Pliny the Elder claimed Roman harbor concrete got "every day stronger" under the sea. Two thousand years later, scientists drilled into Roman harbors and found he had a point: seawater was still growing new minerals inside it.

• ✓doi.org
• ✓news.mit.edu
• ✓nature.com
• ✓news.mit.edu
• ✓news.mit.edu
• ✓newscenter.lbl.gov
• ✓doi.org

Almost two thousand years ago, Pliny the Elder made a bold claim about Roman harbor concrete. Under the sea, he wrote, it became "a single stone mass, impregnable to the waves and every day stronger." It sounds like ancient bragging. But when scientists drilled into Roman harbors, they found the sea really had been changing the concrete, growing new minerals inside it. Other Roman walls held white lumps that can help seal cracks, and a version of that idea has been used on a Swiss motorway. A building site buried by Vesuvius showed how the Romans mixed it. So what did they get right, and why don't we build that way? The Roman architect Vitruvius described the key ingredient: a powder found in the country around Mount Vesuvius. Mixed with lime and rubble, he wrote, it set hard even in piers built in the sea. That powder was volcanic ash. Today it's called pozzolana, after the town of Pozzuoli on the Bay of Naples. The Romans trusted it enough to ship it across the Mediterranean. For King Herod's harbor at Caesarea, in what is now Israel, builders used volcanic ash brought from the Bay of Naples, about twelve hundred miles away, and poured breakwaters out in the open sea. Between 2002 and 2009, a research project drilled cores out of ancient harbor structures around the Mediterranean. These cylinders let scientists see what had happened inside the concrete, beneath the weathered surface. In 2017, a team led by geologist Marie Jackson of the University of Utah published what they found in cores from Italian harbors. Seawater filtering through the concrete had dissolved parts of the volcanic ash. New minerals grew in the spaces left behind. One was a rare mineral called tobermorite. Its tiny plates and fibers helped make the concrete more resistant to fracture. Now, Pliny didn't know why. But he had noticed something real: long after it was poured, the sea was still working on the concrete. So why don't we just build like the Romans? Well, because modern concrete is built for a different job. Portland cement, the binder in most concrete today, gets strong within days, so crews can strip the forms and pour the next floor. Roman marine concrete gained strength slowly, and its compressive strength can be lower than typical modern concrete. And we put steel inside ours. Concrete handles squeezing well, but it cracks when it's pulled. Steel bars carry the pull. That's what makes long bridges and tall towers possible. But that's also the catch. Salt from seawater or winter roads seeps in and reaches the steel. The steel rusts, the rust swells, and the concrete cracks from the inside. The Roman harbor walls had no steel to rust. That doesn't make their recipe better for a bridge. It means they solved a different problem: not getting strong fast, but lasting. And remember, we're looking at the survivors. The Roman structures that failed aren't standing beside them for comparison. The Romans pushed concrete even further on land, and that's where the strangest clue turned up. The Pantheon in Rome has stood for about nineteen hundred years. Its dome, forty-three meters across, is still the largest unreinforced concrete dome in the world. The builders made it lighter as it rose, switching to lighter volcanic stone near the top. But every structure cracks eventually. So another team asked: could Roman concrete repair some of that damage itself? A team at MIT studied mortar from an ancient city wall near Rome. Inside were small white lumps. They had often been treated as signs of sloppy mixing. The researchers suspected they were something else: a reserve of material for repairs. The lumps were rich in calcium. When a crack reached one, water could dissolve some of that calcium. New minerals could then form inside the crack and seal it. An explanation is one thing. But could they actually make it happen? For a study published in 2023, the team made modern mixtures inspired by the Roman recipe, using quicklime, a highly reactive form of lime. They deliberately cracked the samples and ran water through them. The cracks were about half a millimeter wide. That's two hundredths of an inch. Within one to three weeks, water flow through the quicklime samples fell to almost zero as the cracks sealed. Samples made without quicklime kept leaking. These were new laboratory samples, not ancient blocks. But they showed an ancient design principle working in a modern mixture. One question remained: did Roman builders really mix it this way? The answer was buried at Pompeii. When Vesuvius erupted in the year seventy-nine, it buried a construction site, with its materials still waiting to be used. A study published in 2025 examined those piles alongside unfinished walls. The builders had mixed quicklime with dry volcanic ash before adding water. That order matters. When water reaches quicklime, it releases heat inside the mix. This hot mixing can leave calcium-rich lumps behind, the same reserves that can later help seal cracks. What looked like sloppy mixing may have been the fingerprint of the method. Researchers at MIT helped develop a modern version of the approach. A company called Dmat is adapting it for construction. The institute reports that its repair mortar has been used on Switzerland's N13 motorway. That's the practical payoff: repair materials that seal small cracks before they become a recurring maintenance problem. And it matters beyond repairs. Making cement produces around eight percent of the world's carbon dioxide emissions. Concrete that lasts longer has to be replaced less often. The question isn't whether we should abandon modern concrete. It's whether we can give it more of the durability these ancient materials achieved. Early strength and a long working life are different goals. A better material needs both. Pliny didn't have the chemistry. But for some Roman concrete, the day it hardened was only the beginning. Subscribe for more ancient engineering, explained through the evidence.