WEBVTT 00:00:01.000 --> 00:00:06.000 All right. Hi, everyone. It is now 1 p.m, so we're going to go ahead and get started. 00:00:06.000 --> 00:00:10.000 Let's get into some quick housekeeping rules before we jump into the training today. 00:00:10.000 --> 00:00:17.000 Today's training is being recorded and will be offered on demand through Kluin for those who are not able to join us during our live sessions. 00:00:17.000 --> 00:00:21.000 The clue and training page also has the slides available for download. 00:00:21.000 --> 00:00:31.000 While in Zoom, you have access to the Q&A window. Please use the Q&A box to ask questions, make comments, or report technical problems at any time during today's training. 00:00:31.000 --> 00:00:37.000 We have several team members serving as subject matter experts in the background and they will be responding to your questions. 00:00:37.000 --> 00:00:43.000 Please note that time likely will limit us and we expect not to answer all the questions that are submitted today. 00:00:43.000 --> 00:00:48.000 Additionally, our trainers are not able to respond to questions that ask about states. 00:00:48.000 --> 00:00:52.000 Or EPA actions as well as legal or regulatory interpretation questions. 00:00:52.000 --> 00:00:59.000 For those interested in receiving certificates of completion, they are available at the end of the training by completing the feedback form. 00:00:59.000 --> 00:01:05.000 The feedback form for all ITRC courses are located in the Kluent training page for each specific course. 00:01:05.000 --> 00:01:13.000 Today's training is sponsored by the Interstate Technology and Regulatory Council, or ITRC, and is hosted by the US EPA Cleanup Information Network. 00:01:13.000 --> 00:01:19.000 Also known as Kluwen. 00:01:19.000 --> 00:01:25.000 The Interstate Technology and Regulatory Council is a program of the Environmental Council of the States. 00:01:25.000 --> 00:01:30.000 Itrc is a state-led organization composed of over 1,000 members from state agencies. 00:01:30.000 --> 00:01:39.000 Federal government, the private sector, academia, and community stakeholders. Itrc members participate in technical teams which produce tools. 00:01:39.000 --> 00:01:44.000 Resources and training courses such as the one you are participating in today. 00:01:44.000 --> 00:01:50.000 Please visit the ICRC website if you're interested in becoming an active member or need to view the full ITRC disclaimer. 00:01:50.000 --> 00:01:58.000 If you plan to use ITRC materials, we ask that you review those terms and conditions in detail and be sure to credit ITRC. 00:01:58.000 --> 00:02:03.000 Itrc is partially funded by the US government. Itrc nor the US government warranty the material. 00:02:03.000 --> 00:02:18.000 Nor endorse any specific products. On behalf of the Interstate Technology and Regulatory Council, welcome to today's training, PFAS Beyond the Basics, Fate and Transport Site Characterization and Source ID Training. 00:02:18.000 --> 00:02:25.000 Which is based upon ITRC PFAS team online resources. My name is Taylor Vogel and I will be your moderator today. 00:02:25.000 --> 00:02:34.000 This training class builds on the earlier information for introductory PFAS topics presented in the PFAS 101 clue and training. 00:02:34.000 --> 00:02:40.000 The Beyond the Basics class provides more in-depth information for fate and transport, site characterization. 00:02:40.000 --> 00:02:51.000 Characterization, source ID, and some introductory information on environmental forensics. These topics will be presented along with options and a framework for data visualization. 00:02:51.000 --> 00:02:55.000 This training will focus largely on PFAS, fate and transport and groundwater. 00:02:55.000 --> 00:03:12.000 Before we dive in, just a quick note. Some of the links presented in today's training may be broken or out of date. If you run into any issues accessing the link, don't worry, you can always navigate directly to the main guidance document and use the table of contents to find the specific sessions we are referencing. 00:03:12.000 --> 00:03:25.000 I will now be turning it over to Robert with the Alaska Department of Conservation to get us oriented on today's training. 00:03:25.000 --> 00:03:40.000 Okay. Thank you, Taylor. So as Taylor noted, this is the Beyond the Basics training for PFAS fate and transport site characterization source identification and forensics. 00:03:40.000 --> 00:03:48.000 My name is Robert Burgess. I'm from the Department of Alaska Department of Environmental Conservation. 00:03:48.000 --> 00:03:55.000 We also have with us Ted Campbell from the North Carolina DEQ. 00:03:55.000 --> 00:04:02.000 Skyler Soursby with WSP, Dina Drennan, BM Systems, and Chris Oliveris with the University of California, Irvine. 00:04:02.000 --> 00:04:06.000 So those are your trainers today, and we'll just jump right into it here. 00:04:06.000 --> 00:04:18.000 This is, again, part of our beyond the basics trainings. So we have five of these beyond the basics trainings This one, of course. 00:04:18.000 --> 00:04:31.000 Is the… As mentioned, source ID, site characterization, fate and transport. We also have treatment technologies, biosolids, soil leaching and mobility, which has some overlapping content with this one. 00:04:31.000 --> 00:04:44.000 Just a little bit. Human health effects, ecological toxicity and regulations and then a sampling and an analysis beyond the basics course as well. So check those out. There's good content in all of that. 00:04:44.000 --> 00:05:01.000 So for today's for today's Today's training, our objectives are really to help folks understand the fate and transport processes that are unique to PFAS due to the physical and chemical properties of PFAS. 00:05:01.000 --> 00:05:09.000 And then we'll talk a little bit about how to characterize these sites with those processes and properties in mind. 00:05:09.000 --> 00:05:16.000 Keeping in mind how that fate and transport informs our approach to site characterization. 00:05:16.000 --> 00:05:23.000 And then we'll look briefly at the application of forensic techniques for source identification and differentiation. 00:05:23.000 --> 00:05:36.000 The goal is really to provide detailed information about the inputs practitioners can use to develop a robust conceptual site model to help understand fate and transport of PFAS at a particular site. 00:05:36.000 --> 00:05:43.000 And how to gather evidence when we have potential multiple sources. 00:05:43.000 --> 00:05:50.000 We will really focus on this presentation largely on PFAS transport in groundwater or the saturated zone. 00:05:50.000 --> 00:05:56.000 We will touch briefly on transport and other media, but for additional in-depth information. 00:05:56.000 --> 00:06:04.000 You can take a look at the online guidance document in section five for faint and transport, section 10 for site characterization. 00:06:04.000 --> 00:06:11.000 And you can also get some information from the other PFAS Beyond the Basics series. 00:06:11.000 --> 00:06:26.000 So we are going to start by talking about patent transport considerations and At this point, I'll turn it over to Dina Drennan. 00:06:26.000 --> 00:06:49.000 Great. Thank you, Robert. So as Robert alluded to, I'll be talking about fate and transport in terms of PFAS specific phenomena. And so our discussion on fate and transport will focus on how these small scale processes um At the pore scale can govern transport in soils and groundwater. 00:06:49.000 --> 00:06:58.000 And ultimately the sum of all these tiny small effects can result in large scale transport between environmental media. 00:06:58.000 --> 00:07:12.000 We'll cover it in terms of compound specific chemistry. As well as environment chemistry and how those two influence PFAS transport in the environment. 00:07:12.000 --> 00:07:20.000 Okay. That was kind of what I was going to talk about with that green box. 00:07:20.000 --> 00:07:31.000 Fate and transport. Is the environmental behavior of contaminants as informed by both the characteristics of your contaminant and the characteristics of the site. 00:07:31.000 --> 00:07:36.000 And with PFAS in particular, where we have very unique chemistry. 00:07:36.000 --> 00:07:51.000 We should pay special attention to those PFAS characteristics. And on the left side of the slide here, we see some of the properties of PFAS that influence their behavior in the environment. And we'll be going over some of these in more detail. 00:07:51.000 --> 00:08:03.000 But that's really not the complete story without a solid understanding of how site characteristics on the right can interact with the chemistry to influence this behavior. 00:08:03.000 --> 00:08:16.000 So combining these data sets will ensure that appropriate understanding of fate and transport for PFAS is addressed at your conceptual site model. 00:08:16.000 --> 00:08:37.000 Okay, so here we're going to review some of the properties of perfluoroalkyl acids and Perflueroal acids are a subset of PFAS compounds that we think about that typically have screening limits, even though that list is rapidly expanding. So these are your 00:08:37.000 --> 00:08:51.000 Pfos, PFOA. Pfaas, perfluoral alkyl acids, are canonically understood to be surfactants. And by surfactants, they have a hydrophilic head and a hydrophobic tail. 00:08:51.000 --> 00:08:57.000 And these competing properties are going to drive a lot of their behavior in the environment. 00:08:57.000 --> 00:09:06.000 Particularly what we'll focus on here today is that interfacial interaction. 00:09:06.000 --> 00:09:22.000 Surfactants like interfaces and interfaces and this ionic polar head and hydrophobic tail really drive that interfacial activity so Zooming in. 00:09:22.000 --> 00:09:52.000 Even further. How PFAAs assemble at interfaces It's really important to understanding their partitioning in the Vedo zone which is driven by interfaces, air water interfaces, water-soil interfaces soil air interfaces, you have all these interfaces to consider. And probably the most important one in the Veda zone is that air water interface. This is also a relevant interface to think about with respect to surface water. 00:09:53.000 --> 00:10:21.000 And so… Like we said, these are surfactants and many surfactants are hydrocarbon surfactants, which have a CH bond and Because this is a CF bond and has lower polarizability than that CH bond, we see that affinity for interfaces amplified for PFAS compounds. So it's even more important for PFAS than it is for 00:10:21.000 --> 00:10:32.000 What other surfactants may be doing in the environment. And so while we're focusing on these PFAAs and how they align at these interfaces. 00:10:32.000 --> 00:10:50.000 The other side of the coin is what's going on in the bulk solution. So what is the water chemistry and how does that influence either accumulation at the interface or potential to be in the aqueous phase. 00:10:50.000 --> 00:10:58.000 Okay. So moving on from PFAA specific chemistry. 00:10:58.000 --> 00:11:06.000 And looking at environmental factors or site specific factors that will influence sorption on a micro scale. 00:11:06.000 --> 00:11:27.000 These can largely be lumped into two categories, the electrostatic and hydrophobic interactions. And these govern a lot of site-specific PFAS behavior. So when you're building your CSM or doing your site investigation, these are a good thing to have a handle on to understand how PFAS is going to 00:11:27.000 --> 00:11:37.000 Interact with your environment. So on the left-hand side here, we have a summary of some of the electrostatic interactions that influence partitioning. 00:11:37.000 --> 00:12:03.000 Specifically, the charged clay surfaces or cations in a soil. Or groundwater environment to that matter can absorb to negatively charged head groups so for PFAs. And this is also relevant for fluorotelomer sulfonates and fluorotelomer acids which are polyfluorinated as opposed to perfluorinated. 00:12:03.000 --> 00:12:11.000 And then… In contrast to increased absorption due to charge. 00:12:11.000 --> 00:12:32.000 You can have changes in soil pH. For example, if you've done ISCO at your site, you're going to increase your soil pH and that can potentially decrease your absorption capacity in your soil. So again, knowing the history of your site and your soil chemistry can go a long way 00:12:32.000 --> 00:12:38.000 In defining how PFAS absorption manifests out of sight. 00:12:38.000 --> 00:12:47.000 And then on the other side of the slide here, we have a summary of some of the hydrophobic interactions that can drive PFAS absorption. 00:12:47.000 --> 00:12:52.000 And so the big one we think about is absorption to organic matter. 00:12:52.000 --> 00:13:08.000 Particularly, this is relevant for long chain PFAS, so like your PFOS compounds as opposed to like a PFBS. And this is why we general and so there's stronger absorption with those longer chain compounds. And in general, this is why we see 00:13:08.000 --> 00:13:28.000 Short chain compounds will travel further than long chain longer chained PFAS in groundwater because those longer chain PFAS gets stuck on that organic matter. And so you get somewhat of a chromatographic effect where short chain compounds are just transported further as a function of not being as sticky. 