What Made Roman Concrete so Good?
Now with extra added Pozzolana! Ask for it by Name!
I used to bus tourists along the Sorrentine coast and lecture to them about sirens and Hercules and how they shouldn't confuse cement and concrete. That last one was only because we drove by a site on the Castellammare coast road known as "the old cement factory." It has since been converted into a 4-star hotel called the Hotel Crowne Plaza. I don't think it's doing too well. I mean, who wants to stay in a place that looks like an old cement factory, in spite of the admittedly charming post-post-Modernist Alien towers? You think I'm kidding? (image, above left)
The image (above right) shows the arched pier in Roman Puteoli (Pozzuoli).It took anywhere from 15 seconds to four hours to drive by (depending on traffic!), but I could usually make do with a short spiel on the order of "cement is the powder that you mix with water and sand and gravel to get concrete, the building material that sets and hardens." If traffic got slow, I had to "go long," stretch it by adding that cement is made from burnt lime and clay and technically is called a "binder" and that the sand and gravel you add are called "aggregates". Then (now it's really slowed to a crawl) I might have to add that the chemical name for lime is calcium oxide (CaO) and that if you make the aggregates finer, you can get other kinds of material such as mortar, which holds bricks or stones together in masonry, or plaster, used for the protective and decorative coating of walls and ceilings and for molding and shaping decorative elements. Only once, during the Mother of all Traffic Jams, did one of the group say something like "Hey, what about that aluminum tobermorite stuff?" He was an engineer, whom I easily distracted by pointing out the mermaids just off shore.
It is reproduced in various sources and labelled "from a Roman
picture after Bellori" in reference to Fragmenta vestigii veteris
Romae by J.P. Bellori (1615-1696), a French archaeologist.
Also see link to The Roman Port of Pozzuoli at bottom of this page.
Normal modern concrete does not hold up very well in seawater. In as few as 50 years underwater, it may start to decay. Even concrete reinforced with steel rods to increase the load-bearing strength of the concrete and to offset concrete's poor tensile strength will decay. On the other hand, there are submerged Roman piers on the west coast of Italy that are not only still standing, but — and this is the fascinating part — are structurally sounder now than when they were built! (That is what sparked my initial interest.) This is because they used what we call Roman concrete, which has significant differences in what is added to the mix; that is, the quality of the lime and the make-up of the aggregates. Philip Brune, a research engineer at Dupont, has studied the properties of Roman monuments and says that Roman concrete "...is the most durable building material in human history... [and] an extraordinarily rich material in terms of scientific possibility... ."
Marcus Vitruvius Pollio (born c.80–70 BC, died after c.15 BC) was known simply as Vitruvius and was the author of a multi-volume work entitled De architectura. He has been called "the first architect" but today we might call him a civil engineer; he was a true Jack of all Building Trades. Much recent research has gone on in the Gulf of Naples to determine just why he went to such pains to specify in his recipe for concrete a mixture of volcanic ash, lime, seawater, and volcanic rock, and, importantly, that the best volcanic ash comes from the Bay of Pozzuoli. That's the reason so many geological terms start with "pozz...": pozzolans, Pozzolana, pozzolonic ash, pozzolanic ativity, pozzolanic reaction, etc. (Pozzolana is capitalized because it is not generic; it is a specific proper noun — volcanic ash from Pozzuoli.) Vitruvius is most specific: there are two types of concrete: one for use on land and another for underwater. He specifies a ratio of 1 part lime to 3 parts pozzolana for cements used in buildings and a 1:2 ratio of lime to pulvis Puteolanus (same as Pozzolana, volcanic ash) for underwater work. Important in both cases is that Roman concrete was very heavy on the aggregate, so heavy that it might be called "rubble" by modern standards.
