John L. Provis- PSI Center for Nuclear Engineering and Sciences, Paul Scherrer Institute, Villigen, Switzerland
The Structural Materials section of Frontiers in Materials has published almost 1000 articles during the past 10 years, and this article commemorates and assesses the development of thematic trends among this body of works. The three “grand challenges” identified upon launching the section in 2015 remain relevant: Bottom-up design of materials; Sustainable materials supply; Durable materials for extreme environments. Additional topics that have featured extensively in the journal include Natural materials; AI and machine learning; Materials characterisation and performance; Composites and manufacturing. It is anticipated that each of these topics will continue to contribute to the future of Structural Materials research, taking advantage of advances in computational capabilities and the ever-stronger impetus for sustainable selection and use of materials to drive future innovation in this discipline.
Introduction
This short article is intended to mark the 10th year of publication of Frontiers in Materials, and specifically the Structural Materials section within this Journal. As the Specialty Chief Editor of this section, it is a genuine pleasure to be able to describe its growth from a start-up initiative based on a new publishing model, to its current status as an established and successful part of the publishing ecosystem.
In the first year of publication of the Structural Materials section, I was asked to draft a “Perspectives” type of article describing the Grand Challenges facing the field. Ten years on, this article will take the opportunity to revisit that document (Provis, 2015), reassess the challenges that were identified (and the role of the journal in addressing those challenges over the past decade), and provide further thoughts about key directions and trends in research as the journal moves into its second decade.
I would like to preface some of these remarks by repeating a quotation that was used in the conclusions section of Provis (2015), variously attributed (O’Toole, 2013) to Niels Bohr, Yogi Berra, or an anonymous Danish politician – “It is difficult to make predictions, particularly about the future”. Nonetheless, it seems worthwhile to revisit and reassess past predictions, and to provide some new observations, as there are clear trends in materials research directions, published both within and beyond the Structural Materials section, which are relevant to the community.
Reassessment of 2015 trends
The three Grand Challenges identified in the inaugural Grand Challenges article in this Section (Provis, 2015) were:
1. Bottom-up design of materials for improved properties
2. Sustainable supply: Sourcing, recycling, and diversification
3. Materials for extreme environments, and durability
These topics have all formed core aspects of the scope of the Structural Materials section of Frontiers in Materials, but each has also evolved in ways that were maybe not wholly predictable (at least by the current author) a decade ago. To address each in turn:
The bottom-up design and synthesis of materials is now in many ways central to the field of materials science. The application of fundamental chemical insight and computational materials science in this methodology is now very well advanced (Antypov et al., 2025). Materials “design” by artificial intelligence has also been featured very prominently–and rather controversially–in the materials chemistry community in recent years (Cheetham and Seshadri, 2024). In large part, these efforts have tended to focus more on (multi)functional materials rather than on structural materials, but there have been notable advances in alloy design (Ghafarollahi and Buehler, 2025) which are very relevant to the scope of the Structural Materials section. The fact that the bulk structural properties of some other engineering materials with composite microstructures–particularly concrete, biomaterials such as wood, and others - tend to be defined at the microstructural rather than chemical level means that the full implementation of bottom-up materials design by stepping across length scales in a multiscale approach remains a future target rather than a current reality in many areas of structural materials research. However, the use of computational approaches in the design of complex multicomponent alloys–whose nanoscopic structure does play a major role in defining engineering properties - is an obvious target which has featured prominently in the Structural Materials section (Tian, 2017; Li et al., 2020) as well as in broader discussions (George et al., 2019; Ma and Liu, 2024). There is also a highlighted need for better experimental techniques to characterise nanostructured materials, and in particular also validated computational approaches for assessment of data obtained by these techniques (Garboczi and Lura, 2020; Zhang et al., 2024).
The sustainable supply of structural materials continues to be at the forefront of the minds of many in academia and industry. These materials are produced at the giga-scale, and so the global transformation to circular or low-emissions production routes will necessarily be slow, but there are increasingly successful efforts to innovate for sustainability in bulk materials production and structural usage. As notable examples, direct-reduction iron- and steel-making is now being implemented at scale in place of blast furnace with a market share of 8% in 2022 (International Energy Agency, 2023), lower-clinker or zero-clinker cements derived from natural resources or industrial wastes are increasingly being accepted in place of conventional Portland cement (Kanavaris et al., 2023; Rossi et al., 2023; Hooton and Riding, 2025), and the possibility of achieving some project sustainability objectives through high-rise timber construction is leading to increased progress in that application of bio-derived structural materials (Tupenaite et al., 2021; Ilgın, 2024).