00:13:28.000 --> 00:13:40.000 To organic matter as their longer chain counterparts. With respect to head group chemistry, we also have sulfonates, so your PFSAs. This is your PFOS. 00:13:40.000 --> 00:13:49.000 And they resort more strongly to organic matter as compared to their carboxylic counterparts, so your Always. 00:13:49.000 --> 00:14:02.000 And so if you're looking at two compounds of equal chain length, you can look to their head group chemistry to inform how far they would likely transport relative to each other. 00:14:02.000 --> 00:14:28.000 And while not typically analyzed for the branching or lack thereof of this carbon fluorine chain can also influence adsorption to organic matter. So if you have these linear chains, you would get stronger hydrophobic interaction with organic matter but same compound with a branch chain is going to be less sorptive. 00:14:28.000 --> 00:14:38.000 And again, that's not something typically analyzed for, but it can explain if you're seeing something that doesn't make sense here. 00:14:38.000 --> 00:14:47.000 And so if we're thinking about the chemistry of our PFAS, the chemistry of our water, the chemistry of our soil, it begs the question. 00:14:47.000 --> 00:14:53.000 Which partitioning is most important. Where? 00:14:53.000 --> 00:15:15.000 And there we go. And the good news is it's not everything all the time, everywhere, all at once. The relative importance of a compartment's interfacial behavior varies between our source area, which we largely understood to be unsaturated beta zone. 00:15:15.000 --> 00:15:31.000 Versus the plume or the saturated zone so As we talked about in the beta zone PFAS absorption is largely interfacially driven. So in particular, the PFAS want to be on that air water interface. And this is going to drive 00:15:31.000 --> 00:15:43.000 Much of that mass distribution. On the other hand, in the saturated zone where you don't have an air water interface, it's effectively Zero. 00:15:43.000 --> 00:15:56.000 The sorption to the soil is driving a lot of that behavior. So those electrostatic and hydrophobic effects are driving transport. 00:15:56.000 --> 00:16:09.000 More than interfacial adsorption so You think about different zorptive capacities in different compartments. It's not all the same. 00:16:09.000 --> 00:16:18.000 And so this is so the degree to which these retentive properties can attenuate mass flux is an area of active research. 00:16:18.000 --> 00:16:32.000 And I will direct you to section 5.2 of the TCRAG document, which goes into greater detail, but It is rapidly evolving. 00:16:32.000 --> 00:16:40.000 Okay. And so zooming in on some of these interfaces and how they vary dynamically. 00:16:40.000 --> 00:16:59.000 So if we're thinking about our Vedo zone and um how saturation is not always the same in the Vedo zone. Your air water interfacial area is going to vary with, you know, on a day that it rains versus, you know, a drought. 00:16:59.000 --> 00:17:17.000 Scenario. So you're going to have different degrees of saturation And those different degrees of saturation, as you can see on the left-hand side here, result in different air-water interfacial area, which will in hand result in different sorption capacity. 00:17:17.000 --> 00:17:26.000 And so you can think about this in terms of the periodicity, persistence, and intensity of recharge events. 00:17:26.000 --> 00:17:42.000 But also in this slide, and it's kind of to the right here and not as explicit is There are other factors that affect air water interfacial area, including soil type, grain size, and heterogeneity. 00:17:42.000 --> 00:17:55.000 Again, going back to your conceptual site model, understanding your soil type will get you pretty far with understanding how that soil drains. 00:17:55.000 --> 00:18:13.000 What sort of saturation you can expect and knowing your climate can also give you information about recharge. So all of these are built in to our to our conceptual site model. 00:18:13.000 --> 00:18:18.000 Amen. Okay. 00:18:18.000 --> 00:18:36.000 This kind of summarizes the different components of solid phase absorption that we discussed here. And it is an interplay between the chemistry of your contaminant and the chemistry of your soil or water, right? 00:18:36.000 --> 00:18:43.000 Much of these can be considered dynamic and therefore requires robust site characterization. 00:18:43.000 --> 00:18:59.000 So hydrophobic sorption sites can get filled up quickly by those long chain PFAs and result a chromatographic release of shorter chain PFAs downstream. 00:18:59.000 --> 00:19:05.000 We found that our sulfonates absorb more strongly than our carboxylates. 00:19:05.000 --> 00:19:18.000 And all of this is organic carbon dependent so you can have organic carbon rich soil at the top of your soil column and that can decrease with depth. And so you have a spatial component. 00:19:18.000 --> 00:19:24.000 Or if you've done some organic carbon amendments in your soil, such as biosol. 00:19:24.000 --> 00:19:31.000 Application that can also influence hydrophobic partitioning so the history of your site. 00:19:31.000 --> 00:19:39.000 Moving on to electrostatic interactions, considering changes in pH and ionic strength. 00:19:39.000 --> 00:20:03.000 Either naturally induced or because of some remedial action that has taken over that has occurred at the site can be considered dynamic and then There is one thing that PFAS does that other contaminants do is sorption sites become saturated. So it's a concentration dependent nonlinear 00:20:03.000 --> 00:20:23.000 Absorption behavior. We have that one thing and then that one what we're just barely touching on here and will be discussed in more detail later is precursor transformation can increase the observed PFAAs from your time zero so your time zone zero 00:20:23.000 --> 00:20:31.000 You measure your PFAAs and then you get a bunch of precursor transformation, either biotically or oxidatively. 00:20:31.000 --> 00:20:44.000 And now you're measuring more PFAs well those are going to be kicking off your shorter chain compounds and have implications for environmental fractionation as well. So the transformation effects. 00:20:44.000 --> 00:20:55.000 Anything you do to your soil anything you do to your groundwater, these can all influence partitioning. 00:20:55.000 --> 00:21:08.000 And then finally, bringing it back to where we began, this is a very simplified matrix that kind of shows how the characteristics of our contaminant and the characteristics of our site. 00:21:08.000 --> 00:21:19.000 Can result in either enhanced absorption or reduced absorption and reduced absorption Yeah, with that, I'll hand it off. Thank you. 00:21:19.000 --> 00:21:26.000 Thank you, Dina. And so let me go to the next slide. 00:21:26.000 --> 00:21:39.000 So like Dina mentioned, PFAS can behave as surfactants. As any surfactant, one of the behaviors they can form aggregates. So that means that they're still being entirely dissolved in the bulk phase. 00:21:39.000 --> 00:21:46.000 And they start assembling or performing cell assemblies with each other. 00:21:46.000 --> 00:22:01.000 So at certain higher concentrations, some PFAS can group together, forming the structures that you see on your right. So micelles, there would be a sphere-like structure where you have the perfluorinated tails facing inwards. 00:22:01.000 --> 00:22:13.000 But you can have also hemimycyl, which is half of a micelle, as well as bilayers. So thinking about sticking into almost membrane-like behaviors. 00:22:13.000 --> 00:22:43.000 However, unlike other non-fluorinated surfactants, PFAS that can form these aggregates may have different processes than traditional surfactants. And that is because sometimes in some of the studies, aggregations have been reported Below what is called the critical micelle concentration or the concentration at which you would expect to see the formation of this self aggregates. 00:22:43.000 --> 00:22:52.000 And also there are interactions with charged surfaces like Dean was mentioning that lead to sorption. 00:22:52.000 --> 00:23:10.000 As well as charge surfaces. So it might be a little more complex and there is more research that is needed to understand how self-assembly behavior of PFAS What it means for the transport and their fate in general. 00:23:10.000 --> 00:23:33.000 Now looking at the family of the subfamily of non-polymeric PFAS, we see the majority of them are polyfluorinated compounds, which means that not every single carbon is fully saturated with a fluorine Adam. So many of these precursors can be transformed into what are called terminal perfluorinated alkyl acids. 00:23:33.000 --> 00:23:50.000 Perhaps these are the most common types of PFAS that we've heard. So for example, an example of PFCA is PFOA or perfluorinated octanoic acid, as well as PFSAs like PFOS, perfluorinated octane sulfonate. 00:23:50.000 --> 00:24:03.000 So some of these precursors can be transformed and their chemistry, their structure indicate where they are transforming into either a carboxylic acid and a sulfate. 00:24:03.000 --> 00:24:10.000 And as we are learning more about their transformation pathways, we can use this information for site assessment. 00:24:10.000 --> 00:24:25.000 Characterization and see how different PFAS plume might change with sources, with the age and history, perhaps of remediation of a potential site. 00:24:25.000 --> 00:24:51.000 Here we have an example of polyfluorinated compound. So on the structure on the top, we see a fluorotelomer This means that there is a certain number of carbons that are fully fluorinated. In this case, there's eight carbons that have completely saturated with fluorine. But then next door to those carbons, there's two other carbons that do not have fluorine in them, actually have indicated in the green spheres hydrogen. 00:24:51.000 --> 00:25:01.000 So they don't have fluorine. And so we call this an eight to fluorotelomer and the terminal group is on alcohol. So you see an oxygen and a hydrogen. 00:25:01.000 --> 00:25:24.000 This compound, as you can see, has some carbons that are not fully fluorinated and these carbons that are not fluorinated are open to transformation, either through chemical or biological reactions. So usually most of the transformations that we see in the environment are oxidations. So then these polyfluorinated compound can be oxidized into the structure below. 00:25:24.000 --> 00:25:33.000 The equivalent perforated alkyl acid. In this case, it's a carboxylic acid. So we have here an example formation of PF and A. 00:25:33.000 --> 00:25:56.000 And once this compound is formed. It's neither transformable or biodegradable on its own, which is why we call it a terminal product or a terminal perforated alkyl acid. One thing to note, we go from eight Fluorinated carbons but in the name of the structure below is actually nine because the last carbon is the carbon that has the carboxylic 00:25:56.000 --> 00:26:04.000 Asset group. So that's why it forms PFNA. 00:26:04.000 --> 00:26:12.000 Some of the transformations can occur abiotically. So these are entirely chemical transformations. And most of these reactions are oxidations, not all of them. 00:26:12.000 --> 00:26:27.000 But one of them is we can have fluoropolymers that over time can release Side group chains that maybe some of those side group chains are monomeric PFAS, like in this example that we see in the cartoon in the top right 00:26:27.000 --> 00:26:45.000 There is a side group that is the preferred octano We see these reactions, the release of monomeric PFAS, especially under high pHs, but over decades, we can also see it at neutral pHs. 00:26:45.000 --> 00:27:08.000 Other common chemical reactions that we see in the environment is that hydroxyl radicals that are either formed in natural systems or in the built environment can themselves oxidize polypherinated compounds or precursors into the perfluorinated alkyl acid. So in the bottom right, you see a structure here of another fluorotelomer. 00:27:08.000 --> 00:27:17.000 This one is 62-fluorotelomer sulfonate. You see that after that fully fluorinated chain of carbons indicated by those brackets. 00:27:17.000 --> 00:27:21.000 Then it has two carbons next door that do not have fluorine in them. 00:27:21.000 --> 00:27:33.000 And those carbons are open again for potential oxidation and informs the perfluorinated carboxylic acid The structure on the right and the bottom right that you see is a sulfonamide. 00:27:33.000 --> 00:27:46.000 And that's the full ameth also can be oxidized abiotically with hydroxyl radicals and also would cleave that carbon sulfur bond funding forming perforated carboxylic acids. 00:27:46.000 --> 00:27:52.000 Now, it doesn't mean that whenever we have a reaction of a precursor or polyfluorinated compound. 00:27:52.000 --> 00:27:59.000 That we were always going to end 100% yield into the carboxylic acid or perhaps the sulfonate. 