In a newly published scientific paper in the journal Nature Communications on December 9, 2025, an international team led by researchers from MIT and collaborators from the Archaeological Park of Pompeii presents one of the most complete insights into how the Romans truly built – not according to scant archaeological indications or later retellings of ancient texts, but based on a “frozen” construction site stopped by the eruption of Vesuvius in 79 AD. The house in Region IX (Domus IX 10, 1) was discovered at the moment of work: walls in various stages of execution, piles of dry raw materials, ready-mixed mixtures, tools, and – crucial for materials science – mortars in which traces of mixing techniques and reactions that gave the mortar exceptional durability are written. This find not only complements Vitruvius's descriptions but also tangibly confirms that Roman builders applied so-called “hot mixing” with quicklime, a procedure that, besides the speed of installation, had another consequence: the self-healing of cracks decades and centuries after construction.
Construction site caught in the moment before disaster
Unlike most archaeological sites where fragments are found – for example, separate lime kilns, scattered broken amphorae with traces of plaster, or isolated lime slaking pits – here we are dealing with a complete workspace. Two rooms (designated as rooms 2 and 14) contained clearly separated piles of raw materials: dry pozzolanic ash and pyroclastic material, pieces of tephritic and other volcanic tuffs (caementa), as well as mixed dry composites ready for the addition of water immediately before installation. Nearby, tools and measuring weights were also cataloged – traces of everyday life on the construction site – while the walls, some already closed, others still in the “formwork” of the first century, offered an ideal cross-section through the layers of a Roman wall.
Researchers sampled three groups of materials: (1) previously executed, fully hardened walls; (2) walls in the process of masonry; and (3) immediate dry mixtures without water. This typology enabled a correlation between the composition of raw materials, the mixing process on-site, and the resulting microstructure of the hardened mortar. Thanks to such a “triangle of evidence,” the paper documents the complete chain of the Roman working process, from the logistics of storage and sieving to the chemistry that enables the mortar's later reactivity.
“Hot mixing” and lime clasts: a recipe for longevity
The key to the above-average durability of Roman concrete (opus caementicium) is not just in the use of pozzolan and volcanic aggregate. Analyses from Pompeii show that the workers dry-premixed quicklime (CaO) with the pozzolanic component immediately before installation, and then added water. That sequence triggers a strong exothermic hydration of the lime – locally raising the mixture temperature up to several hundred degrees Celsius – which, along with rapid cooling and specific humidity conditions in the pores, leads to the conservation of white “islands” of lime, so-called lime clasts. In traditional interpretation, they were considered flaws or “relics” of insufficiently homogeneous mixing; today, on the contrary, they are recognized as a functional core that remains partially reactive for decades.
When water later penetrates into microcracks and pores – via precipitation, capillary moisture, and even through mechanical vibrations – it dissolves calcium from these clasts and mobilizes it towards the crack. At the boundary with pozzolanic grains, new calcium-aluminate-silicate hydrate phases (C-A-S-H) precipitate, and at the same time, polymorphs of calcium carbonate are formed, from amorphous to crystalline forms like calcite and aragonite. In Pompeii, so-called “reaction rims” around volcanic aggregates – zones of matrix/aggregate interface remodeling – have been precisely documented as a mineral map of ion movement through time. This is a completely different narrative from assumptions that Roman mortars were chemically “dead” after setting.
Reconstruction of the Roman workflow
The archaeological picture and analytical chemistry coincide in a series of steps: dry preparation of the mixture (lime + pozzolan), then addition of water immediately before installation, followed by installation in layers with larger pieces of stone or broken ceramics (caementa) between the formwork slats. Due to hot hydration, part of the lime does not manage to fully transition into Ca(OH)2, but remains as a core around which secondary reactions will later take place. In an environment abundant in silicate and aluminate phases from volcanic ashes and tuffs, that “duration of chemistry” is actually the motto that explains the multi-century resilience of walls in arches, vaults, and domes.