New and more sustainable process routes provide new opportunities for improvements in material performance, but also raise challenges related to some of the synergies between different industries that have been developed over the past century or two. As mentioned in the 2015 Grand Challenges discussion article (Provis, 2015), but still highly relevant, the phase-out of coal-fired electricity generation in some countries is leading to significant shortages of coal fly ash, which is also a valuable constituent in durable and low-carbon cementitious blends. This is driving the use of an increasingly diverse range of metallurgical slags, mining wastes, biomass combustion ashes, natural clays, and other under-utilised resources, in cement and concrete production, both blended with Portland cement and also as a constituent of non-Portland cementitious binders (Criado et al., 2017; Juenger et al., 2019; Wang et al., 2019; Khan et al., 2025). It appears that the way forward for sustainability in cement and concrete production must be fundamentally built on diversification of materials supply, making use of locally available resources and validating the performance of the materials that are produced from them (Provis et al., 2024). A particularly prominent article in the Structural Materials section–currently (mid-2025) the most-cited in the Section to date–also presents insight into the use of biological processes in production of concrete-like materials (Castro-Alonso et al., 2019), which is continuing to be highlighted as an attractive pathway to production of specialist, highly sustainable, structural materials (Armistead et al., 2023).
The design of materials–and combinations of materials, in the structural context–to withstand extreme service environments with high durability remains a central focal point for the research community. This was discussed in 2015 (Provis, 2015) and is increasingly topical now. The need to build sustainably–which means durably–and to fundamentally adapt our built environment to account for the added demands of climate change, is highlighted in key policy discussions (United Nations Environment Programme, 2024; United Nations Environment Programme, 2025). Lightweighting and increased performance (or efficiency) demands in applications ranging from industrial machinery (Zhao et al., 2024) to aerospace or automotive construction (Gao et al., 2021; Jia et al., 2022; Candela et al., 2024) mean that new ways of formulating, producing, processing and finishing materials are needed to withstand higher mechanical loads while using less material. Joining of dissimilar materials is increasingly necessary in this respect (Buffa et al., 2022; Guo et al., 2022; Beygi et al., 2023), and this has also been a successful focus area within the Structural Materials section with three completed Research Topics published in this field.
As a further example of “extreme” service conditions for structural materials, the 2015 Grand Challenges article mentioned the importance of materials under nuclear reactor service conditions, and this remains a valid consideration, particularly when service life extension may lead to reactors exceeding 60 years operational life spans (Busby et al., 2008), and where next-generation reactors (fission and fusion) are likely to require structural materials designed in support of advanced fuel, containment, and cladding systems (Wang et al., 2021; Malerba et al., 2022; Quadling et al., 2022). However, there is increasing consciousness that the availability of suitable pathways to waste disposal must be considered a pre-requisite for ongoing and new-build nuclear operations (Drace et al., 2022). The materials used in a nuclear waste repository (both wasteform and engineered barrier system components) are required to serve a designed purpose until the contained radioisotopes no longer pose a health or environmental hazard, which is in many cases a timescale on the order of 105–106 years (Frankel et al., 2021). This type of timescale is inherently difficult for researchers (and humans in general) to even comprehend, let alone design for Moser et al. (2012), Poirot-Delpech and Raineau (2016). Designing materials to serve for this type of timescale poses extreme challenges in terms of both durability design and engineering performance prediction (Ma et al., 2024), opening significant opportunities for advances in structural materials research.
Figure 1 shows a breakdown of the articles that have been published in the Structural Materials section since 2015 (with articles from 2015 to 2019 grouped together because of the smaller number of published items in those years), showing that around half of the articles published in this section each year could be identified as (fairly broadly) addressing one or more of the three Grand Challenges identified in the 2015 article. The relative proportions of these vary over time, but the three topics collectively do continue to provide a strong platform for the Structural Materials section.