00:27:59.000 --> 00:28:06.000 Sometimes we don't have a complete oxidation reaction. And this is actually quite common in the environment. 00:28:06.000 --> 00:28:11.000 There's an active field of research trying to understand the yield and the kinetics. 00:28:11.000 --> 00:28:25.000 And then lastly, we see also these reactions in the atmosphere. So we have indirect photolysis where we have this fluoroteler alcohol that we were looking at in the previous slide being oxidized by atmospheric reactions. 00:28:25.000 --> 00:28:30.000 Into carboxylic acids. 00:28:30.000 --> 00:28:41.000 Now switching into biological reactions Most of the transformation reactions that lead into the terminal products, the perfluorinated alkyl acids, are oxidations. 00:28:41.000 --> 00:28:50.000 That occur under aerobic conditions. So in the presence of oxygen Here we see two types of structures. In the top part we see fluorotolymers. 00:28:50.000 --> 00:28:56.000 That this one in particular has a very long name, 62FT TAOS. 00:28:56.000 --> 00:29:06.000 Is a component in some firefighting foams. As you can see, we get again the fully fluorinated carbon chain indicating by those brackets, six carbons. 00:29:06.000 --> 00:29:09.000 And then followed by other carbons that do not have fluorine in them. 00:29:09.000 --> 00:29:18.000 Well, it turns out that as we have these transformations occur, we form an intermediate, which is in this case, the six to fluorotolymer sulfonate. 00:29:18.000 --> 00:29:24.000 And eventually that intermediate can be transformed into our terminal products. 00:29:24.000 --> 00:29:44.000 Perfluorinicarboxylic acids. And under biological conditions, we often don't form one equivalent perfluorinated carboxylic acids from the fluorotelomer we actually form a normal distribution, usually anywhere from n to n minus 2. So n being the number of carbons. 00:29:44.000 --> 00:29:53.000 Of the carboxylic acid. So here we see an example where we have six carbon perfluorinated carbon precursor, the S62FTAO. 00:29:53.000 --> 00:30:02.000 Leading into the C4, C5, C6, and C8. Perfluorinated carboxylic acid. 00:30:02.000 --> 00:30:12.000 In the bottom figure, we see an example of another type of structure. This structure is called a sulfonyamide, and it's also a kind of precursor. 00:30:12.000 --> 00:30:18.000 Now, indifference to the fluorotolomers, here we see that we get the fully fluorinated chain. 00:30:18.000 --> 00:30:36.000 And then next door, we don't get another carbon. We actually get a sulfur. So actually for biological reactions that typically leads to mainly the formation of perifluorinated sulfonates. So we see that the final, the terminal product of these biological reactions tends to be perfluorinated 00:30:36.000 --> 00:30:45.000 Hexane sulfonate. This is the six carbon equivalent of PFOS. 00:30:45.000 --> 00:31:08.000 Under anaerobic conditions, there's fewer studies that have looked at these conditions. And the reason is that usually we don' in experimental time scales, enough evidence of transformation all the time. There is some transformation and whenever there is that transformation, it doesn't always lead into the perfluorinated carboxylic acids or the perfluorinated alkyl acids. 00:31:08.000 --> 00:31:22.000 So here we have again in the left an example structure of a precursor found in firefighting foam that only under nitrate reducing conditions can lead eventually into the formation of carboxylic acids. 00:31:22.000 --> 00:31:31.000 Just like the aerobic conditions in presence of oxygen. However, when we go into more reducing conditions, think about sulfate reduction, for example. 00:31:31.000 --> 00:31:40.000 We don't necessarily form the carboxylic acids and sometimes we form other kind of precursors as well. And the reactions are incomplete. 00:31:40.000 --> 00:31:53.000 One example where we see this is in landfills, where oftentimes when intermediate that is commonly detected is the 5, 3 fluorotelomer carboxylic acid or 5-3-FTCA. 00:31:53.000 --> 00:32:03.000 And this is one example of places where we might have anaerobic conditions that do not always lead into perfluorinated carboxylic acids. 00:32:03.000 --> 00:32:19.000 Some recently, there has been reported an anaerobic reaction under a process called Pheamox is the anaerobic reduction of iron coupled to ammonium oxidation by the acid in microbiome species strain A6. 00:32:19.000 --> 00:32:24.000 Which has reported defluorination of PFOA and PFOS with release of fluoride. 00:32:24.000 --> 00:32:38.000 This is an active area of research. Now with that, I will turn it over to Robert for the next slide. 00:32:38.000 --> 00:32:53.000 Thank you, Chris. Great information there. So now that Dina and Chris have reviewed what's happening on a small scale in the subsurface to affect P-phosphate and transport. 00:32:53.000 --> 00:32:58.000 We'll take some time now to discuss the larger scale processes. 00:32:58.000 --> 00:33:04.000 And how those relate to those microscale processes that are happening. 00:33:04.000 --> 00:33:19.000 So we'll use this figure here, this model CSM essentially. To walk through various media Touching briefly on air, soil, groundwater, and surface water, and then sediment. 00:33:19.000 --> 00:33:32.000 The black text on the figure shows the types of processes that we need to think about for each media type. 00:33:32.000 --> 00:33:45.000 So when we're talking about large scale transport in air. The sources that we are concerned about might be things like stack emissions on industrial sites. 00:33:45.000 --> 00:33:51.000 But we can also have transport in air if there are volatile PFAS present. 00:33:51.000 --> 00:33:59.000 Or if PFAS are aerosolized, for example, during fire training or response. 00:33:59.000 --> 00:34:11.000 So… Vapor phase or aerosolized PFAS can undergo transformation while in the atmosphere due to processes such as photo oxidation. 00:34:11.000 --> 00:34:29.000 And non-volatile PFAS and air are usually being transported on particulates. They're attached to particulates that are being transported So short range transport is uh is common at industrial sites. 00:34:29.000 --> 00:34:37.000 Wet or dry deposition can occur a short range from those industrial sites with stack emissions. 00:34:37.000 --> 00:34:41.000 And it can result in large diffuse areas of contaminated soil. 00:34:41.000 --> 00:34:50.000 Which in some cases can leach to groundwater. And there have also been studies that have suggested longer range transport due to static emissions. 00:34:50.000 --> 00:35:00.000 It's still an active area of research. More information on atmospheric transport of PFAS can be found in section 5.2. 00:35:00.000 --> 00:35:11.000 0.4 and section 5.4 of the online guidance document. 00:35:11.000 --> 00:35:27.000 So when we're looking at the Vado zone. There are… that there may be, well, there's a lot to consider here. So Dina talked about partitioning to solid phases being dependent on electrostatic interactions with minerals and cations. 00:35:27.000 --> 00:35:46.000 As well as interaction with organic material through hydro hydrophobic interactions And we touched briefly on interfacial behavior. Chris talked about precursor transformation so Let's think about the factors that affect these processes on a larger scale here. 00:35:46.000 --> 00:35:54.000 Our Vedo zone tends to be aerobic, which is ideal for precursor transformation, as you heard Chris talk about. 00:35:54.000 --> 00:36:01.000 As strongly absorbed cations and zwitterions transform. 00:36:01.000 --> 00:36:10.000 So cationic PFAS, many precursors are cationic or zwitterionic. And as they transform. 00:36:10.000 --> 00:36:19.000 Psaas, which are anionic, are formed then that might lead to desorption from soils because of different interactions. 00:36:19.000 --> 00:36:30.000 And then leaching to groundwater. So it can result in long-term sources if there's retention in the Vedo zone long-term sources to groundwater. 00:36:30.000 --> 00:36:44.000 So site-specific retention in the Vedosone is dependent on all those factors we talked about As well as things like precipitation and infiltration rates and other processes that might occur at specific sites. 00:36:44.000 --> 00:36:49.000 Gina talked about how partitioning to air-water interfaces can have important implications. And again. 00:36:49.000 --> 00:36:54.000 Precipitation and infiltration rates are going to matter for that as well. 00:36:54.000 --> 00:37:10.000 So those infiltration rates, the soil porosity and particle size and biogeochemistry All of those are going to play a role in the extent to which these microscale processes are affected. 00:37:10.000 --> 00:37:27.000 One important thing to note as well when we're talking about Vadosone is that this is where we tend to find plant roots and animal burrows right so that that means that contamination that gets retained in the beta zone could be subject to uptake by biota. 00:37:27.000 --> 00:37:34.000 We're not covering that pathway in this presentation. But it is covered in the biosolids soil leaching and mobility beyond the basics training. 00:37:34.000 --> 00:37:39.000 Which also goes into further detail about Vadosone transport in general. 00:37:39.000 --> 00:37:45.000 Check that out. There's archived recordings available. 00:37:45.000 --> 00:37:58.000 Now, if we do, if PFAS does leach to groundwater Once we're there, transport in the groundwater is expected to be faster relative to transport through the VATO zone. 00:37:58.000 --> 00:38:03.000 But again, this is going to be dependent on the characteristics of the PFAS in groundwater. 00:38:03.000 --> 00:38:09.000 Including things like the chain length and the head group when we're talking about PFAAs. 00:38:09.000 --> 00:38:17.000 And remember, anything affecting geochemistry can impact their transport. 00:38:17.000 --> 00:38:29.000 To understand the partitioning behavior of PFAS in groundwater We need the same types of information that we just talked about for understanding sorption and leaching from the Vedo zone. 00:38:29.000 --> 00:38:37.000 Plus information on groundwater chemistry Things like pH, cation concentration, redox conditions, and so forth. 00:38:37.000 --> 00:38:44.000 We want to consider if there's co-contaminants present. And other parameters that might affect sorption. 00:38:44.000 --> 00:38:51.000 If co-contaminants are present. Oftentimes remediation may have occurred targeting those contaminants. 00:38:51.000 --> 00:38:58.000 Especially if it's a site that's been around for a while and before we were really paying a lot of attention to PFAS. 00:38:58.000 --> 00:39:05.000 And if that remediation has occurred, as Dina mentioned, that can cause changes in our solution chemistry. 00:39:05.000 --> 00:39:12.000 For example, if remediation has driven oxidation. It might increase the rates of precursor transformation. 00:39:12.000 --> 00:39:36.000 Increasing the ratios of PFAAs as a result. Might also facilitate transport or retired transport by changing pH or other geochemical parameters depending on the specific type of remediation that has occurred. 00:39:36.000 --> 00:39:48.000 Other factors that are important to consider and understand with groundwater include, of course, groundwater flow direction and velocity and factors that affect it, such as groundwater extraction or dewatering. 00:39:48.000 --> 00:40:02.000 We want to think about preferential flow pathways and hydrological connections with nearby surface water bodies. 00:40:02.000 --> 00:40:17.000 Now, if we are looking at PFAS, transport and surface water there's the type of water body is of course an important in considering how PFAS might move. 00:40:17.000 --> 00:40:31.000 Not only the type of the water body, but how it's recharged or sustained and whether it's whether it is recharged by groundwater or whether it lends some of its lends some of some of itself to groundwater. 00:40:31.000 --> 00:40:41.000 So slow moving water and fast moving water, for example, are going to have different concerns, right? The extent to which sediments or particulate matter are transported will be different. 00:40:41.000 --> 00:40:46.000 We have precipitation events. When we're looking at surface water. 00:40:46.000 --> 00:40:53.000 They may dilute PFAS, an important consideration for timing of sampling. 00:40:53.000 --> 00:41:03.000 Or it might cause additional transport to PFAS of PFAS to water bodies through runoff or storm drains. 00:41:03.000 --> 00:41:15.000 Perhaps a particulate transport as well. Fish and other biota of course can uptake PFAS. It does bioaccumulate by binding to proteins. 00:41:15.000 --> 00:41:21.000 And there's potential for biomagnification of the food chain as a result. 00:41:21.000 --> 00:41:30.000 Really important consideration for surface water is that that tendency of PFAS to aggregate at the air-water interface. 00:41:30.000 --> 00:41:39.000 The concentrations can be very high at the surface And there's often stratification throughout the water column as a result. 