In Pompeii, additionally, masonry tools and vessels were found suggesting that slaking lime in large pits was not necessarily the rule. Vitruvius in De architectura describes the practice of “slaked” lime (calx restincta), but chronology and local tradition could change, especially in post-earthquake reconstruction after the earthquake of 62 AD. In Domus IX 10, 1, we find primarily logistics adapted to speed: dry piles of standardized mixtures, ready for “activation” with water and immediate installation.
What microscopes and spectrometers say
The team analysis encompassed multiscale methods: optical and electron microscopy, X-ray diffraction, and spectroscopies for phase mapping. Special attention was paid to the boundary between glassy (vitric) tuffs and the mortar matrix. Precisely there, concentric “rims” enriched with calcium and silicates were registered – proof that mobile calcium from lime clasts reached the aggregates and “did the job” of secondary binding chemistry there. In certain zones, CaCO3 polymorphs (calcite, aragonite) were also recorded, which often “heal” microcracks by filling them with new precipitates. In older literature, this was sometimes attributed exclusively to modern-era infiltrations; the comparison of three groups of samples in this paper refutes that and chronologically ties the process to the early and middle phase of the mortar's “life.”
From the chemical side, this is a confirmation of the concept of “self-healing” which was demonstrated in 2023 in modern analogies on concrete samples formulated according to the Roman model. In experiments where the mortar was intentionally damaged, versions with the addition of quicklime and “hot mixing” showed crack closure and restoration of impermeability, while control samples with classically slaked lime remained with open cracks. The Pompeii samples give archaeological-material proof that such functionality is not a laboratory trick, but a real feature of the original technology.
Implications: heritage conservation and low-carbon concrete
Why is this important today? First, the conservation aspect: restoration plasters and injection mixtures that mimic Roman chemistry could be more compatible with originals, reducing the risk of harmful interactions and extending the life of heritage. Second, the climate aspect: Portland cement concrete is responsible for a significant part of global CO2 emissions. If part of the functionality – for example, microcrack self-healing and long-term post-pozzolanic reactivity – can be achieved with a smaller amount of clinker and selected pozzolans, space opens up for lower-carbon mixtures with a longer service life. It is not about a romantic return to “Roman concrete,” but about the translation of principles into standardized systems adapted to today's regulations.
Pompeii offers more than one “case study” in this regard. In recent years, the Park has gradually opened new sites and tours through active excavations in Region IX, enabling the documentation of contexts that were previously out of reach. Along with this paper, parallel findings – for example, larger private thermal complexes – remind us that construction techniques were diverse and socially “networked” with daily life, political representations, and the economic interests of the city's elites.
What changes in the textbooks of construction history
The biggest shift concerns the understanding of the source of our knowledge. Vitruvius and Pliny are key, but they are not an encyclopedia of all local practices. Roman construction was an "ecosystem" – the speed of reconstruction after the earthquake of 62 AD, the availability of raw materials, work rhythms, and the logistics of reusing ceramics and stone all led toward sustainable and pragmatic choices. “Hot mixing” in that landscape is more than an exotic detail: it is an operational compromise between productivity, durability, and the technology of that time for dealing with moisture and cracks in the wall. Therefore, it is wrong to interpret it as a “mistake” or “poor homogenization”; on the contrary, in Pompeii we see that it is a deliberate strategy.
Archaeology of the process, not just the product
The Pompeian construction site allows us to read archaeology as “process forensics.” Piles of materials are arranged so that dry mixtures stand closest to the masonry site – lime already in contact with ash, but without water – while larger pieces of stone and broken ceramics are arranged for quick insertion. Once water is added, the chemical clock starts ticking: the mix heats up, viscosity changes, and the mortar gains workability suitable for installation in layers. Due to that heat and partial dehydration at the micro-level, part of the lime remains “protected” from complete dissolution and turns into a long-term reservoir of calcium. When the wall meets rain, condensation, or capillarity, that reservoir activates and “feeds” new phases that support bridges across cracks and micropores.
Precisely these details – the sequence, proximity of piles, arrangement of tools – often missed in older interpretations. The result was often an anachronism: lime slaking as a universal rule, neglecting the time pressures of the construction site, or reducing the pozzolanic component to “gray dust.” Pompeii allows us to correct that scheme and replace it with a dynamic picture in which logistics, chemistry, and practice unite into a unique workflow.