Figure 1. Breakdown of research articles published in the Structural Materials section of Frontiers in Materials since 2015, classified broadly by topic according to the three Grand Challenges mentioned above, as well as other prominent topics within the Section. Data for 2025 are for papers published up to mid-September 2025. Classification was carried out manually based on the author’s assessment of the title and abstract of each paper.
Of the remaining articles, the majority fall into four main categories:
• Natural materials–an area which is rapidly growing in prominence in the journal, and including timber, stone, and soil-based materials in structural applications;
• Application of artificial intelligence (AI) or machine learning to the assessment of structural materials–which is an area of growth that will be discussed in more detail below;
• Tools and techniques to assess materials and their structural properties–a foundational aspect of the field of materials science which also underpins many of the Grand Challenge activities;
• Composites and investigation of manufacturing/processing–grouped together because so many of the important processes applied to materials result in composites, and gaining synergy from the interaction between multiple system constituents in a structural sense.
The classification presented in Figure 1 should be considered semi-quantitative at best, because it results from manually classifying each published paper into a single category only, where obviously many of the papers published are addressing more than one of the seven main listed themes. Nonetheless, it could be argued that a hallmark of a successful broad-based materials science/engineering-focused journal is its ability to provide articles addressing both key challenge topics and foundational aspects of the discipline, and the evidence provided here indicates that the Structural Materials section is continuing to provide a useful platform for the community in this regard.
Perspectives and current trends
Each of the points discussed in the preceding section remains in many ways relevant to the future of materials research, and each has formed an important part of the contribution of the Structural Materials section over the past decade. However, it is important to comment specifically on what has become the central focus of a very large field of activity in materials research: the use of machine learning models and their application to “big data” datasets, to optimise various characteristics of materials–in many cases from purely empirical viewpoints, although with increasing recognition that “physics-informed” models potentially offer significant advances in insight and predictive capabilities beyond solely conducting numerical interpolation in a black-box type model (Wagner and Rondinelli, 2016; Stoll and Benner, 2021; Li et al., 2022). This is particularly important when addressing materials behaviour under the types of extreme conditions discussed above, where the collection of experimentally measured data under the target conditions may be very difficult, and so it becomes necessary to extrapolate from data measured at less extreme pressures, or temperatures, or material history of exposure to potential damage (e.g., mechanical stress, irradiation, fatigue). When this extrapolation can be carried out using a physics-informed model, this has the potential to provide much greater insight, and potentially a more accurate description of material characteristics.
Nonetheless, the ability to predict target optimum compositional (or microstructural) design of materials based on purely data-driven approaches does appear to hold significant appeal to the research community. The “bestiary” of machine-learning algorithms, so named because many algorithms claim to make reference to different aspects of animal behaviour (Campelo and Aranha, 2023) (see also https://fcampelo.github.io/EC-Bestiary/ for an extensive list of examples), may appear to provide opportunities for numerous publications by application of a particular model to a given (often open-source dataset), but almost invariably these are of limited novelty. As a point of interest here, it could be noted that the paper describing the UC Irvine dataset for compressive strength of concretes (Yeh I. 1998; Yeh I. C. 1998), containing 1030 curated records, has been cited more than 1000 times, and this sits in addition to numerous articles which have evidently used the same dataset for model training without correct attribution, as becomes clear from searching for “1030 entries” as a keyword in Google Scholar in conjunction with other terms related to concrete strength and machine learning. It is maybe relevant to ask how much further value (or how many useful publications) can be gained from fitting and re-fitting the same data set, compiled almost 30 years ago from data older than that, with an ever-increasing number of model variants–but this does still appear to be fertile ground for authors.
It has been decided that the Structural Materials section will only publish papers of this type, fitting an existing dataset using a previously published model architecture, when true advancements in technical insight are gained; such cases remain rare, and so do such publications in this journal. It is also important to note the very helpful guidance provided by the authors of the “REFORMS” checklist for machine-learning-based science (Kapoor et al., 2024), which gives a set of questions and corresponding guidelines that are recommended to be followed by researchers intending to make the best possible use of the opportunities presented by machine learning in their research.