00:41:39.000 --> 00:42:00.000 One interesting thing that can happen is with wind or turbulence, PFAS in this high concentration surface micro layer can form foams. It's kind of similar to how a triple f works you mix water, PFAS, and air or agitation and you can create 00:42:00.000 --> 00:42:12.000 Foam but this is it's important to note though this is distinct from a triple F foam. This is a different phenomenon this is not A triple F, just a similar process for creating that foam. 00:42:12.000 --> 00:42:29.000 And then foams can float downstream Or… end up on shores, which can spread PFAS kind of create new areas of contamination and different surface water bodies or in soils or sediments. 00:42:29.000 --> 00:42:44.000 So when we're talking about foam Again, we can see these foam islands get transported to a new location and can travel downstream or be blown across a lake. 00:42:44.000 --> 00:43:02.000 Or other water body depending on, of course, the flow rate and the wind action once it is transported that way, it can collapse back into the water column or if it's onshore, it can collapse and create an area of soil contamination. 00:43:02.000 --> 00:43:17.000 And it's really critical to understand that the amount of PFAS that we find in foams tends to be enriched compared to the underlying water column, often by multiple orders of magnitude. 00:43:17.000 --> 00:43:34.000 There are also exposure concerns to consider with foams. You know can imagine uh dogs, pets, kids, adults playing in the foams when they find them. 00:43:34.000 --> 00:43:51.000 So there's a lot that we still need to study about that exposure route, but it is something to consider if you're thinking about conceptual site model out of sight. 00:43:51.000 --> 00:44:00.000 Um so One thing that's important to consider is transport between surface water and groundwater. It can go either way. 00:44:00.000 --> 00:44:04.000 Either from groundwater to surface water or surface water to groundwater. 00:44:04.000 --> 00:44:22.000 But this is kind of an active area of research We do have some information from peer reviewed articles and case studies that can help us to understand how PFAS behave in this transition zone, the hyperic Just that transition between 00:44:22.000 --> 00:44:31.000 Groundwater and surface water So, for example, groundwater is often, not always, but often it's anoxic or reducing. 00:44:31.000 --> 00:44:48.000 And surface water in general tends to have a higher concentration of dissolved oxygen Or there might be changes in the dissolved organics, the types and concentrations of ions in your groundwater or surface water systems. 00:44:48.000 --> 00:44:56.000 And other changes that can happen over that. Between surface water and groundwater. 00:44:56.000 --> 00:45:09.000 So that increasing oxygen that may occur, as well as the potential for increased biological activity can lead to faster rates of precursor transformation. 00:45:09.000 --> 00:45:25.000 And some studies have documented increased proportions of PFAs and surface water bodies compared to gradient groundwater, which might indicate that precursor transformation is occurring. 00:45:25.000 --> 00:45:37.000 So the differences in the organic or mineral content of saturated soils versus sediments of a water body can also have important implications. 00:45:37.000 --> 00:45:46.000 If we see higher cations or increased salinity. That can lead to increased sorption As can increased organic content. 00:45:46.000 --> 00:45:56.000 At PFAS, sorb 2 sediments Depending, again, on the what type of water body you have, how fast it moves and so forth. 00:45:56.000 --> 00:46:13.000 Those sediments can then be transported downstream. Or there could be uptake of PFAS by the biota that live in those sediments. 00:46:13.000 --> 00:46:27.000 So we also want to consider co-contaminants when we're thinking about PFAS. This is pretty common to see co-contamination particularly at A triple F sites, but of course it can occur at other PFAS sites as well. 00:46:27.000 --> 00:46:37.000 Pfas might partition into LNAPL or they can accumulate, again, at interfaces, including the water NAPL interface. 00:46:37.000 --> 00:46:43.000 That might lead to greater retention of PFAS if you have petroleum present. 00:46:43.000 --> 00:46:50.000 Petroleum contamination also will lead to reducing conditions Right. 00:46:50.000 --> 00:47:04.000 Anaerobic reducing conditions, which can slow down precursor transformation and change the pathways from you know aerobic pathways to more anaerobic processes and pathways. 00:47:04.000 --> 00:47:12.000 Which might, as Chris mentioned earlier, result in different products being formed during that transformation. 00:47:12.000 --> 00:47:19.000 And again, I think we've hit on this a few times and we'll hit on it a few more, but if petroleum remediation has occurred. 00:47:19.000 --> 00:47:26.000 It can alter those redox conditions to a more oxidizing conditions. 00:47:26.000 --> 00:47:43.000 For example, if air sparge has occurred, or it might change the ionic concentrations and that can change the rates of both precursor transformation in the case of oxidizing conditions or retention of PFAS when we change the ionic concentrations. 00:47:43.000 --> 00:47:57.000 It is important. It's maybe a little less studied but dean apple co-contamination can also result in increased PFAS retention. 00:47:57.000 --> 00:48:22.000 So using this hypothetical hypothetical made-up example of an A triple f site We can use this kind of cartoon, this diagram to talk about how co-contamination affects PFAS migration in groundwater And really, just more generally how differential transport can occur. 00:48:22.000 --> 00:48:36.000 When you have a mix of different PFAS. So on the left side of this cartoon, we have our fire training area That's our source area we have lnapple present. It's anaerobic and reducing conditions. 00:48:36.000 --> 00:48:52.000 We have highly concentrated PFAS and petroleum constituents There may be some retention of some of the PFAS in the beta zone as well as the air water interface. 00:48:52.000 --> 00:48:58.000 And depending on what's going on in the subsurface, you might have retention in the saturated zone as well. 00:48:58.000 --> 00:49:21.000 Groundwater is flowing in this example from left to right And the oxidation reduction potential is changing along the axis of the petroleum loom so more reducing conditions near the source And more oxidizing conditions the further away you get from the source. 00:49:21.000 --> 00:49:33.000 And that, of course, as mentioned earlier, is due to biodegradation of petroleum and the consumption of oxygen and other favorable electron receptors that occurs during that biodegradation. 00:49:33.000 --> 00:49:40.000 Near the source area. We may or may not have elm apple. 00:49:40.000 --> 00:49:53.000 If it is present, it may slow down PFAS migration. Particularly at the water nappal interface We also would tend, like I said, we have reducing conditions there generated by consumption of oxygen. 00:49:53.000 --> 00:50:00.000 Slowing precursor transformation, potentially increasing retention due to associated changes in groundwater chemistry. 00:50:00.000 --> 00:50:09.000 But as we move from the source, we start to see lower concentrations of hydrocarbons Conditions that are less reducing, just as is common with petroleum plumes. 00:50:09.000 --> 00:50:14.000 And then precursor transformation rates might increase as we shift to aerobic pathways. 00:50:14.000 --> 00:50:19.000 That's if the precursors got that far in the first place, right? 00:50:19.000 --> 00:50:27.000 Beyond the petroleum plume, oxygen will tend to be higher. Precursors that have traveled this far will break down more readily. 00:50:27.000 --> 00:50:37.000 And then due to the differential transport of short versus long chains, we'll also start to see a higher proportion of shorter chain PFAAs relative to the source area. 00:50:37.000 --> 00:50:41.000 And branched BFAAs will also be enriched compared to the source. 00:50:41.000 --> 00:50:52.000 Relative to linear isomers. Again. 00:50:52.000 --> 00:51:00.000 This is dependent on the properties of the molecules, right? So shorter chain PFA tend to migrate faster. They sorb less to soils. 00:51:00.000 --> 00:51:12.000 So you're going to see an increased proportion of those relative to total PFAS. I'm talking proportions here, not concentrations. Important to note that difference. 00:51:12.000 --> 00:51:20.000 And again, branched isomers, as Dina mentioned, also tend to travel faster and farther than their linear counterparts. 00:51:20.000 --> 00:51:27.000 So in addition to increased short chain BFAAs, we might see increased proportions of branched isomers. 00:51:27.000 --> 00:51:43.000 If they are present and measured. All of this we refer to this process as environmental fractionation as the fraction of specific PFAS species that we see down gradient is often different from the fraction at the source. 00:51:43.000 --> 00:51:57.000 In other words, we should not expect the composition of PFAS in the source area to be the same as the composition or relative proportions down gradient. You get a rarefied set of those original constituents. 00:51:57.000 --> 00:52:03.000 And that's important to note when we talk about things like site characterization. 00:52:03.000 --> 00:52:11.000 And the use and application of forensics, which we'll get to a little bit later in the presentation. 00:52:11.000 --> 00:52:25.000 At this point, we're going to switch a little bit to site characterization, and I'm going to turn this over to Ted Campbell. 00:52:25.000 --> 00:52:33.000 All right. Thank you, Robert. That was an excellent presentation. And really tees up you know. 00:52:33.000 --> 00:52:40.000 The material that I'm going to cover. Which is key components of PFAS site characterization. 00:52:40.000 --> 00:52:48.000 Dina, Chris, and Robert have covered a number of details on PFAS chemistry, fate, and transport. 00:52:48.000 --> 00:52:57.000 They explained how the physical properties of PFAS And site characteristics affect retention and mobility. 00:52:57.000 --> 00:53:04.000 This information is a perfect foundation for our discussion of site characterization of PFAS. 00:53:04.000 --> 00:53:20.000 So as you'll see, these themes inform how we think about specific information needed for site characterization. 00:53:20.000 --> 00:53:27.000 So in this section of the training, we'll look at common source types and their associated transport pathways. 00:53:27.000 --> 00:53:33.000 We'll talk about how the hydrogeologic setting informs our understanding of PFAS sites. 00:53:33.000 --> 00:53:39.000 And what kind of information should be gathered to understand PFAS transport? 00:53:39.000 --> 00:53:46.000 We'll also look at a few special considerations related to the sampling and analysis of PFAS. 00:53:46.000 --> 00:53:54.000 Note that in the interest of time, we won't go into a lot of details on sampling analysis, but we will drop a link in the chat box. 00:53:54.000 --> 00:54:09.000 To an excellent archived ITRC training on this very topic. And we'll look at methods for interpreting and visualizing data to help us understand our conceptual site model. 00:54:09.000 --> 00:54:14.000 Now, before we get started, let's recall the purpose of a site characterization. 00:54:14.000 --> 00:54:27.000 Which is to define the nature, sources, and extent of contamination, understand transport pathways and identify potential receptors. 00:54:27.000 --> 00:54:45.000 This information is needed for risk assessment and remediation planning. With that in mind, let's now look at site characterization of PFAS. 00:54:45.000 --> 00:54:54.000 Okay, so site characterization usually begins with the development of an initial conceptual site model or CSM for short. 00:54:54.000 --> 00:55:04.000 Csm pulls together our conceptual understanding of site conditions including PFAS releases, hydrogeologic setting. 00:55:04.000 --> 00:55:17.000 Transport characteristics. Exposure pathways And it's here that we want to focus on site-specific characteristics and properties of PFAS or mixtures at the site. 00:55:17.000 --> 00:55:22.000 Which, taken together will affect Titan transport, as we're hearing. 00:55:22.000 --> 00:55:29.000 One important use of the CSM is that it helps us define our data needs. 00:55:29.000 --> 00:55:40.000 Including numbers, types, and data quality objectives. It's often appropriate to develop an initial CSM prior to or perhaps as part of? 00:55:40.000 --> 00:55:52.000 Preparing the site investigation work plan And we'll discuss this more in later slides. 00:55:52.000 --> 00:56:00.000 So just to quickly review, the ITRC PFAS document identifies four primary types of PFAS sources to the environment. 00:56:00.000 --> 00:56:07.000 And they're listed here. We'll briefly run through some generalized CSMs. 00:56:07.000 --> 00:56:15.000 For these source types and discuss some of the important differences or unique characteristics of each. 00:56:15.000 --> 00:56:21.000 Now, ITRC has identified these as the most common four types of PFAS sites. 00:56:21.000 --> 00:56:28.000 But it's important to note that there are potentially other sources of PFAS besides these. 