Reaction rims as a “diagnostic fingerprint”
For materials scientists, perhaps the most intriguing are the so-called reaction rims around fragments of volcanic aggregate. These are zones where the mortar matrix has undergone secondary mineralization over time – from calcium-enriched solutions arriving from lime clasts. In Pompeii, these rims are multi-layered: in some cases amorphous phases transition into crystalline ones, and spatial variations point to cyclic wetting and drying conditions. These “fingerprints” allow macro-environmental conditions, and even space usage regimes, to be read from the microstructure. For example, walls more exposed to precipitation moisture may develop “thicker” carbonate zones than those in protected spaces.
Such “wall geology” is a first-class tool for conservators as well. Instead of generic injections, it is possible to design mixtures that targetedly replicate the chemical potential of the original. In that sense, laboratories developing “Rome-inspired” concretes have already been showing for several years that with a combination of quicklime and carefully selected pozzolan, measurable crack closure can be achieved without external adhesive additives. The Pompeii findings give this direction historical verification.
The bigger picture: from the Pantheon to 21st-century urban infrastructure
Often, the popular explanation of Roman concrete is reduced to a “secret ingredient” and mentioning the Pantheon. New data show that the “secret” is not in one substance, but in a combination of procedures and in the architecture of the process. Romans recycled to exhaustion: broken ceramics and stone returned to the wall; dry mixtures were prepared in advance; and water – the trigger of chemistry – was added when logistically most rational. On the other hand, modern construction often approaches its concrete as a “mono-product” with a short designed life and high initial carbon cost. If anything, Pompeii suggests that longevity is not an accident, but a designed consequence of the workflow.
This does not mean that the “Roman recipe” should be uncritically given a place in norms. Standards of safety, resistance to freezing, salt and sulfate attacks, as well as compatibility with reinforcement, condition modern systems. But the principle of “chemistry that remains alive,” paired with an appropriate choice of pozzolans (e.g., silicate ashes, vitric tuffs), can reduce the need for high-clinker binders and give concrete internal mechanisms of “forgiving” microcracks, which is particularly important for infrastructure exposed to load cycles and environmental stresses.
Contexts and chronology: 62 – 79 AD
The paper draws attention to the seismic event of 62 AD which triggered a wave of reconstruction in Pompeii. Many houses and public buildings were reconstructed in the years up to the eruption, which explains why we find construction sites “in real time.” That chronology is also essential for interpreting Vitruvius: his descriptions are a valuable source, but they describe the practice of the late Republic and early Empire, not necessarily what – due to time pressure – was being done in Pompeii immediately before 79 AD. In that sense, Domus IX 10, 1 becomes a study of local, time-specific know-how, and not a Platonistic “essence of the Roman wall.”
Open questions for the next campaigns
How widespread was “hot mixing” outside Campania? Are shifts in microstructure dependent on the mineralogy of local pozzolans? How did recipes for walls, floors, and water features differ? And finally: can modern “Rome-inspired” systems satisfy design requirements for reinforced concrete without compromise in alkali-silica reaction and compatibility with steel reinforcement? These Pompeii findings give a roadmap – but also a task – to interdisciplinary teams combining archaeology, materials science, and civil engineering.
What this means for the history of technology
The most valuable thing in this discovery is the shift from “recipe” to “construction site ecology.” We see how the arrangement of materials and tools, the sequence of steps, the rhythm of mixing and installation, and the chemical echoes of those decisions form one system. That system is robust, since it relies on long-term reactivity and on secondary reactions activated by the environment, and not on the rigid ideal of a perfectly homogeneous binder. In that sense, Pompeii teaches us that longevity is an emergent feature of the system, and not the result of a magic ingredient. Therefore, this paper is not just news about one house, but a lesson on how technologies become sustainable when they synchronize logistics and chemistry.
For experts in conservation and engineering, the benefit is most practical: now there is a reference, dated, and contextualized set of samples with which museum, laboratory, and field methods can be calibrated. Furthermore, the public character of the research means that it is possible to compare everything – from the arrangement of raw material piles to the mineralogy of reaction zones – with other sites and historical periods. This opens the field of “comparative process archaeology” in which it will be verified which dimensions are universal and which are local adaptations. Ultimately, from that matrix, a new class of mixtures for remediation and for new construction can arise, designed to last at least as long as the Roman walls in Campania lasted.