The principle applied in this discussion is that it is essential to be very careful when moving away from fundamental physically-based expressions, to rely solely on collected data as the basis for a predictive model. Such models can have impressive power of interpolation within the domain of the existing dataset, but are very much more limited when used to extrapolate beyond the scope of previous observations. This does recall the comment in the Grand Challenges article of 10 years ago (Provis, 2015) that “greater crosstalk is needed between the theoretical and practical aspects of the field of structural materials science” – and this becomes even more important when the theoretical side of the relationship is less closely connected to underlying physical laws via the use of empirical or surrogate models. These do clearly offer important advantages in speed-up of calculations, e.g., a factor of 500 speed-up from a surrogate model replacing a full geochemical code for prediction of cement hydrate phase assemblages in Boiger et al. (2025), which makes calculations feasible that may have otherwise been computationally impractical (or at least inconveniently slow). However, it remains essential to close the loop on these models, re-checking the outputs against a physically-based method, to ensure that the results are meaningful (Churakov et al., 2024).
As a toy example of this challenge, Figure 2 shows two pictures generated by a current (September 2025) implementation of the large language “artificial intelligence” model ChatGPT–one on a materials-science based question (Figure 2a), and one on a more general topic (Figure 2b). It can be seen from these illustrations that the AI-generated answers are superficially plausible but incorrect in key details. The sketch of concrete has an aggregate particle labelled as “hardened cement paste”, and the word “aggregates” misspelled; the map of Australia is missing the second-largest city (showing the first, third and fourth largest) and the names of one state and one territory, despite the prompt specifically asking for these to be included.
Figure 2. Diagrams generated by ChatGPT (current as of Sept. 2025) in response to the following prompts: (a) “Please draw a detailed sketch of the microstructure of reinforced concrete, labelling the remnant cement grains, aggregates, hardened cement paste, and other important features”. This contains numerous errors as outlined in the text. (b) “Please draw a map of Australia, showing state borders, and naming all states, territories, and major cities”. This is mostly correct with only small errors in terms of the shoreline shape and positioning of borders, but omits the name of the state of Tasmania and of the Australian Capital Territory, and omits the city of Melbourne (population >5.3 million), which is the second-largest city in Australia.
This is one of the main risks in the broader use of AI-based tools in research more generally: the generation of answers which are sufficiently close to plausible that a non-expert will believe them to be true, but fundamentally flawed. There are well-known examples of AI “hallucinations” causing obvious errors on technical points, but it is these subtler errors which are in many ways more insidious–and more difficult to keep out of the scientific literature when generative AI is used to prepare manuscripts–because they are close enough to the truth to be believable, yet actually wrong in critically important details.
Setting aside questions of AI and moving more to view more broadly current trends in the field of materials research, it is clear that the use of natural or bio-inspired materials–including the use of engineered biological processes for production of chemicals or materials that may not themselves be naturally occurring–will continue to gain prominence in coming years. This is both because of the unique engineered (or architected) structures that can be gained from this type of approach, and also relates to the possibility to reduce the environmental footprint of current processes (Sandak and Butina Ogorelec, 2023). It is of course not necessarily guaranteed that a “bio” method to produce a material will necessarily be more sustainable or efficient, but the chance at least to innovate and explore new possibilities can spark creativity and advancement in the field of structural materials. Although the large-scale uptake of truly biomimetic materials or processes in the field of structural materials is not currently seeing the same rate of growth as it is in some other fields of materials science (Chayaamor-Heil et al., 2023) - particularly in topics such as surfaces, coatings and pigments (Sandak and Butina Ogorelec, 2023) - there is significant development at scale of “bio-inspired” materials and processes which combine some aspects of a biological basis into an otherwise relatively more familiar engineering approach. Self-cleaning, self-healing, and lightweight structural or engineering materials can be highlighted as topics that have benefited from bio-inspiration; structural colour and other optical properties are also interesting in initial practical demonstration applications, and aspects such as control of moisture movement offer interesting opportunities at the interface of architecture, building physics, and structural engineering (Chayaamor-Heil et al., 2023).
