Friday is actually the day I do the least.

In a typical week, I am shattered by Friday morning. I’ve spent the week staring down a microscope, collected a lot images, and/or passed a couple afternoons staring at a green and white screen of the SEM. I have learned that it’s not good to spend too much time in the microscale. It’s important to remember that while the work I am doing is very detailed and requires a good bit of microanalysis, there also is a wider cultural component that is important on the macroscale. This morning, I have been reading some archaeological theory about the spread of ancient technology, operational chains, and the cultural importance of material production. This is the part I enjoy the least, but, as I’ve been reminded on several occasions, is just as important as the analytical details. In fact, there is little reason to even begin the analysis without a Bigger Picture reason for doing so.


Figure 1. This is my desk in its clean state.  That stack of papers is what I’ve been reading for the last 2 weeks. Usually there is a stone cold half-cup of coffee sitting there as well.




Actually, this is where the real work is done. All of the sample analysis and data collection would mean nothing if I didn’t spend many more hours pouring over the results, trying to tease out some sort of meaning from black and white images and long series of numbers. This is the part where I get to make charts! And data tables! This is the part I enjoy most.

In the afternoon, I was able to escape back to the lab and do some initial prep work for the petrographic work I am planning for next week. Here is the rig I am using:


Figure 2. I’m using a polarizing microscope with an attached video camera. There are four lenses from 4X to 40X, and the homogeneity of the concrete samples and variable particle sizes of the inclusions  mean that all lenses are useful at some point or another.


Before starting a quantitative point count, I take some time to get the know the sample. That involves having a look to see what the components are, if there is anything unusual about the sample, any unexpected components, etc. I take notes on a high resolution scan of the thin section and capture images of what I’m seeing down the eyepiece (you can sort of see an image on the screen there). For these initial examinations I usually work with the 4X objective lens and move up in magnification as needed. When I feel like I have a good idea what I’m dealing with, I set up the parameters I’ll need for the point count.

Figure 3. Here, I’ve mounted a computerized microstepper to the microscope stage, which moves the slide a specific distance. This way, I can be sure that the step distance between each is point is standardized and precise. The microstepper and PETROG software have been generously loaned to me for the duration of my research by the manufacturers.


At each point I identify and record what I am seeing with the 10X objective lens. The software allows me to record the data in a database for subsequent analysis. (I also keep paper records because the computer I am using is rather old and prone to crashes) What I’ve done this afternoon is program in the area of the thin section I want to count, the number of points I am aiming for (500-600), and the step distance (1 mm). Now I’m all set to do the full point count on Monday, which should take me about 6 hours.

When I’m finished with the microscope work, I will have generated a good description of the concrete fabric, recorded the size and nature of voids and cracks, and quantified the types of aggregate. I’ll also have a good idea which areas I want to target with finer detailed analysis with the SEM. All of this will (hopefully!) go a long way toward determining the original mix designs of the concrete, which I will compare across the site.

So that’s a typical Friday for me in the lab. A bit of reading, a bit of playing with data, a bit of exploration, and some planning for the week ahead. I have a few more samples to get through this summer and then I’ll be collecting more samples at the end of the summer. I am hoping to have everything analyzed by this time next year and to have my thesis submitted by the end of next fall. Fingers perpetually crossed.


Ancient concrete? Really?

Yes, really.  I first fell in love with old buildings in Pompeii, where I spent summers working as an excavator from 2002-2008. Every day it struck me that I was in a place that still looked and felt like a real city. To my mind, this was down to the fact that the buildings are still standing. After more than 2000 years. Someone did something very, very right when making those buildings and I want to know more.

For my D.Phil research, I have landed in an opportunity to study structures in Ostia, Italy, which is also a preserved city-sized site.  The structures I’m investigating are all brick and mortar masonry, with concrete filling up the center wall core. This is what Vitruvius called opus caementicium. To be honest, I’m most interested in the people who made it: the builders who developed this wonderful, magical material that is still performing successfully more than 2000 years after it was first installed. Where did they get their materials? Why were certain materials preferred over others? How were the materials processed and mixed together? How did builders’ choices affect the concrete and its performance? Were the same mix types used for both public and private structures? Why is this stuff still standing? These are the questions driving my research, and I am looking to answer them by investigating the material itself.

To give a quick overview, the mortar and concrete I am analyzing was made of lime, volcanic sand aggregate, and water. Sounds rather simple, however, the combination of materials they were using produced complex chemical reactions, known to modern concrete scientists as pozzolanic reactions, which resulted in a sophisticated, high quality material. My sample collection was collected from a series of structures in Ostia from the 2nd century CE, by which time – at least in Rome – concrete was well-developed and had been employed in large-scale Imperial building projects. My task now is to analyze the Ostian structures to determine how well-developed their concrete industry had become by that time. The benefit of a site like Ostia is that the ancient city is left largely in tact without modern development. This means that unlike in Rome, where centuries of modern development has destroyed all but the most protected monumental structures, it will be possible to evaluate the buildings within their original cultural context.

The analytical techniques employed for my research are borrowed from geology and concrete science, which makes this a truly interdisciplinary project. My samples are essentially synthetic composites of natural materials that can be investigated with traditional petrography. I’m using light microscopy of thin sections to identify and quantify the aggregate, to describe the cementitious matrix, and to identify any  obvious degradation features or alteration products. Today I’m working on point counting one of the samples, which is pretty straight forward. I move across the sample in 1 mm steps, and at each location I record what I see in the cross hairs of the eyepiece. Besides the obvious benefit of quantifying each of the different components, I’m also getting to the know the sample really well. As I go, I’m recording information about the state of degradation or alteration, the shape and fillings of any cracks or holes, particle size and shape, and any other details that may give me a clue about what the builders were doing when they made the concrete.

I am also using scanning electron microscopy (SEM) to collect high-resolution, high-magnification backscatter images of the samples. At this scale I can get a better look at the binder-aggregate interface to see how well-bonded these components are. It is also possible to see any microscopic cements that have formed in pores, cracks, and the vesicles of aggregate clasts that would otherwise not be visible. The SEM also detects the atomic weights of everything in the sample, which show up as differences in the greyscale colour of the image. It  also can calculate the chemical composition of the different components, so using a combination of chemical data and backscatter images, I can determine what types of cements have formed (strengthening) and how much leaching has occurred across the matrix (degradation). The ratio of calcium to silica is key in both cases.

X-ray diffraction is also on the menu, assuming I can find the funding to pay for it. This technique is incredibly useful for identifying the mineral assemblage in rocks and materials. In this case, I will use it to confirm the original petrographic identification of minerals in the aggregate and to find any other alteration minerals that could not be seen in thin section. The presence of certain minerals like gypsum or ettringite usually indicate alteration of the mortar itself, but minerals such as stratlingite and calcium-aluminum-silicate-hydrates suggest the mortar was rather well-formed in the first place.

So today, I’ll be giving an account of what it’s like for me in the lab. I realize that being stuck in the lab sounds like a death sentence to some people, but for me, it’s where the magic happens.