Additional sources and field availability
The Pompeii Park has in recent years enabled tours of active works in Region IX, offering experts and the public insight into excavation, protection, and interpretation procedures. For those who want to understand the science behind “hot mixing” more deeply, papers that experimentally confirmed self-healing in Rome-inspired mixtures in 2023 are recommended, as well as concise overviews in journals dedicated to materials. Comparison of those papers with the Pompeian find from 2025 shows how a hypothesis becomes a confirmed model when the laboratory and the archaeological site speak the same language – the language of reaction rims, lime clasts, and phase transitions in carbonates.
At that point, it is understandable why news of the paper was also picked up by popular science media highlighting the potential for safer and longer-lasting infrastructure. But for the profession, it is more important that the discussion can now be conducted on the basis of microscopically and geochemically precisely documented facts, and not just according to tradition or the authority of ancient text. And that is perhaps the greatest value of Domus IX 10, 1: from it we learn not only what the Romans did, but how they thought about time, material, and duration.
Technical notes for practice
For conservation teams and engineers thinking about the transfer of principles into practice, three points are key: (1) mixing sequence – dry premixing of lime and pozzolan, and controlled addition of water; (2) pozzolan mineralogy – glassy, silicate fractions that favor long-term reactivity; (3) moisture management – allowing the system to truly “breathe” and for water to trigger secondary reactions without permanent saturation that would degrade the structure. In laboratory analogies, this means careful dosing, heat control, and comparisons with reference samples from Pompeii; in the field, this means designing details that introduce a “smart” relationship with water, instead of a fight for complete hermeticity.
This view does not oppose "Roman" and "modern," but proposes a bridge: from an archaeologically verified process toward contemporary norms. And although the question of how resistant to chloride ions or abrasion a modern system relying on similar principles can be will remain open for a long time, the fact that today we have a “classroom” in Domus IX 10, 1 makes that bridge stronger. Ultimately, the biggest gain is that we can finally move away from the binary dilemma of “secret ingredient vs. myth” and deal with what made the Romans masters of durability: the choreography of materials and time.
Why Domus IX 10, 1 is a unique case
Even in archaeologically rich Pompeii, we rarely find a construction site with such a complete context: neat piles of dry mixtures, tools within reach, and walls in a series of phases. That combination enabled an unusually clear attribution: where what was mixed, when water was added, how the installation proceeded, and where the “signatures” of reactions remained. Museum specimens from other sites preserve products; Domus IX 10, 1 preserves the process. Because of this, this paper will long remain a reference point in the literature on Roman construction, and give modern laboratories a stable platform for comparisons and validation of new, more sustainable concrete systems.
In the following seasons, additional studies are expected that will examine variations in the composition of dry mixtures, the relationship to local ashes, and the modulation of “hot mixing” depending on the purpose of the wall. In parallel, engineering groups are already testing how to integrate self-healing mechanisms into standard reinforced concrete systems without compromise regarding reinforcement corrosion. Thus, Pompeii becomes a living laboratory – archaeological and technological – where the past and future meet in one, surprisingly practical, lesson on durability.
For field visits and additional professional materials, we recommend the official publications of the Pompeii Park, as well as review texts on the self-healing of Rome-inspired composites in relevant scientific journals. Whoever wants to go deeper into microstructural details should pay attention to maps of reaction zones around vitric tuffs and to the role of lime clasts in initiating secondary carbonation – these are "maps" by which it is possible to read the history of a wall just as precisely as geographical maps read a landscape.
In the end, Domus IX 10, 1 shows that great discoveries are not hidden only in monumental buildings, but also in humble rooms full of piles of dry sand, ash, and lime. In them hides the logic of construction that enabled the Romans to build for centuries with “living chemistry.”
The complete paper in the journal Nature Communications (December 9, 2025) offers detailed methods, microscopy displays, and phase mapping, plus a rich supplement of images from the location, valuable for both archaeologists and materials scientists.