Sustainability in materials production has also moved over the past decades from a “nice-to-have” characteristic to a property of existential importance. Put simply, materials that are not able to be produced sustainably are progressively being removed from the market, and recycling rates (considering secondary materials as inputs in production of e.g., steel, asphalt, and cement/concrete) are being driven both by financial and political levers. It is of course necessary for scientists and engineers to carefully assess whether reuse, recycling, down-cycling, or other end-of-life avenues are the most efficient and effective for each class of materials, considering the Waste Hierarchy principle as well as questions of contamination, (re)processing costs (energy, resources, and financial), and suitability for purpose. Many of these questions remain under-explored in a large percentage of academic published studies of “recycled” materials, and this can be highlighted as another opportunity for future consideration.
The engineering of multifunctionality in structural materials, the ability to manipulate interfaces in materials to enhance composite properties, and the increasing availability (and affordability) of 2-dimensional materials, are to a significant extent interconnected and also provide scope for future advances. As an example of this which appears potentially mundane but actually offers enormous scope for materials scientists to enhance material performance in service: it could be argued that reinforced concrete is a massively multi-phase composite, comprising several residual unreacted clinker phases and supplementary cementitious materials as particulate inclusions (whose surfaces may be modified by polymeric additives), a complex array of hydration products as a continuous phase (but which also contains both liquid-filled and vapour-filled pores across several orders of magnitude in size), fine and coarse aggregates, and steel bars and/or fibre reinforcing elements, which may themselves have associated passive surface films or manufactured coatings. Each of the interfaces involved in this material, spanning length scales from Ångstroms up to centimetres, can potentially be analysed and engineered to drive improvements in material properties; even if the bulk behaviour of each phase is relatively well characterised (which is increasingly, but not completely, the case at present), the application of interfacial science and engineering to the advancement of structural material performance is evidently a major opportunity.
Concluding remarks
In this document, I have attempted to provide insights and predictions related to trends in materials research as a discipline, and more specifically related to the Structural Materials section of Frontiers in Materials as the journal celebrates 10 years of contributions to the technical community. It is clear that AI, sustainable materials, and the design and selection of materials according to optimised functionality (including durability under challenging conditions), will remain at the forefront of the minds of researchers in the near-term. The selection and use of structural materials will continue to evolve along with the needs and capabilities of society and its research community, and materials that better serve the needs of a resource-constrained and emissions-constrained future world will continue to be developed and described. The clearly stated intention of the journal team is that we plan to play a significant role in continuing to disseminate high quality and relevant information about structural materials to the relevant stakeholders within and beyond the academic community, publishing open science and enabling access by all.
Data availability statement
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.
Author contributions
JP: Formal Analysis, Writing – review and editing, Writing – original draft, Visualization, Investigation, Conceptualization.
Funding
The author(s) declared that financial support was received for this work and/or its publication. Open access funding by PSI - Paul Scherrer Institute.
Acknowledgements
I particularly thank Emily Young of Frontiers, who has, throughout the life of the Structural Materials section, provided excellent input, insight and support in editorial and strategic matters for the section and the journal, which has enabled us to maintain a high technical standard in the papers published, and to make a useful and high-quality contribution to the academic literature (as summarised in this article).
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author JP declared that they were an editorial board member of Frontiers at the time of submission. This had no impact on the peer review process and the final decision.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. Generative AI was used to generate Figure 2, as discussed in the text.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Antypov, D., Vasylenko, A., Collins, C. M., Daniels, L. M., Darling, G. R., Dyer, M. S., et al. (2025). Discovery of crystalline inorganic solids in the digital age. Accounts Chem. Res. 58 (9), 1355–1365. doi:10.1021/acs.accounts.4c00694
Armistead, S. J., Mikofsky, R. A., and Srubar, W. V., III (2023). Toward biomimetic and living earth materials. Matter 6 (12), 4124–4127. doi:10.1016/j.matt.2023.11.003
Beygi, R., Galvão, I., Akhavan-Safar, A., Pouraliakbar, H., Fallah, V., and da Silva, L. F. M. (2023). Effect of alloying elements on intermetallic formation during friction stir welding of dissimilar metals: a critical review on aluminum/steel. Metals 13 (4), 768. doi:10.3390/met13040768
Boiger, R., Xi, B., Miron, G.-D., Bonvin, M., Provis, J. L., Churakov, S. V., et al. (2025). Machine learning-accelerated discovery of green cement recipes. Mater. Struct. 58 (5), 173. doi:10.1617/s11527-025-02684-z
Buffa, G., Fratini, L., La Commare, U., Römisch, D., Wiesenmayer, S., Wituschek, S., et al. (2022). Joining by forming technologies: current solutions and future trends. Int. J. Material Form. 15 (3), 27. doi:10.1007/s12289-022-01674-8
Busby, J. T., Nanstad, R. K., Stoller, R. E., Feng, Z., and Naus, D. J. (2008). Materials degradation in light water reactors: life after 60 (ORNL/TM-2008/170). Oak Ridge TN: Oak Ridge National Laboratory. (ORNL).
