This lecture concludes our look at isotopic analysis, or CSIA. We're looking at some of the more advanced ways that it's used during this lecture. So Chuck, I know that demonstrating attenuation using CSIA can be complicated by the fact that degradation is not the only processes that cause a shift in the isotopic ratio of a compound, meaning that the Raleigh equation that is used to quantify and describe these systems is probably an oversimplification of a more complex set of processes. >> Yeah, you're right, that's one of the major challenges of trying to use CSIA. In a case for monitored natural attenuation. >> Yeah, and so that problem we sort of layout on this first slide. It's part of an over arching issue that the data can be sort of difficult to interpret at times. Meaning, that if you were to give these sorts of isotopic data to two different people, you might end up with two different interpretations. >> So that sort of thing can be difficult when you're just trying to make an argument to a regulator or to stakeholders. >> Yeah. The other thing is that there's sort of this increasing recognition that fractionation occurs with processes that aren't destructive. >> So we want to demonstrate that attenuation is occurring by these destructive processes, but there's other processes that might actually cause shifts in this isotopic signal. >> Okay, and some of those might be dispersion, diffusion, sorption, and volatilization, right? So these are pretty important processes for understanding natural tenuation. And so, how do we address those? >> Well it all comes back to this idea that the Rolle equation is applicable to all situations. In reality it's suitable to closed systems, but probably an over simplification for complex heterogeneous systems like an aquifer. Dispersion and diffusion are issues in heterogeneous aquifers, as we all know. Dispersion doesn't cause fractionation, but it does shift the isotopic signal due to the fact that you've got mixing between compounds that have traveled at different rates. So it especially at plume fringes. Paul Hatzinger from CBNI wrote a paper in 2013 that has a nice explanation of why. So I'll read a quote here he says, typical field samples are mixtures of water and solutes that may have different source areas, flow pass, and travel times. That is individual samples must be interpreted as mixtures of sub-parcels with varying reaction progress values. >> Mixtures of sub-parcels, a really nice way to phrase that. >> Yeah and there's other processes like bottlilization that do cause some fractionation, because lighter isotopes are prefered on the effects of dispersion in isotope ratios in ground water has been established since at least, 2008 there was a great paper by Eric La Bolle and other researchers at UC Davis. And I'd also like to show on this slide, in the upper right-hand corner there's this graphic from a paper by Boris van Breukelen at University of Amsterdam. It shows how diffusion induced fractionation or DIF is an important process. >> Okay, so this sort of complicated graphic. But they're showing what would happen to a source of vinyl chloride in an that's flowing from left to right. Now the depth is shown on the vertical axis from the surface to the top of the deeper intervals moving downward. The vinyl chloride source is located between the two different domains. And the one at the top is labeled the reactive fringe, where mixing and degradation occurs. So in this reactive zone, there's plenty of oxygen to promote this biological oxidation of vinyl chloride. So that's why you see the concentration go from zero in this C versus depth graph on the left side of the graphic. But they've also defined this second zone below the source. One that they label conservative because only mixing occurs, no degradation. In this conservative zone, the concentration decreases with depth due to diffusion. >> But they've also included in a graph, with the isotopic ratios versus depth on the right hand side of that graphic. In the reactive zone, you see that the degradation causes an enrichment in C13 and CL37, the chlorine isotope as expected. And that effect is strong enough that it doesn't really change whether or not you assume that diffusion induced fractionation, that's that DIF term, whether or not you assume that's occurring. But then if you go down to the deeper that conservative zone, you see that there's a shift in the other direction, lighter molecules are preferentially diffusing into this layer that results in a different isotopic ratio that what you'd expect if you neglected diffusion induced fractionation. So the fractionation factor for carbon isn't large compared to some processes when we're talking about diffusion, but it is interesting to note that it's at least twice as large for chlorine, which makes sense because there's a lot more difference in the relative size of those. >> Okay, so all right, that's a lot of time for one slide, but I think it illustrates that there needs to be some thought in how these data are analyzed. So one method that can help us is to evaluating some of this other isotopes, more than one isotope at a time, right? >> Yeah, and that's what's shown sort of on this slide. Remember, these are largely organic compounds. You can always analyze for carbon isotopes. But by combining with other isotopes, things like hydrogen or chloride, you can get some really valuable diagnostic information. So let's walk through a few of those benefits. >> Okay, so this slide shows the primary benefit. Which is stronger evidence that fractionation is occurring. >> Yeah, so this plot is showing the isotopic ratios for hydrogen on the y axis and carbon on the x axis for benzene in groundwater samples from different wells. The enrichment of carbon when benzene is being degraded, whether it's aerobic or anaerobic isn't too larger, so that's reflected in those epsilon values that are somewhere between -1.5 and -3.6. >> But enrichment of the heavier hydrogen isotopes is much bigger. That's apparent in this large epsilon value that describes this hydrogen fractionation. Another way to tell is just by looking at the slope of those lines. They're pretty big. >> Yeah. So the take home message here is that, the shifts in one isotope might be more apparent than the shifts in the other isotopes. So another benefit of this sort of dual isotope approach is, it allows you to better estimate the parameters of interest, these