00:56:28.000 --> 00:56:40.000 Additionally, biosolids from wastewater treatment plants maybe land applied in agricultural settings So this would represent a unique PFAS source type. 00:56:40.000 --> 00:57:00.000 Now, we won't touch much on biosolids in my session, but a recent ITRC training focused specifically on this topic And we will drop a link to it in the chat box. 00:57:00.000 --> 00:57:09.000 So here's the first of three generalized CSMs for PFAS. This one shows a fire training and response site. 00:57:09.000 --> 00:57:15.000 These are sites that are associated with A triple F, aqueous film forming foam. 00:57:15.000 --> 00:57:20.000 It contains high concentrations of PFAS. For fire suppression. 00:57:20.000 --> 00:57:27.000 These sites also contain a hydrocarbon fuel source, which co-mingles with PFAS released at the site. 00:57:27.000 --> 00:57:38.000 Importantly, fire training sites often are used over and over, which multiplies the A triple F and PFAS that may be released at these sites. 00:57:38.000 --> 00:57:51.000 Now, because of the high solubility, mobility, and persistence of some PFAS, Careful attention must be paid to any pathway that may transport PFAS from the release area. 00:57:51.000 --> 00:58:02.000 This might include surface water runoff. Infiltration of groundwater and groundwater transport to supply wells or surface waters. 00:58:02.000 --> 00:58:10.000 Multiple investigations have reported long-range transport of PFAS in surface water from A triple F releases. 00:58:10.000 --> 00:58:19.000 In some cases, even secondary groundwater plumes have been created miles away from the original source. 00:58:19.000 --> 00:58:28.000 This occurred, for example, at a site in Minnesota where surface water infiltrated the groundwater at a distant downstream location. 00:58:28.000 --> 00:58:44.000 And it resulted in a very large secondary groundwater plume. It's not known how significant the air emission pathway may be at a triple F sites. 00:58:44.000 --> 00:58:49.000 So to think a little bit more about this in the context of site characterization. 00:58:49.000 --> 00:58:59.000 One important thing to note about a triple F impacted sites is the highly diverse mixtures of PFAS that are commonly found. 00:58:59.000 --> 00:59:07.000 And the differences in PFAS composition that result from different manufacturing methods over time. 00:59:07.000 --> 00:59:16.000 For example, PFAS are commonly produced by two manufacturing processes. Electrochemical fluorination, ECF. 00:59:16.000 --> 00:59:22.000 And fluorotelerization. Ecf dominated earlier manufacture. 00:59:22.000 --> 00:59:31.000 But starting in the late 90s and early 2000s. Multiple manufacturers began using fluorotelerization. 00:59:31.000 --> 00:59:49.000 And with the phase out of PFOS, manufacturing in the early 2000s, newer foam formulations have lower proportions of long chain PFAS and higher proportions of PFAS with six or fewer carbons. 00:59:49.000 --> 01:00:00.000 Note that many manufacturers have begun shifting to fluorine-free foams. 01:00:00.000 --> 01:00:06.000 Now let's touch on forensic analysis of PFAS from a triple F.'s. 01:00:06.000 --> 01:00:14.000 This is based on fingerprinting the composition of unknown samples. And comparing them with published fingerprints. 01:00:14.000 --> 01:00:23.000 However, caveats warranted. Precursor transformations can over time alter that original fingerprint. 01:00:23.000 --> 01:00:33.000 And there's batch variability. And impurities associated with the manufacturing process that can also affect that original fingerprint. 01:00:33.000 --> 01:00:38.000 Now, because of the complexities and variability of A triple F sources. 01:00:38.000 --> 01:00:50.000 And their associated environmental behaviors. We'll want to use multiple lines of evidence and analytical methods for fingerprinting. 01:00:50.000 --> 01:00:55.000 We should consider each site based on its unique features, chemical data. 01:00:55.000 --> 01:01:06.000 And historical information. So all this is just a teaser. We have an entire forensic section that will be covered by Skyler Sorsby later in the presentation. 01:01:06.000 --> 01:01:17.000 Just wanted to briefly mention it here in the context of site characterization. 01:01:17.000 --> 01:01:30.000 All right, let's shift now to industrial sites. Here we show a conceptualized CSM with pathways that you may want to consider when characterizing an industrial PFAS source. 01:01:30.000 --> 01:01:38.000 The actual PFAS released and the pathways they follow depends, of course, on the type of industry involved. 01:01:38.000 --> 01:01:44.000 Note that air emissions can be a significant pathway for industrial sources. 01:01:44.000 --> 01:01:56.000 Even though terminal perfluoroalkyl acids are not considered to be volatile, they can be released from stacks either in vapor droplets or bound to particles. 01:01:56.000 --> 01:02:03.000 Volatile precursors, however, may be emitted and transformed to terminal perfluoroalkylal acids. 01:02:03.000 --> 01:02:21.000 That are then deposited across potentially large areas. In addition, discharges to surface water from groundwater stormwater runoff and effluent can spread contamination long distances from these source facilities. 01:02:21.000 --> 01:02:29.000 A special case of this is PFAS ending up in downstream drinking water intakes and associated supply lines. 01:02:29.000 --> 01:02:40.000 Which can leak and infiltrate PFAS to groundwater. Resulting in very large and distant discontiguous PFAS footprints. 01:02:40.000 --> 01:02:49.000 This, in fact, occurred at a PFAS manufacturing facility in North Carolina and resulted in a four-county PFAS footprint. 01:02:49.000 --> 01:03:00.000 Far downstream of the site. 01:03:00.000 --> 01:03:12.000 A shift now to landfills and wastewater treatment plants. This CSM is a generalized assessment of key PFAS pathways at landfills and treatment plants. 01:03:12.000 --> 01:03:17.000 Landfills are aggregators. Of society's waste. 01:03:17.000 --> 01:03:25.000 Widespread use of pfos and a host of commercial products means they're likely present in nearly all landfills. 01:03:25.000 --> 01:03:31.000 Their concentrations will depend on the particular waste streams that have entered that landfill. 01:03:31.000 --> 01:03:38.000 Higher concentrations would be expected in areas with industrial waste sources. 01:03:38.000 --> 01:03:46.000 Note that landfills and wastewater treatment plants commonly have a mutual dependence. What do I mean by that? 01:03:46.000 --> 01:03:53.000 Most landfills truck or discharge their leachate to treatment plants via sewer systems. 01:03:53.000 --> 01:04:01.000 This, along with other sources, may contribute to the presence of PFAS in treatment plant effluent. 01:04:01.000 --> 01:04:13.000 And biosolids. The biosolids often end up back in the landfills where the PFAS can once again enter landfill each egg So it can form a sort of cycle. 01:04:13.000 --> 01:04:31.000 Note that the oxidation and digestion processes in treatment plants can enhance precursor transformation, resulting in some cases to greater measurable PFAS concentrations In the effluent than that of the influent. 01:04:31.000 --> 01:04:42.000 So one other note, PFAS have been detected at elevated concentrations say parts per billion in runoff from composting sites in the United States. 01:04:42.000 --> 01:04:56.000 Indicating that even residentially sourced paper products can create PFAS contaminated leachate. 01:04:56.000 --> 01:05:10.000 So just to recap, we've covered source specific considerations in a bit of detail and We see that A triple F sources can be a diverse mix of PFAS that are highly dependent on the manufacturing methods. 01:05:10.000 --> 01:05:19.000 Of that a triple f of that We also see that fate and transport can be affected by co-contaminants. 01:05:19.000 --> 01:05:29.000 For an industrial source, PFAS composition may be less diverse. And will depend on the type of PFAS produced or used by that facility. 01:05:29.000 --> 01:05:35.000 Release mechanisms will be site-specific. And depending on facility operations. 01:05:35.000 --> 01:05:46.000 Short or long range air transport. Might be an important pathway to consider when performing a site characterization. 01:05:46.000 --> 01:05:55.000 For a landfill source, PFAS composition and concentrations will vary depending on the wastes accepted. 01:05:55.000 --> 01:06:06.000 Leechate will tend to be predominated by PFAS with fewer than eight carbons because they're less hydrophobic and therefore more likely to be in the aqueous phase. 01:06:06.000 --> 01:06:12.000 In particular, 5.3 floor telomer carboxylic acid FTCA. 01:06:12.000 --> 01:06:23.000 Is a common and often dominant PFAS found in landfills. Pfas may also be released to air from landfills. 01:06:23.000 --> 01:06:33.000 Predominantly as a floor telomer alcohols, FTOHs, and perfluorobutanoate. Pfba. 01:06:33.000 --> 01:06:42.000 So for treatment plants and biosolids, the PFAS mix will depend on the PFAS entering the plant. 01:06:42.000 --> 01:06:57.000 And the plant's treatment processes. Biological and chemical transformation of precursor compounds can create intermediate products and terminal perfluoraalkyl acids. 01:06:57.000 --> 01:07:06.000 These generalized CSMs for common source types provide a context for developing our site-specific conceptual site models. 01:07:06.000 --> 01:07:19.000 And these CSMs are prelude to, and I'll say an integral part of the site characterization process. 01:07:19.000 --> 01:07:27.000 So let's look briefly at our site-specific CSM and see how it fits into our site characterization. 01:07:27.000 --> 01:07:35.000 Shown here, some basic elements of the CSM. For starters, prior to or even part of an initial site visit. 01:07:35.000 --> 01:07:40.000 It's good to gather information on historical use of PFAS at the site. 01:07:40.000 --> 01:07:51.000 What type of PFAS site are we dealing with? What are current and historical site operations, chemical inventories, waste handling processes? 01:07:51.000 --> 01:08:04.000 Are there permits in place? These help us understand what types of PFAS mixtures and potential commingled contaminants to look for and where to look. 01:08:04.000 --> 01:08:09.000 We also need to understand site features that impact fate and transport. 01:08:09.000 --> 01:08:18.000 Including certain hydrogeologic details and surface features that may result in runoff to surface waters. 01:08:18.000 --> 01:08:23.000 What are the nature of surface water features and their capacity to convey PFAS downstream? 01:08:23.000 --> 01:08:32.000 How might rainfall patterns affect runoff and surface water transport? His groundwater impacted and to what extent? 01:08:32.000 --> 01:08:40.000 How susceptible are the nearest supply wells based on local geology, proximity, and pumping characteristics? 01:08:40.000 --> 01:08:48.000 If atmospheric deposition, if atmospheric releases have occurred, do we understand wind and deposition patterns? 01:08:48.000 --> 01:09:06.000 In short, have we identified all susceptible receptors? One reminder, it's critical during site characterization of PFAS that the nearest groundwater and surface water drinking water supplies be identified. 01:09:06.000 --> 01:09:10.000 So all of this will drive how and where to sample. 01:09:10.000 --> 01:09:19.000 To evaluate the nature and extent of PFAS impacts to soil groundwater, surface water, sediment, and air. 01:09:19.000 --> 01:09:25.000 Note also that it's important that we identify at the outset any permit limits. 01:09:25.000 --> 01:09:33.000 Or federal or state regulations or concentration thresholds that apply to the site. 01:09:33.000 --> 01:09:40.000 And finally, we want to update and improve our CSM as new information becomes available. 01:09:40.000 --> 01:09:58.000 We will lean on it to understand additional data needs throughout site characterization risk assessment and remediation planning. 01:09:58.000 --> 01:10:08.000 So let's now take a look at some key data needed to understand transport pathways, focusing here on releases to groundwater. 01:10:08.000 --> 01:10:29.000 And factors that may affect migration or retention. Any discussion of groundwater transport should start with the Vedos zone and factors that may inhibit or enhance movement of PFAS through this zone We'll cover some of these factors in the next slide. And of course, we've heard quite a bit about this 01:10:29.000 --> 01:10:34.000 In the sessions from the prior speakers. 01:10:34.000 --> 01:10:41.000 So focusing on groundwater we will want to understand how and where PFAS will move. 01:10:41.000 --> 01:10:49.000 Through advection, dispersion, and potential retention. Including matrix diffusion. 01:10:49.000 --> 01:10:56.000 At the most basic level, we will need to understand groundwater flow directions and gradients based on water level measurements. 01:10:56.000 --> 01:11:07.000 And we'll want to assess potential subsurface heterogeneities. Do we have soil layering with clays or perhaps fractured bedrock or karst? 