pure PVDF. The culprit, visible in both electron micrographs and elemental maps, was agglomeration—defects that concentrate stress and sever efficient load transfer through the network.

Wettability measurements added an unexpected twist. All composites remained hydrophobic, with static water contact angles of 114 degrees for pure PVDF and, apart from a slight dip at 3 percent clay, rising values of 115, 119, 122, and 123 degrees as clay content climbed to 5, 10, 20, and 30 percent. Because montmorillonite is inherently hydrophilic, the trend runs against intuition, but the authors attribute it to increased surface roughness imparted by clay particles, which amplifies the low surface energy conferred by PVDF’s strongly bonded carbon-fluorine groups. A water-ethanol mixture reduced the contact angles, offering a simple lever for tuning wettability in applications such as filtration and separation without changing the material’s composition.

The study’s broader message is twofold. First, Solution Blow Spinning is a genuinely viable and scalable route to PVDF-based fibrous nanocomposites, producing sub-200-nanometer fibers at productivity levels electrospinning cannot match—here, 5 milliliters of a 30 percent solution yielded mats roughly 500 micrometers thick. Second, the technique’s tolerance for inorganic fillers has hard limits. The 30 percent formulation, with its thin, fragile mats, abundant aggregates, and depressed mechanical performance, marks a boundary that future work should not cross without improved dispersion strategies. For now, the sweet spot sits at low clay contents, where composites combine hydrophobic surfaces, controllable porosity, and enhanced stiffness, positioning them for filtration, environmental remediation, and membrane applications. The authors suggest that next steps should focus on refining filler dispersion, optimizing processing parameters, and evaluating functional metrics such as permeability, selective adsorption, and long-term stability under realistic operating conditions.

Beyond the specific findings, the study offers a useful methodological template for anyone probing the limits of high-throughput fiber fabrication. Rather than presenting a single optimized formulation, the authors deliberately swept a wide compositional range and documented where each characterization technique registered a change in behavior. This threshold-mapping approach—tracking solution viscosity, fiber diameter, mat thickness, crystalline phase content, degradation onset, tensile response, and contact angle as a function of filler loading—makes the results directly comparable across laboratories and provides benchmarks against which future dispersion strategies can be judged.

The work also situates itself within a broader shift in fibrous membrane research. Electrospinning remains the reference technique for producing submicron polymer fibers, but its dependence on high voltages and modest throughput has long constrained industrial adoption. Solution Blow Spinning, by contrast, requires only a pressurized gas supply, a syringe pump, and a rotating collector, all of which are inexpensive and straightforward to operate. The demonstration that this setup can accommodate inorganic loadings up to 30 percent by weight—while still yielding self-supported mats with sub-200-nanometer fibers—suggests that the technique’s real bottleneck is not fiber formation itself but the rheology and homogeneity of the precursor solutions.

For applications, the combination of hydrophobic surfaces, interconnected porosity, and tunable stiffness is particularly relevant to membrane-based separations, where water-repellent fibrous mats are sought for treatments involving oily wastewater or for supporting catalytic and adsorptive phases. The observation that a simple water-ethanol test liquid lowers contact angles hints at how wetting behavior could be tuned in practice, either through liquid choice or through post-processing that modifies surface texture. The authors’ open-access publication, released in the first volume of the Journal of Materials Science: Polymers, makes the full dataset available to groups working on scalable nanocomposite fibers, and the identification of agglomeration as the dominant failure mode above roughly 10 percent clay gives a clear target for subsequent studies employing surface-modified or organophilic clays, alternative solvents, or in-line dispersion methods.

Subject of Research: Fabrication of hydrophobic PVDF/clay fibrous nanocomposites via solution blow spinning with tunable mechanical properties

Article Title: Hydrophobic PVDF/clay fibrous nanocomposites prepared by solution blow spinning with tunable mechanical properties

Article References: Dias, G. D. C., Zadorosny, L., Sanches, A. O., de Paula, F. R., dos Santos, M. C., & Malmonge, L. F. (2026). Hydrophobic PVDF/clay fibrous nanocomposites prepared by solution blow spinning with tunable mechanical properties. Journal of Materials Science: Polymers, 1(1), Article 19. https://doi.org/10.1007/s44493-026-00018-1

Image Credits: AI Generated

DOI: 10.1007/s44493-026-00018-1

Keywords: PVDF, montmorillonite clay, solution blow spinning, nanocomposites, nanofibers, hydrophobicity, contact angle, tensile properties, thermal stability, crystalline phases, fibrous membranes, morphology