Campelo, F., and Aranha, C. (2023). Lessons from the evolutionary computation bestiary. Artif. Life 29 (4), 421–432. doi:10.1162/artl_a_00402
Candela, A., Sandrini, G., Gadola, M., Chindamo, D., and Magri, P. (2024). Lightweighting in the automotive industry as a measure for energy efficiency: review of the main materials and methods. Heliyon 10 (8), e29728. doi:10.1016/j.heliyon.2024.e29728
Castro-Alonso, M. J., Montañez-Hernandez, L. E., Sanchez-Muñoz, M. A., Macias Franco, M. R., Narayanasamy, R., and Balagurusamy, N. (2019). Microbially induced calcium carbonate precipitation (MICP) and its potential in bioconcrete: microbiological and molecular concepts. Front. Mater. 6, 126. doi:10.3389/fmats.2019.00126
Chayaamor-Heil, N., Perricone, V., Gruber, P., and Guéna, F. (2023). Bioinspired, biobased and living material designs: a review of recent research in architecture and construction. Bioinspiration and Biomimetics 18 (4), 041001. doi:10.1088/1748-3190/acd82e
Cheetham, A. K., and Seshadri, R. (2024). Artificial intelligence driving materials discovery? Perspective on the article: scaling deep learning for materials discovery. Chem. Mater. 36 (8), 3490–3495. doi:10.1021/acs.chemmater.4c00643
Churakov, S. V., Claret, F., Idiart, A., Jacques, D., Govaerts, J., Kolditz, O., et al. (2024). Position paper on high fidelity simulations for coupled processes, multi-physics and chemistry in geological disposal of nuclear waste. Environ. Earth Sci. 83 (17), 521. doi:10.1007/s12665-024-11832-7
Criado, M., Ke, X., Provis, J. L., Bernal, S. A., Savastano, H., Jr, Fiorelli, J., et al. (2017). “Alternative inorganic binders based on alkali-activated metallurgical slags,” in Sustainable and nonconventional construction materials using inorganic bonded fiber composites. UK: Woodhead, 185–220.
Drace, Z., Ojovan, M. I., and Samanta, S. K. (2022). Challenges in planning of integrated nuclear waste management. Sustainability 14 (21), 14204. doi:10.3390/su142114204
Frankel, G. S., Vienna, J. D., Lian, J., Guo, X., Gin, S., Kim, S. H., et al. (2021). Recent advances in corrosion science applicable to disposal of high-level nuclear waste. Chem. Rev. 121 (20), 12327–12383. doi:10.1021/acs.chemrev.0c00990
Gao, J., Jiang, S., Zhang, H., Huang, Y., Guan, D., Xu, Y., et al. (2021). Facile route to bulk ultrafine-grain steels for high strength and ductility. Nature 590 (7845), 262–267. doi:10.1038/s41586-021-03246-3
Garboczi, E. J., and Lura, P. (2020). Local elastic moduli of simple random composites computed at different length scales. Mater. Struct. 53 (6), 150. doi:10.1617/s11527-020-01592-8
George, E. P., Raabe, D., and Ritchie, R. O. (2019). High-entropy alloys. Nat. Rev. Mater. 4 (8), 515–534. doi:10.1038/s41578-019-0121-4
Ghafarollahi, A., and Buehler, M. J. (2025). Automating alloy design and discovery with physics-aware multimodal multiagent AI. Proc. Natl. Acad. Sci. 122 (4), e2414074122. doi:10.1073/pnas.2414074122
Guo, S., Li, X., Dong, K., Gu, J., Zhou, Q., and Wang, K. (2022). Microstructure and shear property of high-temperature brazing of TiAl alloy and 316L stainless steel with a Ni-based amorphous filler. Front. Mater. 9, #1098397. doi:10.3389/fmats.2022.1098397
Hooton, R. D., and Riding, K. (2025). Type IL cement use in precast, prestressed concrete. PCI J. 70, 23–36. doi:10.15554/pcij70.2-04
Ilgın, H. E. (2024). High-rise residential timber buildings: emerging architectural and structural design trends. Buildings 14 (1), 25. doi:10.3390/buildings14010025
International Energy Agency (2023). Production of iron by route and scrap share of metallic inputs in the net zero scenario. Paris, France: IEA, 2018–2030. Available online at: https://www.iea.org/data-and-statistics/charts/production-of-iron-by-route-and-scrap-share-of-metallic-inputs-in-the-net-zero-scenario-2018-2030.