enrichment factors. So when you're analyzing two isotopes, you use this lambda, that's there on the lower right, to express the ratio between the enrichment factors for the two different isotopes. So, essentially larger numbers help give you more confidence that you're seeing real fractionation, something above the noises associated with the analytical uncertainty or other things. So, this in terms sort of gives you a better basis for estimating the reaction rate, the degree of degradation and some of those other calculations that we described in the previous lecture. >> And finally, there's this idea that you can plot data on two different isotopes and get an idea of what type of pathways occurring. This is another one of those things that showing on this graph from this fisher paper, where the anaerobic pathway shows a lot more fractionation of a hydrogen ion than the aerobic pathway. So the samples that they collected and analysed followed this anaerobic curve pretty closely and told them, that's what the pathway was. >> Yeah, and we should note that this is something that's easier done with lab studies then field data. And there definitely is some argument about whether you could actually distinguish between pathways using isotopic data, using field data. So example of a study where this has been attempted is shown on this next slide. This is a 2D plot, very recent paper by Palau et al. A ST paper. But they were looking at what happen to the carbon and chlorine isotopic ratios in 1,1,1-TCA along the plume transect. So based on lab studies, they knew that if TCA was being oxidized, it would follow this reaction pathway C that you see there. So that's the green one, right? But it's being degraded through hydrolysis or dehydrohalogenation to 1,1-DCE, we follow that pathway A. And then pathway B, is abiotic reductive dechlorination, using zero valent iron. So when they got their data, they plotted it, and it followed this sort of black line that you see, in between the A and the B pathways. So based on these data, as well as other data they had, like concentration patterns, and daughter products, they confirm that pathway A was the most representative in this case. >> So pretty neat but in reality, you might not have enough information to establish this at many sites. Remember, you've got the contributions of the non destructive processes that you may also have to worry about. So that's why we label it as a possible benefit. Maybe something to take a look at, if you got enough data. >> Okay. A couple of additional points on dual isotope analysis are shown on this next slide and that's using this data to help you identify sources or differentiate between sources. There's lots of examples of this but a good one is another study of lab in Amsterdam. So in this case, they're showing the chlorine isotopic ratio on the y-axis and oxygen isotopic ratio on the x-axis. In this case, they were measuring these ratios in samples of perchlorate. And they were easily able to distinguish between samples that originated from a natural perchlorate, so that's source A and from a synthetic perchlorate, source B. And so what's more they use this model that they developed, mixing model, to predict what happened as each degree didn't mix. So that FA refers to the fraction of source A that is present, relative to source B. And based on the extent degraded, or ED, that's what shown on here. As that increased, you can see the type of enrichment that you'd be expected to see, based on the amount of source that was present. >> So this is really source of portionate, right? Which isn't necessarily a big priority for M&A, right? >> Yeah, but it shows an example of sort of multidimensional modeling, and that's why we've sort of included this as a final topic, because it's showing where this field is heading. Making CSI more applicable to sort of ground water, where you have bait and transport issues to deal with. And so, we're sort of showing that then here on this this final slide. So this is an example of data that's been proposed as Generation 3 MNA Analysis. Chuck, I know you were involved in this project, so maybe explain what the objective was here. >> Okay, so this is an ESTCP project that was lead by Paul Phillip and Thomas Kuter at the Oklahoma, along with Boris VonBruchlen at VU Amsterdam and GSI. It's just wrapped up and you can see that guidance is already available. But the basic idea here was sort of way to incorporate this CSIA data into a reactive transport model to simulate this isotopic shifts and space and time, including these degradation byproducts, so that's what the team there labeled this Generation 3 MNA Analysis. >> So that means there's a Generation 1 and a Generation 2 MNA Analysis as well. >> Right, you can sort of think about it, Gen one is the simples concentration versus time and concentration versus distance evaluations. Generation two is using CSIA as a line of evidence. >> And generation three then is CSIA with reactive transport modeling. So Chuck, what's going on in these graphs here? >> Okay, lots of colors here but the left panel is the actual contaminants themselves starting with PCE on the top. Going all the way down to ethene on the bottom. Y axis is depth. The X axis is distance away from the source. The more red you get, the higher concentrations that you have, and so you can sort of see this plume in all the daughter products and different places and in space. Then in the middle panel, that's the isotope information you can look at. And so, here they are looking at the Carbon 13. And you can sort of see how much of these different constituents are degraded by their color. Red, again, represents more degradation. All this was done by combining this CSIA data at a site with reactive transport model, in this case it's PHAST-2 and they're assuming in this case, they want to complete the reductive dechlorination to Ethene. >> Well, really great way to shows serve the power of some of these data and some of things that you can do with it. So let's wrap up, some are key points from this lecture. Analyzing and plotting data from two or more isotopes can provide additional forensic information. >> Okay, and then one potential benefit of this type of dual isotope analysis is to identify or differentiate. Relevant attenuation pathways such as biotic versus abiotic. >> And then finally reactive transport modeling is emerging as a powerful method for interpreting CSIA data.