The key here is that the area of Pozzuoli was (and still is) an active volcanic site; it was the site of the mammoth Campiflegrean caldera collapse that created that part of the bay of Naples 50,000 years ago. If you want volcanic ash, this is still the place to come. Vitruvius was particularly concerned with a series of harbors he envisioned being built from Naples along the coast northward for about 300 km (200 mi) to the Orbetello peninsula, where Tuscany begins, harbors that would flank Rome and protect her on both sides. That would, indeed, entail putting into place giant blocks of "the most durable building material in human history." And Marie Jackson, a geologist and research engineer at the University of California at Berkeley reminds us that “The Romans shipped thousands and thousands of tons of that volcanic ash around the Mediterranean to build harbors from the coast of Italy to Israel to Alexandria in Egypt to Pompeiopolis in Turkey,”
Recent research has shown that the secret has to do with what happens chemically in the interaction between sea water and Pozzolana, the volcanic ash, dissolved in the water near Pozzuoli: new binding minerals actually grow in the concrete and produce within a short time a rare hydrothermal mineral called aluminum tobermorite. Al-tobermorite can be made in the lab but is difficult to incorporate into concrete. Yet when open seawater percolates through a cement matrix, it reacts with the volcanic ash to form Al-tobermorite and a porous mineral called phillipsite, both of which are necessary to fit Vitruvius' description of concrete exposed to seawater as: “a single stone mass, impregnable to the waves and everyday stronger.” Essentially, he invented what is called hydraulic (or hydraulic-setting) cement, which can be poured into wooden molds on land and then put in place underwater, or even poured underwater (!) into pre-placed molds as long as the cement is accurately guided into place by a conduit. (One presumes that the 1:2 ratio (underwater) is about where the aluminum tobermorite "threshold" was, the point where the crystals started linking and strengthening the mixture. Anything below that, you could use on land.
When Pliny the Elder wrote (in his Ode to Concrete!) that "as soon as it comes into contact with the waves of the sea and is submerged, it becomes a single stone mass, impregnable to the waves") he was describing an ideal solution to many modern engineering problems. If the seas keep rising, think of the sea-walls that might be necessary to shelter shoreline environments. Think of the advantage of not needing reinforced steel to strengthen concrete when the concrete itself is packed with tiny crystals (that don't rust) that grow, expand, and spread, linking up to actually strengthen the concrete as the years pass.
De Agostini Picture Library, Getty Images
It bears emphasizing that this entry was primarily meant to deal with the successful use of a very special kind of Roman concrete used in underwater environments. Modern engineering reinforces concrete for use on land — buildings, bridges, etc. — to deal with the conflicting properties of concrete and steel: concrete is said to be strong when compressed, but weak under tension; steel is strong in tension but weak under compression. Thus, structural engineers try to place the "rebar" (short for 'reinforcing bar,' the metal used, usually steel or iron) where the finished product — the building or bridge — strikes a balance between tension and compression.
That has little to do with structures meant to be constantly exposed to seawater percolating through a concrete matrix (such as Roman seawalls on the coast) and growing stronger by the year through the magic chemistry of Pozzolana volcanic ash. Yet the Romans were very adept at building on land and we should not forget the role that Pozzolana played in the countless Roman bridges, aqueducts, and buildings. Even if not constantly reinforcing itself through constant exposure to seawater, the 1:3-ratio concrete used on land was infused with Pozzolana and was strong. The Pantheon in Rome (image above) still contains the largest unreinforced concrete dome in the world more because of the incredibly refined architectural devices used to minimize the tensile forces on the concrete — that is, nothing is pulling the concrete apart; the concrete dome is practically all under compression, not under tension. A simple look at the cutaway image reveals some of the tricks the Romans used: the outside of the dome is stepped for example; every step represents concrete that was cut away and jettisoned, so to speak, to save weight. The "oculus" (eye) at the top (1) saves weight (it's a hole!) and (2) lets a shaft of sunlight in, one of the most attractive aesthetic features of the Pantheon. (It might have been a heavy slab of concrete!) Very refined is the fact the wall of the dome gets progressively thinner (thus lighter) as it goes up. Inside there is no extra baggage, no ceramic pots, busts, none of that. On the contrary, there are a great number of "coffers", decorative sunken panels, each one a spot where even more weight was cut out and thrown away. By some estimates, the engineers saved 80% (!) of what might have been the weight of a full-blown don't-spare-the-concrete version of the dome with no attempt to trim it down, which, we note, wouldn't have lasted very long.
