Roman concrete mineralisation explored

Materials World magazine
1 Aug 2017

A study into the structure of Roman concrete examines its long-term chemical interactions with seawater. Simon Frost reports. 

A research group at the Department of Energy's Lawrence Berkeley National Laboratory, USA, claims to have gained further insight into the long-pondered durability of Roman marine concrete. 

They found that seawater dissolved volcanic ash inside the concrete, creating an environment for new minerals to grow from highly alkaline leached fluids, particularly calcium silicate hydrates (Al-tobermorite) and phillipsite, which reinforce the cement matrix. The paper, published in American Mineralogist, claims that the findings could contribute to more robust formulations for structural marine concrete, CO2 storage reservoirs and cementitious barriers for hazardous waste.

Many ancient Roman marine structures remain intact, attesting to naturalist Pliny the Elder’s declaration two millennia ago that concrete, ‘As soon as it comes into contact with the waves of the sea and is submerged, becomes a single stone mass, impregnable to the waves and every day stronger.’ The Berkley Lab team’s findings support Pliny’s belief that seawater bolstered the material. 

In concrete samples taken from an ancient Roman pier off the west coast of Italy, the researchers found that the structure continued to produce alkaline pore fluids through low-temperature interactions between seawater and pumiceous volcanic ash, driving ongoing zeolite and Al-tobermorite crystallisation and allowing the material to flex rather than shatter when stressed. The concrete’s chemical interactions with seawater also enhance its pore space, pumice clast bonding and ability to sequester alkali cations of sodium and potassium. 

‘Contrary to the principles of modern cement-based concrete, the Romans created a rock-like concrete that thrives in open chemical exchange with seawater,’ said the study’s lead author, Dr Marie Jackson, a geology and geophysics professor at the University of Utah, USA. Modern Portland cement-based concretes combine materials such as limestone, sandstone, ash, chalk, iron and clay, which are heated, finely ground and combined with an aggregate that is intended to be chemically inert – they are designed not to react and mineralise in situ. 

Using samples of the concrete in thin, polished slices, the researchers mapped the distribution of elements in the mineral microstructures and examined their crystal structure through a combination of electron probe microanalysis, synchrotron-based X-ray microdiffraction, scanning electron microscopy and Raman spectroscopy. ‘We can go into the tiny natural laboratories in the concrete, map the minerals that are present, the succession of the crystals that occur and their crystallographic properties,’ Jackson said. ‘It’s been astounding what we’ve been able to find.’ 

Jackson is now attempting to replicate Roman concrete recipes, mixing seawater from the San Francisco Bay and volcanic rock from the western USA, as well as leading a drilling project to examine the production of tobermorite and related minerals at the Surtsey volcano in Iceland. She hopes this will lead to test structures that can be used to compare Roman-inspired concrete formulations with modern steel-reinforced concrete. 

Paul Lambert, Head of Materials and Corrosion Technology at Mott MacDonald, UK, played down the novelty of the study, however, telling Materials World that its claims that ‘the mineral fabrics provide a geo-archaeological prototype for developing cementitious processes through low-temperature rock-fluid interactions could be used to improve concrete in seawater’ were ‘somewhat fanciful’ and, although describing it as ‘quite an interesting paper’, said that these claims are ‘the very parts I, had I been a peer reviewer, would have requested removed.’ 

‘We don’t have problems with concrete in seawater relating to the calcium silicate hydrates (tobermorite, if you prefer, although it is not a term commonly used by cement chemists),’ he said. ‘Sulphate attack of the calcium aluminate phases, yes – but this has long been sorted. The same goes for corrosion of the reinforcement. Refinement of the cement matrix by exposure to seawater was thoroughly researched and described by the Concrete in the Oceans Programme in the late 1980s.’  

Jackson recently suggested that the proposed Swansea Bay Tidal Lagoon, UK, could be the ideal application for a Roman-inspired concrete, claiming that the lagoon would need to operate for 120 years to recoup its building costs, by which time ‘it would be a mass of corroding steel.’

To read Phillipsite and Al-tobermorite mineral cements produced through low-temperature water-rock reactions in Roman marine concrete in full, visit