Roman concrete self-healed with quicklime, MIT study finds

Roman concrete self-healed thanks to quicklime

Why does a two-millennia-old aqueduct still stand while a fifteen-year-old building crumbles? The answer, according to a study published in Science Advances by a team from MIT, Harvard, and several Italian and Swiss labs, lies in tiny lime clasts embedded in the mixture. This was no manufacturing flaw. It was the mechanism that allowed the material to seal its own cracks when water entered.

The finding overturns the previously accepted explanation. The durability of Roman concrete was attributed to pozzolana, the volcanic ash from the Bay of Naples that, reacting with calcium hydroxide, formed a hydraulic compound similar to that generated when modern cement clinker hydrates. True, but insufficient. Spectroscopic analysis of the samples revealed that the lime clasts had formed at extreme temperatures, consistent with the use of quicklime rather than slaked lime. During this hot mixing, the clasts develop a fragile and reactive nanoparticle architecture: a calcium reserve ready to fracture and seal any fissure.

Why modern reinforced concrete lasts less

The comparison with current construction is inevitable. Modern concrete is reinforced with metal rebar and works primarily under tension. Here lies its Achilles' heel: as soon as the concrete cracks, the steel is exposed, rusts, and the structure degrades from within. Roman concrete contained no steel. It consolidated under compression, using stone and mortar, and the mixture itself closed its wounds. The lifespan of steel encased in concrete is, comparatively, ridiculous.

The case of the Pantheon of Agrippa is cited as proof: two millennia with not a single sustancia ilegal. In contrast, Valencia's City of Arts and Sciences, designed by Calatrava, has accumulated problems fifteen years after its inauguration. The irony comes from the calendar, not the author.

Can we replicate the Roman concrete formula today?

Technically, the recipe is on the table: quicklime, pozzolana, aggregates, and hot mixing. The question is whether it interests anyone. The cement industry moves huge volumes with standardized processes and regulations designed for reinforced concrete. A material that self-repairs over centuries does not fit a model that thrives on new construction, maintenance, and periodic replacement. Some argue the formula was never lost: it simply was not wanted back. The argument has logic, though it also carries unverified suspicion.

What is documented is that the Romans did not improvise. They trinc recipes optimized over centuries and invested effort at every step. The question posed by one of the researchers is pertinent: if they cared so much about the mixture, why would they leave lime unslaked by carelessness? It was not carelessness. It was design.

The bridge that still bears trucks

Among the circulating examples, one is particularly illustrative: a Roman bridge in the Ebro reservoir area, near Reinosa, used for vehicle passage until relatively recently. Next to it, a modern bridge was built to divert traffic. The modern one collapsed in a week. The Roman one remains. The anecdote, repeated in conversation, summarizes better than any report the distance between two ways of understanding engineering.

Medieval bridges also appear, often mistakenly attributed to the Romans. They are narrower, rougher, and less practical: designed for an ox cart, not a legion with full equipment. The comparison does not do justice to the Middle Ages, but it is not fair either: the Romans had an empire financing public works on a continental scale.

---

With these elements, the question is not whether Roman concrete was better. It was. The question is how long it will take someone to industrialize the formula. And the answer, given the sector's real interest, probably will not come in this decade.