A final silly note, and I include this only as a quantum afterthought (by which I mean that nothing is so outlandish that it is not possible) is that the profile of the Pantheon dome sure looks like a Fresnel lens in optics and maybe this is where Augustin-Jean Fresnel (1788–1827), the French engineer, got the idea for his annular ring design (allowing the construction of multi-part, thin, light lenses without the size and weight required by lenses of conventional design. If I had a time machine, I would go back and ask him "Augie, c'mon. I won't tell. You got this stepped idea from the Romans, right?"
One last thing (only partially silly!). The confusion between cement and concrete is due to the fact the Romans, themselves, in Latin, called the building mixture (what English speakers call "concrete") opus caementicium. Modern Italian uses calcestruzzo (calcium + structure) for all 'concrete' (Roman or otherwise) but also cemento armato — armored cement (for Roman concrete) presumably because it is armored with bits and chunks of volcanic aggregate. Italian has no noun concreto, just the adjective concreto, but only in the sense of a "concrete solution." Don't ask about Spanish. Really, please. Don't ask. Oh, in Italian, struzzo also means ostrich, in which case you've got yourself a bridge with added protein! The plot thickens. Eventually it will harden.
My thanks to Preston Lamp for his comments and help. Any mistakes or oversimplifications are mine.
sources and further reading/viewing:
The Secrets of Ancient Rome's Buildings in Smithsonian Magazine.
Heather Lechtman and Linn Hobbs. "Roman Concrete and the Roman Architectural Revolution", Ceramics and Civilization Volume 3: High Technology Ceramics: Past, Present, Future, edited by W.D. Kingery, published by the American Ceramics Society, 1986; and Vitruvius, Book II:v,1; Book V:xii2
Jackson, Marie D.; Mulcahy, Sean R.; Chen, Heng; Li, Yao; Li, Qinfei; Cappelletti, Piergiulio; Wenk, Hans-Rudolf (2017). "Phillipsite and Al-tobermorite mineral cements produced through low-temperature water-rock reactions in Roman marine concrete". American Mineralogist. 102 (7): 1435–1450. doi:10.2138/am-2017-5993CCBY. ISSN 0003-004X.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
added Jan. 2023. My thanks, again, to Jeff Miller for this added information. When it comes to concrete, this guy is really solid.
All that was the long of it, but there is also a "short of it" and it's important. Here.
Very recent studies speak of a "mystery ingredient", previously ignored, that helps explain how the majestic structures of ancient Rome have survived for millennia even in docks, sewers and earthquake zones. The study team, including researchers from the United States, Italy and Switzerland, analyzed 2,000-year-old concrete samples that were taken from a city wall at the archaeological site of Privernum (image, right), in central Italy. They are similar in composition to other concrete found throughout the Roman Empire. The team found that white chunks in the concrete, referred to as LIME CLASTS, gave the concrete the ability to HEAL CRACKS that formed over time. The white chunks were previously overlooked as evidence of sloppy mixing or poor-quality raw material. (A "clast" is a piece of rock, as in "pyroclast", a burning rock from a volcano.)
What that means is the concrete structures were healing their wounds, a concept that is one associate with living organisms! It's what trees do. It's what you and I do. This type of concrete "allowed the Romans to have an architectural revolution and create and turn their cities into something that is extraordinary and beautiful to live in. And that revolution basically changed completely the way humans live." (Citation: Admir Masic, associate professor of civil and environmental engineering at MIT.)
Analysis of the concrete showed that the lime clasts formed at extreme temperatures from the use of lime (aka quicklime), and "hot mixing" were the key to the concrete's durability. This suggests that the lime clasts can dissolve into cracks and recrystallize after exposure to water, healing cracks created by weathering before they spread. Researchers said this self-healing potential could pave the way to more long-lasting, and thus more sustainable, modern concrete. It is true that volcanic ash from the area of Pozzuoli, on the Bay of Naples, was what made Roman concrete so strong, but it was one key ingredient but not the only one. The Romans didn't tell us about the lime.
The original research was published in the journal Science Advances at
Vol. 9, No. 1. "Hot mixing: Mechanistic insights into the durability of ancient Roman concrete"
[Also see related item, The Roman Port of Pozzuoli.]