01:11:07.000 --> 01:11:21.000 How permeable are these materials? Here we would rely on, for example, slug or aquifer testing Note that perfluoroalkyl acids tend to move readily once they enter groundwater. 01:11:21.000 --> 01:11:28.000 But certain factors can affect this, as we heard earlier. Including the PFAS compound. 01:11:28.000 --> 01:11:42.000 Soil type, pH, total dissolved solids, co-contaminants, and others. 01:11:42.000 --> 01:11:48.000 Let's now look at what we may want to analyze to understand the movement of PFAS through the VATO zone. 01:11:48.000 --> 01:11:56.000 Of course, site-specific and climate related factors like precipitation, depth to water. 01:11:56.000 --> 01:12:05.000 Infiltration rates will affect the rate the system flushes PFAS or most any contaminant for that matter. 01:12:05.000 --> 01:12:15.000 Also, soil properties like surface charge, pH, carbon content generally will affect the transport of most contaminants through the Vedos zone. 01:12:15.000 --> 01:12:20.000 That said, there are a number of unique attributes specific to PFAS. 01:12:20.000 --> 01:12:30.000 That affect its movement through the Vedo zone. Chief among them the propensity for PFAS compounds to accumulate at the air-water interface. 01:12:30.000 --> 01:12:36.000 Information collected as part of the site characterization should account for these. 01:12:36.000 --> 01:12:42.000 For example, we may want to analyze soil porosity organic carbon content. 01:12:42.000 --> 01:12:50.000 Bulk density, ionic composition. Anion and cation exchange capacity. 01:12:50.000 --> 01:13:01.000 Grain size, mineralogy, water content. And we would want information on subsurface geochemistry, including ionic strength and composition. 01:13:01.000 --> 01:13:07.000 Dissolved oxygen. Oxidation reduction conditions and pH. 01:13:07.000 --> 01:13:16.000 Taken together, this information helps us evaluate how PFAS would be expected to migrate through the Vedosome. 01:13:16.000 --> 01:13:35.000 Note that we may see pulses of PFAS concentrations entering groundwater with time based on flushing events and PFAS properties like chain links. 01:13:35.000 --> 01:13:49.000 So another thing we want to be aware of during site characterization is that PFAS sampling is different from other sampling events due to an increased potential for inadvertent sample contamination. 01:13:49.000 --> 01:13:55.000 So what I mean by contamination here is the unintentional introduction of PFAS into your sample container. 01:13:55.000 --> 01:14:03.000 During sample collection, shipping, or storage. This can occur due to three main reasons. 01:14:03.000 --> 01:14:11.000 First, some sampling equipment and materials may contain PFAS. Shown here are a few. 01:14:11.000 --> 01:14:17.000 Second, our samples are analyzed at extremely low levels, parts per trillion for aqueous. 01:14:17.000 --> 01:14:30.000 Parts per billion for soils. And third, PFAS can exist at a sampling location at concentrations that are orders of magnitude higher than our screening or regulatory threshold. 01:14:30.000 --> 01:14:40.000 So there's an increased risk that routine decontamination processes are inadequate and result in cross-contaminated samples. 01:14:40.000 --> 01:14:49.000 So many of the state-specific guidance documents detail specific materials and products to avoid during sampling. 01:14:49.000 --> 01:14:55.000 But the bottom line is be meticulous with sampling equipment, materials, and procedures. 01:14:55.000 --> 01:15:09.000 And collect plenty of equipment rinse aid and related QC samples to avoid false positive results. 01:15:09.000 --> 01:15:18.000 So when sampling for PFAS in surface water, another important consideration is foam containing PFAS. 01:15:18.000 --> 01:15:27.000 This foam results from dissolve phase PFAS compounds in surface waters that have been agitated by wind or wave action. 01:15:27.000 --> 01:15:40.000 And aggregate it into a mass above the surface water. Note that this is not the same as a triple F phones used for firefighting that can be transported to surface waters. 01:15:40.000 --> 01:15:50.000 So concentrations of PFAS containing foam can be tens of thousands of times higher than concentrations in water below it. 01:15:50.000 --> 01:16:00.000 Fact, one Minnesota study found that PFAS and foam enriched by over 100,000 times that of surface water in which it floated. 01:16:00.000 --> 01:16:06.000 And was enriched in long-chain PFAS over that of short chains. 01:16:06.000 --> 01:16:12.000 Foam on surface water can be found in all seasons and can be white, brown, or orange. 01:16:12.000 --> 01:16:22.000 Note that these foams are Easily windblown, resulting in further PFAS transport across the water body. 01:16:22.000 --> 01:16:27.000 Shown here are the four layers to consider when collecting samples. 01:16:27.000 --> 01:16:35.000 On top is the highly enriched phone. Below this is the surface micro layer, which includes the air water interface. 01:16:35.000 --> 01:16:41.000 And likely contains the highest PFAS concentrations in the water column itself. 01:16:41.000 --> 01:16:47.000 Below this is the Neustan layer, which contains minute aquatic organisms. 01:16:47.000 --> 01:16:58.000 And this layer is important because the aquatic community is potentially in contact with the highest concentration of PFAS that occurs in that surface micro layer. 01:16:58.000 --> 01:17:03.000 And then underlying this is the water column with sediment on the bottom. 01:17:03.000 --> 01:17:08.000 So remember that even if PFAS is not detected in surface water. 01:17:08.000 --> 01:17:19.000 It may be detected in the foam located immediately above it. 01:17:19.000 --> 01:17:28.000 So we also wanted to touch briefly on the importance in understanding the variety of analytical methods available for PFAS detection. 01:17:28.000 --> 01:17:37.000 These methods can differ in how the samples are prepared, overall quantitation schemes using internal or external standards. 01:17:37.000 --> 01:17:53.000 And even some incorporating isotope dilution. As we see in the table here, the number of PFAS analyzed can vary in these methods from two analytes up to 40 analytes using, for example, EPA method 1633. 01:17:53.000 --> 01:18:00.000 In all cases, the quantitation limits will be dependent upon the specific PFAS evaluated. 01:18:00.000 --> 01:18:08.000 The matrix that's analyzed and the methods utilized. For our purposes here in site characterization. 01:18:08.000 --> 01:18:15.000 We must consider all these analytical differences when choosing the most appropriate method for a given site. 01:18:15.000 --> 01:18:31.000 To ensure that project objectives are met. I mentioned earlier that ITRC has a training on sampling analysis that looks at this topic in detail and see the chat box for the link to that. 01:18:31.000 --> 01:18:46.000 And my last slide, before I turn it back over to Robert, describes a useful and somewhat STEMI qualitative analysis called the TOP assay. 01:18:46.000 --> 01:18:53.000 Shown here. Which stands for total oxidizable precursors. 01:18:53.000 --> 01:19:07.000 This is an analysis that helps us evaluate the amount of precursors that may be contributing or contribute in the future to perfluoralkyl acid concentrations we are measuring at a site. 01:19:07.000 --> 01:19:18.000 This lab method converts precursor compounds to terminal Perfluoroalkyl acids through oxidative digestion. 01:19:18.000 --> 01:19:31.000 The increase in perfluoroalkyl acids measured after the TOP assay relative to before is an estimate of the total oxidizable precursors present in a sample. 01:19:31.000 --> 01:19:42.000 Note that TOP digestion is relatively harsh and thus may overestimate the amount of precursor oxidation that would be expected under field conditions. 01:19:42.000 --> 01:19:51.000 Still, it can be a useful tool in assessing potential transformations that are or may occur at a site. 01:19:51.000 --> 01:20:02.000 Note also that the perfluoroalkyl acids generated have perfluoroalkyl chain links equal to or shorter than those present in the precursors. 01:20:02.000 --> 01:20:12.000 So again, from a site characterization perspective. Top assay can provide an additional line of evidence for source differentiation. 01:20:12.000 --> 01:20:23.000 Now, I'll say caution should be used in its application because of the likelihood of cationic and zwitterionic precursors to sorba soils. 01:20:23.000 --> 01:20:33.000 In other words, comparison of source samples to down gradient samples may be complicated by environmental fractionation, which we heard about earlier. 01:20:33.000 --> 01:20:38.000 The ITRC sampling and analysis training goes into this topic as well. 01:20:38.000 --> 01:20:53.000 So with that, I will turn it back over to Robert. Thank you. 01:20:53.000 --> 01:21:12.000 Okay. So we've talked about fate and transport. We've talked about how that informs how you're going to take your approach to site characterization talked about what data we want to gather and how to gather it so Now let's talk about what we do with that data. 01:21:12.000 --> 01:21:25.000 With those data. These PFAS PFAS… data sets can be large and complex. 01:21:25.000 --> 01:21:31.000 So we need to kind of think of ways to help us interpret and visualize the data. 01:21:31.000 --> 01:21:40.000 And so that's what we're going to cover here is ways that data visualization can help us to understand what's going on at our site a little bit better. 01:21:40.000 --> 01:21:48.000 Or just help us translate that those large data sets into something that we can really sink our teeth into. 01:21:48.000 --> 01:21:55.000 So, for example, we have on this slide Bar charts. 01:21:55.000 --> 01:22:04.000 Which can be a great way to show actual concentrations while also comparing the presence of PFAS species within samples. 01:22:04.000 --> 01:22:23.000 So you can see these stacked bar charts from different samples showing the different PFAS concentrations and different composition of PFAS with some some species showing up in some samples and not others. 01:22:23.000 --> 01:22:35.000 Pie charts and these radar plots or spider charts or radial diagrams they're sometimes called. These are great at showing proportions. 01:22:35.000 --> 01:22:50.000 Of a total, right? In a pie chart you know of course we have wedges of different colors representing different PFAS species. 01:22:50.000 --> 01:23:02.000 Categories of PFAS. So, you know, you can see that In red, for example, on that first one we have the carboxylates, PFCAs. 01:23:02.000 --> 01:23:11.000 With essays in yellow and so forth. 01:23:11.000 --> 01:23:18.000 So again, these are for showing proportions of the total with wedges of varying sizes. 01:23:18.000 --> 01:23:36.000 Radar plots um can either show actual values, actual concentrations plotted on multiple axes or possibly more commonly, they can show proportions or ratios of specific compounds or groups relative to the total measured. 01:23:36.000 --> 01:23:49.000 The latter is accomplished by dividing the concentration of an individual compound by the total concentration of all measured PFAS. It's a method of normalizing data. 01:23:49.000 --> 01:24:07.000 So it's important to remember that analytical precision gets less certain with very low levels of PFAS or really any contaminants so When those levels are very low, it could affect those ratios and give you less certainty that that's a true ratio. 01:24:07.000 --> 01:24:18.000 Because of analytical imprecision. So just important to remember that, you know, when we're using data with low levels. 01:24:18.000 --> 01:24:39.000 Viewing data in these ways can have many purposes from the mundane to the cutting edge. And the next few slides will include some case studies that used some of these tools to evaluate environmental fractionation and the effects of in situ petroleum remediation on PFAS composition in PFAS plumes. 01:24:39.000 --> 01:24:47.000 There is a little overlap with these data visualization techniques and some forensics techniques that we can apply. 01:24:47.000 --> 01:24:54.000 Skylar will talk more about that. In the upcoming forensics section. 01:24:54.000 --> 01:24:59.000 So Sticking with radial diagrams for a moment. 01:24:59.000 --> 01:25:19.000 Or those spider plots. Return to this generalized hypothetical scenario for an A triple F site that we brought up earlier in the fate and transport section As mentioned then, we can expect our PFAS composition or proportions of various PFAS species to change over the length of a plume. 01:25:19.000 --> 01:25:25.000 And these radial diagrams are one way to visualize the compositional changes. 01:25:25.000 --> 01:25:35.000 Now, in this example, we've selected a subset of PFAS, just four in this radial diagram, and that's primarily for simplicity. 01:25:35.000 --> 01:25:48.000 And that's often how it's done. We might have a data set with dozens of PFAS in it depending on what analytical method we used and what we actually have at our site, of course. 01:25:48.000 --> 01:25:56.000 But we can simplify the data by just choosing the most common constituents. 01:25:56.000 --> 01:26:03.000 And that can really tell us a lot and make our data visualization just kind of easier to absorb. 