The secrets of Roman concrete

Across the territory of the former Roman Empire, structures built some two thousand years ago are still standing: harbour walls, aqueducts, bridges and, most famously, the Pantheon in Rome, whose vast dome remains the largest ever built of unreinforced concrete. By contrast, many modern concrete structures begin to crack and crumble within a few decades, and the cost of repairing roads, bridges and buildings has become a major burden for governments. It is hardly surprising, then, that engineers have become increasingly curious about the material the Romans used, and about whether its secrets might be useful today.

Roman concrete differed from its modern equivalent in several important respects. Modern concrete is based on Portland cement, which is made by heating limestone and clay to around 1,450°C in a kiln. The Romans, by contrast, combined lime with volcanic ash, and then mixed this mortar with chunks of rock or broken brick. The most prized ash came from the area around the Bay of Naples, near the town of Pozzuoli, and the Romans recognised that it gave their mortar unusual strength. The first-century architect Vitruvius described its properties in his writings on building, although, like other ancient authors, he could not explain why it worked as it did.

Part of the explanation emerged only recently, through studies of Roman harbour structures that have stood in seawater for two millennia. For modern concrete, seawater is destructive: salt penetrates the material and corrodes the steel bars that are usually embedded within it for reinforcement. In the Roman examples, however, researchers found that seawater percolating through the concrete had reacted with the volcanic ash to form new minerals, which grew in the cracks and pores and actually made the material more cohesive over time. Rather than being steadily weakened by its environment, the concrete had, in a sense, been strengthened by it.

A second discovery concerned small white lumps of lime, known as lime clasts, which are found throughout many samples of Roman concrete. These had long been regarded as evidence of careless mixing or poor-quality ingredients. A team led by the materials scientist Dr Paolo Ferri proposed a different interpretation. In their view, the lumps were the result of a technique called hot mixing, in which quicklime, a highly reactive form of lime, was added directly to the mix instead of being combined with water first. The heat produced by the reaction would have allowed the concrete to set faster, but its most remarkable consequence came much later. When cracks formed and water seeped in, the lime clasts dissolved and the calcium they contained recrystallised, filling the cracks. The concrete, in other words, was capable of repairing itself.

These findings have generated considerable excitement, partly because of concrete's environmental impact. Cement production is one of the largest industrial sources of carbon dioxide, responsible by most estimates for around seven to eight per cent of global emissions. Much of this comes from the chemical reaction that occurs when limestone is heated, and much of the rest from the fuel burned to reach such high temperatures. Roman-style mixtures, which require lower kiln temperatures and use volcanic ash or similar materials in place of some cement, could in principle reduce these emissions. Concrete that lasts longer would also need replacing less often, which would further reduce the environmental cost.

Nevertheless, experts warn against romanticising ancient technology. Roman concrete is considerably weaker than modern concrete when compressed, and it gains strength slowly, which would make it unsuitable for many modern purposes, from tall buildings to motorway bridges that must carry heavy traffic. The Romans compensated for these weaknesses with very thick walls and arches, an approach that would be impractical and expensive today. Moreover, the Roman structures that survive are, by definition, the successful ones; countless others have collapsed or been demolished, and we have no way of knowing how many failed. As the civil engineer Dr Hannah Brook puts it, “We are looking at the winners of a two-thousand-year contest.”

Rather than copying the Romans directly, most researchers hope to borrow particular principles from them. Some laboratories are experimenting with adding lime clasts to modern mixes to give them a degree of self-healing ability; others are investigating volcanic ash and industrial by-products as partial replacements for Portland cement. Whether any of these approaches will be adopted on a large scale remains to be seen. The construction industry is cautious, for good reason: buildings must be safe, and new materials must be tested for many years before engineers can trust them. But the lesson of Roman concrete may be less about any single ingredient than about a different way of thinking: designing materials not simply to resist their environment, but to work with it.