Characterizing ancient materials for the modern day
Linda Seymour of the Masic Group investigates the longevity of ancient Roman concrete
Courtesy of the Masic Group
Lab of the Week
The Masic Group: Laboratory for Multiscale Characterization and Materials Design
Room 1-347
Course 1: Civil and Environmental Engineering
Ancient buildings are useful for learning how past civilizations lived, as well as the aesthetic values and architecture of their respective time periods. Surprisingly, we can also glean technological information from these grand structures — in particular, we can examine the materials with which they were built to learn how and why they remain standing today.
Part of the Masic Group’s work focuses on analyzing these structural materials, hoping to draw inspiration from ancient materials as well as biological ones to inform the production of new materials. Linda Seymour, a graduate student in Masic Group, is particularly interested in Roman concrete and mortars. “There’s no denying that Roman concrete, Roman structures, have been standing for 2,000 years. I can’t exactly mimic that in a lab in six months. So we’re taking these known examples and trying to figure out — but why? How has it been standing for 2,000 years when our concrete structures fail after 50 to 100 years?”
The many workbenches in the Masic Group contain a variety of specimens, from Roman mortar to terracotta tile. At one corner of the lab sits a Raman microscope that can analyze structural connections between atoms, and at the other end, tools that can distinguish between types of mortar at an archeological dig site.
For Seymour, these tools are invaluable for learning from ancient materials: “What we’re trying to do with these machines is to chemically characterize ancient structural materials, such as Roman mortar, that we have clear proof that they’ve lasted for millennia.” She hopes to reverse-engineer the chemical aspects of Roman materials to find out what makes them so durable. Her first project in Masic Group involved examining cocciopesto, a material made from crushed pottery. The ancient Romans used cocciopesto in structures exposed to water to help them last longer.
Seymour attempted to determine the chemical interaction occurring in cocciopesto by adding the material to Ordinary Portland Cement (OPC), which is the cement conventionally used in structures today. Using the Raman microscope, she found that the OPC interacted with the cocciopesto to create calcium silica hydrate (CSH), a primary structural component of modern cement. “The cocciopesto was not only acting as an aggregate, but it was also acting as a reactive material within the concrete, which is really exciting. It was really cool to see that this ancient material or ancient-inspired material was working in unison with our modern material. That was the first glimmer of hope that we would be able to do this.”
More recently, Seymour has been intrigued by a phenomenon that occurs in Roman concrete but not in Ordinary Portland Cement. She notes that the binding phases we observe in the ancient Roman concrete are “gel, amorphous phases — they’re not really crystalline ... Even after 2,000 years, we have these gel-like phases present, they haven’t all crystallized over time.”
Seymour hopes to get to the bottom of the crystallization mystery by looking holistically at the concrete samples to understand their chemical composition and how these chemicals interact with each other. She is limited, however, by the samples’ time scale. “The problem with trying to understand this is that all of our Roman samples are over 2,000 years old. We don’t have a 500-year-old sample to compare it with, or even a 1,000-year-old sample to compare it with. We don’t necessarily have too many snapshots in time, unless they’re just a couple centuries apart.”
Seymour continues to work toward a better understanding of these ancient materials by reconstructing them. As part of the Course 1 mini-UROP program, Seymour supervised two undergraduate students as they mixed Roman concrete and Ordinary Portland Cement to compare their properties. The cement blocks are in the shape of small cylinders, with the Roman concrete crumbling at the touch in comparison to its rigid OPC counterpart.
The differences between these two concrete blocks explains why we don’t use Roman concrete today. “Ordinary Portland Cement chemically sets much faster than Roman concrete does. Roman cement takes on the order of months to set, whereas Ordinary Portland Cement reaches 98 percent of its strength in 28 days, much quicker.” OPC is also much stronger than Roman cement, but according to Seymour, Roman cement's longevity may be more valuable. “Do we really need to be able to hold all of this stuff, if we’re just building a road? Is that necessary to have such high strength?”
In addition to investigating antiqua-inspired and bioinspired materials, the Masic Group provides learning opportunities to undergraduates by sponsoring a trip to Italy for research fieldwork. “We’re all about teaching others that we can look at these materials differently than we’ve traditionally thought. A lot of times we’re looking at old materials to try and figure out, ‘What were they thinking? What were they doing?’ for that historical perspective, which is really important, and has informed a lot of really interesting research, but we’re trying to take it to that next level and say, ‘What were they doing and how can we make it better?'”