Jia, H., Guo, Z., Wu, R., and Zhang, S. (2022). Grain refinement of Mg-14Li alloy by Sn addition and large strain rolling. Front. Mater. 9, 1050764. doi:10.3389/fmats.2022.1050764
Juenger, M. C. G., Snellings, R., and Bernal, S. A. (2019). Supplementary cementitious materials: new sources, characterization, and performance insights. Cem. Concr. Res. 122, 257–273. doi:10.1016/j.cemconres.2019.05.008
Kanavaris, F., Vieira, M., Bishnoi, S., Zhao, Z., Wilson, W., Tagnit Hamou, A., et al. (2023). Standardisation of low clinker cements containing calcined clay and limestone: a review by RILEM TC-282 CCL. Mater. Struct. 56 (9), 169. doi:10.1617/s11527-023-02257-y
Kapoor, S., Cantrell, E. M., Peng, K., Pham, T. H., Bail, C. A., Gundersen, O. E., et al. (2024). REFORMS: consensus-based recommendations for machine-learning-based science. Sci. Adv. 10 (18), eadk3452. doi:10.1126/sciadv.adk3452
Khan, A. H., Firoozi, A. A., Alam, P., Agwa, I. S., Rahmawati, C., and Hasan, M. A. (2025). Editorial: supplementary cementitious materials: recent developments, performance insights and potential applicability. Front. Mater. 12, 1608358. doi:10.3389/fmats.2025.1608358
Li, R., Xie, L., Wang, W. Y., Liaw, P. K., and Zhang, Y. (2020). High-throughput calculations for high-entropy alloys: a brief review. Front. Mater. 7, 290. doi:10.3389/fmats.2020.00290
Li, Z., Yoon, J., Zhang, R., Rajabipour, F., Srubar, W. V., III, Dabo, I., et al. (2022). Machine learning in concrete science: applications, challenges, and best practices. Npj Comput. Mater. 8 (1), 127. doi:10.1038/s41524-022-00810-x
Ma, E., and Liu, C. (2024). Chemical inhomogeneities in high-entropy alloys help mitigate the strength-ductility trade-off. Prog. Mater. Sci. 143, #101252. doi:10.1016/j.pmatsci.2024.101252
Ma, B., Provis, J. L., Wang, D., and Kosakowski, G. (2024). The essential role of cement-based materials in a radioactive waste repository. Npj Mater. Sustain. 2 (1), 21. doi:10.1038/s44296-024-00025-9
Malerba, L., Al Mazouzi, A., Bertolus, M., Cologna, M., Efsing, P., Jianu, A., et al. (2022). Materials for sustainable nuclear energy: a European strategic research and innovation agenda for all reactor generations. Energies 15 (5), 1845. doi:10.3390/en15051845
Moser, C., Stauffacher, M., Krütli, P., and Scholz, R. W. (2012). The crucial role of nomothetic and idiographic conceptions of time: interdisciplinary collaboration in nuclear waste management. Risk Anal. 32 (1), 138–154. doi:10.1111/j.1539-6924.2011.01639.x
O'Toole, G. (2013). “Quote investigator - Quote origin: it’s difficult to make predictions, especially about the future. Available online at: https://quoteinvestigator.com/2013/10/20/no-predict/.”Retrieved .”Retrieved 13/November/2024.