01:26:03.000 --> 01:26:15.000 So, um. So again, four compounds being plotted on these radial diagrams. We have eight two fluorotelomer sulfonate. 01:26:15.000 --> 01:26:31.000 Efna on the top, PFOS on the right, and PFHXA. On the bottom. And so we can see if you look As we look along the axis of the plume. 01:26:31.000 --> 01:26:55.000 We can see that you can use this tool to visualize the changes in ratios of these various constituents so you can see in the beginning we have a lot of A2 fluorotelomer sulfonate For example, as we move down gradient, that precursor gets broken down and we see less of it relative to total PFAS. 01:26:55.000 --> 01:27:03.000 And instead, we see increasing proportions of those carboxylates and sulfonates. 01:27:03.000 --> 01:27:23.000 And of course, we also see more of an increase in the carboxylates because sulfonates tend to be retained more than carboxylates. If we plotted additional short chain we would see an increase in those two. Again, this is made up data, so don't don't 01:27:23.000 --> 01:27:33.000 Don't do too much with it. But this is just a hypothetical example of what we might see if we use these plots to visualize data. 01:27:33.000 --> 01:27:42.000 At a site like this. And in fact, we have real life examples of having used that data at sites like this. 01:27:42.000 --> 01:27:54.000 This first case study is from Sweden. Where data visualization using radial diagrams was useful to help them understand what was going on at their site. 01:27:54.000 --> 01:28:02.000 In this case, they are plotting concentrations of three groups of PFAS. 01:28:02.000 --> 01:28:16.000 And one specific compound of interest. It's plotted on four axes again, using a logarithmic scale So make sure you make note of that logarithmic scale, of course. 01:28:16.000 --> 01:28:28.000 The groups are the groups some of the measured PFCAs the sum of the measured pf essays And then other precursors. 01:28:28.000 --> 01:28:36.000 So precursors besides six to fluorotelomer sulfonate, which is their single compound of interest. 01:28:36.000 --> 01:28:44.000 And that was just differentiated in this case because of site-specific considerations that we don't need to get into. 01:28:44.000 --> 01:28:51.000 But the patterns in the shape that result from plotting these on this axis like this. 01:28:51.000 --> 01:29:13.000 Pertain to the precursor and PFAA relationships. And again, these are connected just because it's a visual tool there's no the lines drawn between the different axes are just to help us visualize. There's no other meaning there. 01:29:13.000 --> 01:29:24.000 So if you look at would actually happen at the site I know it might be a lot to look at on the screen, but there are some some patterns that are evident from the data. 01:29:24.000 --> 01:29:31.000 First, we can see that the presence of PFSAs, PFCAs. And fluorotelomer sulphonates. 01:29:31.000 --> 01:29:39.000 Indicates that the foams produced The foams used here were produced by telomerization. 01:29:39.000 --> 01:29:44.000 Right? Because we see those pleurotelomer sulfonates. 01:29:44.000 --> 01:30:06.000 And just the composition in general of those foams. So additionally, a decrease in concentration of PFAAs can be seen along the main flow channel around the source area is around G8 and G9. It's circled in black here. 01:30:06.000 --> 01:30:20.000 And in the inset that purple box we can see that precursor concentrations are higher at the fire training areas and decrease down gradient. 01:30:20.000 --> 01:30:31.000 And that might be a result of sorption And less transport of those precursors, or it might be a result of biotransformation or abiotic transformation. 01:30:31.000 --> 01:30:38.000 Or a combination of those things. 01:30:38.000 --> 01:30:44.000 And this next example This next case study. 01:30:44.000 --> 01:30:50.000 We have four specific. Pfas that are plotted, kind of like my hypothetical example. 01:30:50.000 --> 01:31:10.000 So the use of the dark blue outline represents the source area samples. So that's what we that dark blue outline is what we see in the source area. So they're actually preserving that On every location that they've plotted on this map. 01:31:10.000 --> 01:31:22.000 For an easy visual comparison. Whereas the shaded lighter blue color represents the concentrations at the actual sample. 01:31:22.000 --> 01:31:29.000 So one pattern that emerges from this data is that we see that the precursor compound FHXSA. 01:31:29.000 --> 01:31:41.000 Is reduced in concentrations near the dissolved oxygen infusion wells that were used here. So again. 01:31:41.000 --> 01:32:00.000 Just looking at the legend there, those are the little triangles around there so you can see how the the um that introduction of DO infusion wells reduce the precursor ratios and concentrations. 01:32:00.000 --> 01:32:12.000 Probably by increasing transformation rates. So that wraps up our section on site characterization and data visualization. 01:32:12.000 --> 01:32:17.000 And we're going to take a quick break before getting into forensics. 01:32:17.000 --> 01:32:25.000 Give us a chance to rest our brains before getting into the real dense stuff. 01:32:25.000 --> 01:32:31.000 All right. Thank you, Robert. So as Robert said, we've now reached our Q&A break. 01:32:31.000 --> 01:32:36.000 Our first Q&A break of the training. We will have one more at the end if we have time. 01:32:36.000 --> 01:32:41.000 So I do have a few questions for our trainers. We don't have too much time. 01:32:41.000 --> 01:32:53.000 First question, can you explain why PFSAs absorb more strongly than PFCAs to organic carbon. 01:32:53.000 --> 01:33:09.000 Sure, this is uh Sorry, this is Dina. I can take that question. So the sulfonate head group is more hydrophobic and so it has stronger electrostatic interactions with the organic carbon. 01:33:09.000 --> 01:33:16.000 Than the carboxylate group. So this generally will just give you a higher KD or KOC. 01:33:16.000 --> 01:33:27.000 And then, of course, like we discussed earlier, this effect would be amplified for those longer chain ones. 01:33:27.000 --> 01:33:39.000 Thank you, Dina. Any of the other trainers have any other comments for that question? If not, I can move on to another one. 01:33:39.000 --> 01:33:59.000 All right. For PFAS sources to air from stack emissions, what are the transport pathways that we should be looking at? 01:33:59.000 --> 01:34:04.000 Any of the trainers have any thoughts for that one? 01:34:04.000 --> 01:34:24.000 Sure. I could take a stab at it. When we're talking about stack emissions, we're looking at PFAS that have basically been forced into air even if they're not ordinarily particularly volatile. 01:34:24.000 --> 01:34:32.000 And so short range transport on particulates or, you know. 01:34:32.000 --> 01:34:40.000 Short-range transport in general and processes like wet and dry deposition. 01:34:40.000 --> 01:34:44.000 Are some of the processes that you'd want to look at. 01:34:44.000 --> 01:35:01.000 In terms of air transport, you're going to want to understand um the uh the wind patterns Of course, in the area If there's any shifts in that on a daily or seasonal basis. 01:35:01.000 --> 01:35:09.000 To make sure that you're targeting those those right the correct areas for your samples. 01:35:09.000 --> 01:35:25.000 All right. Thank you, Robert. And if none of the other trainers have any comments, I will go ahead and move forward with the training just so we can keep on track with the time. But we will hopefully have a little more time at the end to ask some more questions. 01:35:25.000 --> 01:35:30.000 So thank you all for continuing to interact with that Q&A box. 01:35:30.000 --> 01:35:44.000 All right, and I'm now going to turn this over to Skylar to introduce forensics for us. 01:35:44.000 --> 01:35:50.000 All right, so we've talked about characterization. We've talked about fate and transport. 01:35:50.000 --> 01:35:57.000 Now we'll talk about forensics, which is kind of coming at things from a little bit of a different, but very complementary perspective. 01:35:57.000 --> 01:36:12.000 So one of the goals for this part of the module is to kind of walk out of it understanding more about source ID, what goes into that or what could go into that And a few techniques for distinguishing sources from each other 01:36:12.000 --> 01:36:17.000 And also from other processes. 01:36:17.000 --> 01:36:27.000 So what is environmental forensics? As a field, it's a fairly mature approach to assessing environmental chemistry data. 01:36:27.000 --> 01:36:34.000 With regards to PFAS, it's still relatively in its nascent stages. It's still, you know. 01:36:34.000 --> 01:36:58.000 We're still bringing the practices over from other constituents other areas of the field and applying them to PFAS. So this discussion is going to draw upon some fairly established methods and show how they relate to PFAS. But definitely consider these methods carefully and carefully 01:36:58.000 --> 01:37:09.000 Use practitioners who know how to apply these methods and definitely keep the nuances that we've talked about regarding fate and transport in mind. 01:37:09.000 --> 01:37:28.000 So kind of the goals of forensics is to take your data and maybe collect ancillary data sets You want to be able to identify what the sources are on your site, if there are any. You want to be able to distinguish between multiple sources if there are any. 01:37:28.000 --> 01:37:39.000 On-site versus offsite sources. And often this kind of culminates in in unraveling commingled plumes. If you're looking in groundwater and you've got multiple sources. 01:37:39.000 --> 01:37:52.000 Now, in order to be able to comment on sources and identify and isolate what they are and where they are, you also have to address natural attenuation processes, fate and transport processes, et cetera. 01:37:52.000 --> 01:37:59.000 And ultimately, this does play back into the csm where you have to have a good idea of what's going on at the site. 01:37:59.000 --> 01:38:14.000 So we'll get into how this applies to PFAS next. As I said, you know, we're still bringing these kind of time-honored methods over into PFAS. So you'll look in the literature and find a lot of recent literature. 01:38:14.000 --> 01:38:21.000 We're going to talk about a few kind of the key methods that have definitely distinguished themselves. 01:38:21.000 --> 01:38:36.000 So specifically to PFAS, we want to know where the sources are. This could relate to firefighting training areas. This could relate to multiple different formulations of a triple F. 01:38:36.000 --> 01:38:48.000 Used in firefighting. This could relate to leachate at landfills We want to know what the sources are. We want to identify them and identify where they are. 01:38:48.000 --> 01:38:54.000 We want to be able to unravel if there are overlapping PFAS sources. 01:38:54.000 --> 01:39:03.000 And we also want to understand a little bit about environmental fractionation. Is this signature I'm seeing in this particular sample due to a source. 01:39:03.000 --> 01:39:07.000 Is it due to an environmental process that's just modifying the flavor of my source? 01:39:07.000 --> 01:39:20.000 That's definitely very important to have an understanding of. And then again, also kind of tracking that evolution of these signatures, of these fingerprints, so to speak. 01:39:20.000 --> 01:39:28.000 A long flow path to your site to be able to tell if it's a source or if it's a transport mechanism. So understanding the history of your site is really important. 01:39:28.000 --> 01:39:38.000 And being able to bring the results of the analyses where we've already talked about and the ones that we're yet to talk about back to the historical site records. 01:39:38.000 --> 01:39:44.000 And the CSM and reconcile them all. 01:39:44.000 --> 01:39:55.000 So idea between identifying the source and differentiating that source from other things Identifying the source is kind of a main objective here. 01:39:55.000 --> 01:40:07.000 We want to know where it is. Um and to get to that point, we may have to untangle the data from multiple sources again. 01:40:07.000 --> 01:40:19.000 Talking about commingling or overlapping sources Let's say you're sampling along a plume and you are starting to get some results that just look different. 01:40:19.000 --> 01:40:35.000 You want to be able to say, is this a source? Is this a transport pattern? Or even still, is this potentially some sort of non-point contribution that's getting into the site. 01:40:35.000 --> 01:40:49.000 So starting to get into some of these methods, there are kind of two limbs to a forensic analysis and one is transforming the data, how you transform the data definitely affects the results you'll get from these methods. 01:40:49.000 --> 01:40:54.000 And then the other is kind of taking all of the pieces as a whole. 