Poirot-Delpech, S., and Raineau, L. (2016). Nuclear waste facing the test of time: the case of the French deep geological repository project. Sci. Eng. Ethics 22 (6), 1813–1830. doi:10.1007/s11948-015-9739-9
Provis, J. L. (2015). Grand challenges in structural materials. Front. Mater. 2–31. doi:10.3389/fmats.2015.00031
Provis, J. L., Bernal, S. A., and Zhang, Z. (2024). The decarbonization of construction — how can alkali-activated materials contribute? Engineering 37 (6), 18–21. doi:10.1016/j.eng.2023.09.014
Quadling, A., Lee, W. E., and Astbury, J. (2022). Materials challenges for successful roll-out of commercial fusion reactors. J. Phys. Energy 4 (3), 030401. doi:10.1088/2515-7655/ac73b2
Rossi, L., Miranda de Lima, L., Sun, Y., Dehn, F., Provis, J., Ye, G., et al. (2023). Future perspectives for alkali-activated materials: from existing standards to structural applications. RILEM Tech. Lett. 7, 159–177. doi:10.21809/rilemtechlett.2022.160
Sandak, A., and Butina Ogorelec, K. (2023). Bioinspired building materials - lessons from nature. Front. Mater. 10, 1283163. doi:10.3389/fmats.2023.1283163
Stoll, A., and Benner, P. (2021). Machine learning for material characterization with an application for predicting mechanical properties. Mittl. 44 (1), e202100003. doi:10.1002/gamm.202100003
Tian, F. (2017). A review of solid-solution models of high-entropy alloys based on ab initio calculations. Front. Mater. 4, 36. doi:10.3389/fmats.2017.00036
Tupenaite, L., Zilenaite, V., Kanapeckiene, L., Gecys, T., and Geipele, I. (2021). Sustainability assessment of modern high-rise timber buildings. Sustainability 13 (16), 8719. doi:10.3390/su13168719
United Nations Environment Programme (2024). Adaptation gap report 2024: come hell and high water — as fires and floods hit the poor hardest, it is time for the world to step up adaptation actions. Nairobi: UNEP.
United Nations Environment Programme (2025). “Not just another brick in the wall: the solutions exist - scaling them will build on progress and cut emissions fast,” in Global status report for buildings and construction 2024/2025. Paris: UNEP.
Wagner, N., and Rondinelli, J. M. (2016). Theory-guided machine learning in materials science. Front. Mater. 3 (28). doi:10.3389/fmats.2016.00028
Wang, Y.-S., Alrefaei, Y., and Dai, J.-G. (2019). Silico-aluminophosphate and alkali-aluminosilicate geopolymers: a comparative review. Front. Mater. 6, 106. doi:10.3389/fmats.2019.00106
Wang, P., Qiao, Y., Qi, W., Du, S., Liu, Z., Meng, F., et al. (2021). Preparation and properties study of Cr on FeCrAl cladding materials. Front. Mater. 8, 621086. doi:10.3389/fmats.2021.621086
Yeh, I. C. (1998b). Modeling of strength of high-performance concrete using artificial neural networks. Cem. Concr. Res. 28 (12), 1797–1808. doi:10.1016/s0008-8846(98)00165-3
Zhang, C., Bos, C., Sandfeld, S., and Schwaiger, R. (2024). Unsupervised learning of nanoindentation data to infer microstructural details of complex materials. Front. Mater. 11, 1440608. doi:10.3389/fmats.2024.1440608
Keywords: artificial intelligence, future developments, retrospective, structural materials, trends in research
Citation: Provis JL (2026) Structural materials in Frontiers in Materials, 10 years on. Front. Mater. 12:1713993. doi: 10.3389/fmats.2025.1713993
Received: 26 September 2025; Accepted: 17 December 2025;
Published: 07 January 2026.
Edited by:
Nicola Maria Pugno, University of Trento, ItalyReviewed by:
Blas Uberuaga, Los Alamos National Laboratory (DOE), United StatesCopyright © 2026 Provis. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: John L. Provis, am9obi5wcm92aXNAcHNpLmNo