01:40:54.000 --> 01:41:01.000 So looking at you know traditionally in the environment industry, we'll look at one thing at a time. We'll make a map of PFOS. 01:41:01.000 --> 01:41:14.000 We'll make a trend graph, a PFOS. And then we'll rinse and repeat for many other things. So we could do this 40 times, 80 times. And we usually take things one at a time. And what these methods are going to do is kind of successively 01:41:14.000 --> 01:41:22.000 Group the full data set together. So these are kind of whole data set methods as we get into them. 01:41:22.000 --> 01:41:40.000 So we'll start kind of more simply, you know, looking at the relative the relative amounts of things with ratios getting farther into data visualization, which can handle a little bit more complexity and a few more samples and then getting into a few specialized analytical techniques 01:41:40.000 --> 01:41:49.000 And then kind of finally landing at these kind of whole data set techniques like PCA or cluster analysis. 01:41:49.000 --> 01:41:56.000 So kind of off the bat What is a diagnostic ratio? 01:41:56.000 --> 01:42:02.000 You can make ratios out of a lot of different things in your data set here. 01:42:02.000 --> 01:42:20.000 Really what it is, is kind of two main ways of going about it. You either compare the relative composition of your samples. So here, if you see sample A and sample B, These are two hypothetical samples that we're trying to distinguish and say, are they the same or are they different? 01:42:20.000 --> 01:42:25.000 You can see we have the carboxylates here. We have the sulfonates here. 01:42:25.000 --> 01:42:30.000 And floratelomer sulfonates. So we're looking at relative abundance for the first three rows. 01:42:30.000 --> 01:42:36.000 And the idea being you add You group those constituents together. 01:42:36.000 --> 01:42:46.000 And you take the sum totals and divide by the total of the concentration of the entire sample. And what you get is something between zero and one. 01:42:46.000 --> 01:42:57.000 So focusing on the first three rows here, sample a is enriched in PFSAs and sample B is enriched with PFCAs. 01:42:57.000 --> 01:43:05.000 Here that's very clearly different. Those are two very clearly different samples because those compositional ratios are so different. 01:43:05.000 --> 01:43:22.000 Now, a different way of going about it is to compare to constituents or groups of constituents together. So instead of a total percentage You could look at, let's say, divide the PFCAs by the PFSAs. 01:43:22.000 --> 01:43:30.000 And instead, you get similar information out of it, you know, looking at sample A, it's a very small number. 01:43:30.000 --> 01:43:35.000 But instead, this type of approach isn't limited to a max value of one. 01:43:35.000 --> 01:43:38.000 So you get a much larger number, 22 for sample B. 01:43:38.000 --> 01:43:59.000 And then if you wanted to kind of get even more granular and step down to two compounds, specifically looking at PFOA and PFOS Again, very small number versus very large number. Now, these are all ratios. These are all ways of going about it. They take 01:43:59.000 --> 01:44:06.000 Two things and compare them. So that you can then compare that number, that ratio between other samples. 01:44:06.000 --> 01:44:10.000 And you can, you know, you can bake your whole data set into it, you know. 01:44:10.000 --> 01:44:15.000 Tallying the different types of compounds together before you do the ratio. 01:44:15.000 --> 01:44:22.000 But ultimately, you're comparing two things together. And you're comparing two samples together. 01:44:22.000 --> 01:44:42.000 So if you want to scale beyond ratios, you start getting into data visualization and data visualization These are a few we've already talked about, but these are also very common environmental forensics. You're looking for your kind of fingerprinting file and here this is a bar chart, an unstacked bar chart 01:44:42.000 --> 01:44:55.000 And it's normalized so that it's normalized The… y-axis is a percentage rather than a total concentration. And that lets us kind of stack them together. 01:44:55.000 --> 01:45:01.000 So you can see in sample one, we're dominated by PFOA. 01:45:01.000 --> 01:45:08.000 You have HPA, whereas in Sample two, we're dominated by PFPEA. 01:45:08.000 --> 01:45:18.000 Pfhxa. And not just those, but you can see the overall mood is different between the two samples, just the profile. If you're looking at the full profile, they're a little bit different. 01:45:18.000 --> 01:45:28.000 So this is, again, normalized to 100%. You know, oftentimes you'll probably see something similar to this in the form of a pie chart. 01:45:28.000 --> 01:45:31.000 And a pie chart that you can show on a map. 01:45:31.000 --> 01:45:47.000 So the idea being you can quickly compare not just two things, but here you're comparing a number of different constituents together But you're still kind of only showing two samples easily, at least on the same page. 01:45:47.000 --> 01:45:52.000 So if you want to scale beyond that a little bit more, this is a stacked bar chart. 01:45:52.000 --> 01:46:02.000 Now the right panel here, the right-hand panel is a little different. There's a couple of nuances to it. The upper one. 01:46:02.000 --> 01:46:14.000 Yeah, so these would appear to be stacked by concentration. You can see they have varying heights to them. So that's something you can do. We've got the numbers on the top, in fact. 01:46:14.000 --> 01:46:24.000 Indicating the raw concentration. So that's a way that you can tell different samples apart. You've got these, so in the stack chart specifically. 01:46:24.000 --> 01:46:28.000 The different colors now are what the individual bars were on the previous slide. 01:46:28.000 --> 01:46:48.000 So a blue bar is different from a blue light blue red bar. And again, those colors relate two different constituents. So again, compositionally, you're looking across and you're looking for basically different patterns in color And here, instead of just two samples, you've got six different samples. 01:46:48.000 --> 01:47:05.000 And you can say, well, these ratios look similar but one sample is much higher in concentration than another. And that's where you start to get into what's source like versus not source like So here you've scaled, you've kept the number of constituents 01:47:05.000 --> 01:47:17.000 And you're now increasing the number of samples. But even then you start to hit a bit of a threshold of how much can actually be practical to look at. 01:47:17.000 --> 01:47:25.000 Scaling up you know this is Honestly, about the same level of scale. Looking at a radial diagram, we've seen these before. 01:47:25.000 --> 01:47:31.000 This is where I think it's important to note that your data transformations do have an effect. 01:47:31.000 --> 01:47:47.000 On what the visual looks like. So radar chart, you read instead of up from the bottom like you would on a bar chart, you're reading out from the center. So you can see if you trace out from the center for these different constituents, sample A 01:47:47.000 --> 01:47:52.000 Is dominated by PFOS and a little bit less by PFHXS. 01:47:52.000 --> 01:47:57.000 Sample B is dominated by PFNA. Primarily. 01:47:57.000 --> 01:48:03.000 So in this case, the data are normalized to tally to a total, you know, to a percentage again. 01:48:03.000 --> 01:48:08.000 In this case, a fraction. So the highest things is like 60%. 01:48:08.000 --> 01:48:16.000 And that would be PFOS. Now, there's another common transformation. Folks are probably familiar with, which is a logarithmic transformation. 01:48:16.000 --> 01:48:23.000 In that case, going out from the center isn't a percentage. It's a concentration. 01:48:23.000 --> 01:48:33.000 So patterns farther from the center would be higher concentration samples. Patterns closer to the center would be lower concentration samples. 01:48:33.000 --> 01:48:51.000 So you really do have to track your data transformations. And remember the effect that they could have on these graphs. But even so, you've got some patterns. This is a fairly compelling a difference. But let's say you start getting more and more samples, it does start to become cluttered. 01:48:51.000 --> 01:48:59.000 And if you have more constituents like let's say a larger analytical suite, it just becomes kind of incomprehensible at some point. 01:48:59.000 --> 01:49:05.000 So what do you do when you need to scale up from this? 01:49:05.000 --> 01:49:14.000 This is where we would start talking about multivariate approaches. Multivariate's a large word. It can be intimidating. 01:49:14.000 --> 01:49:21.000 But really, at the end of the day. The kind of key players in these analyses are your data transformations. 01:49:21.000 --> 01:49:40.000 And understanding of what these methods do. So again. Definitely keeping in mind that when you do PCA or factor analysis or cluster analysis, have an analyst to do it who's Who's familiar with the analysis or be prepared to really do some homework to make sure you're understanding what you're doing. 01:49:40.000 --> 01:49:53.000 So for PCA and factor analysis these are Also called dimensionality reduction method. So imagine, you know, you make 100 graphs For a hundred different constituents. 01:49:53.000 --> 01:49:58.000 And instead, you could make one graph that shows them all. That's kind of the goal. 01:49:58.000 --> 01:50:01.000 Or you could have a couple of graphs that show them all. 01:50:01.000 --> 01:50:14.000 That's the goal of a dimensionality reduction method is it collapses the information into a much smaller data set or a much smaller number of things to consider. It extracts kind of a latent trend from the data. 01:50:14.000 --> 01:50:18.000 Now, cluster analysis extracts latent groups from the data. 01:50:18.000 --> 01:50:35.000 It focuses on similarity and dissimilarity so for the the following examples we'll be relying on this table at the right. So imagine in your table, every row is a different sample and every column is a different analyte. I think for each example we have. 01:50:35.000 --> 01:50:50.000 We have about eight data points. So again, principal components, you can think about as if you had your multivariate data cloud let's say There are how many constituents? One, two, three, four, five. 01:50:50.000 --> 01:50:55.000 Let's, you know, we can't see in 5d These methods can. 01:50:55.000 --> 01:51:06.000 So if you have… a data cloud in 5D. What Principal Components is trying to do is it's going to spear the data along the direction of the greatest spread. 01:51:06.000 --> 01:51:22.000 This may or may not align with a particular source. It's just the greatest spread in the data cloud. And then what it does successively is it spears the data through the next direction of greatest spread, but it locks it in at basically right angles at orthogonality. 01:51:22.000 --> 01:51:31.000 So what you get essentially is this thing on the right This is a representation of the data using the outputs of PCA. 01:51:31.000 --> 01:51:46.000 You can see our data cloud is eight data points. The black axes, PC1 and 2, are basically those spears that I mentioned that went through the data set in the directions of greatest spread, or that's what the analysis should do. 01:51:46.000 --> 01:51:50.000 And then… it represents it in 2D. 01:51:50.000 --> 01:51:55.000 Or if you want to in 3D, depending how many PCs you keep. 01:51:55.000 --> 01:52:05.000 Now, what this does is it collapses the samples. So every row instead of having five things to go with it. There's just one data point. 01:52:05.000 --> 01:52:09.000 Per sample now. So those are the eight data points shown on the graph. 01:52:09.000 --> 01:52:27.000 And the columns are shown as arrows. That's the five red arrows here. So you can quickly see the trends in your data you can see which samples are aligning with which constituents. You've got samples one, two, and three going with PFOA and PFBA. 01:52:27.000 --> 01:52:41.000 Pfos is loading for samples four and six. And then PFBS and PFNA are loading for samples five, seven, and eight. And that can quickly tell you some fairly salient patterns in your data. 01:52:41.000 --> 01:52:59.000 Keep in mind too, along these axes, we have a percent of variance explained pca inherently does not capture 100% of the information into two degrees here, unless you're only looking at two constituents, which really at that point, just make a scatter plot. 01:52:59.000 --> 01:53:17.000 So just definitely, you have to take the plot that you get with a grain of salt, but definitely if you do pca make a buy plot. That's like the standard output and it can help you unpack some of these major trends in your data. 01:53:17.000 --> 01:53:24.000 Clustering is a bit of a different animal. Instead of maximizing the variance, looking for the directions of biggest spread. 01:53:24.000 --> 01:53:38.000 It looks at sample similarity. So it looks for things that are close together in 5D or let's say if you have a large assay, maybe 30D, 40D, 60D, all these dimensions, again. 01:53:38.000 --> 01:53:52.000 It does something that we can't intuitively do. But the machine can. It'll compare the concentrations of each sample and just check which ones are close and which ones are far away. That's kind of the nuts and bolts of it. 01:53:52.000 --> 01:53:57.000 Now here, this thing is called a dendrogram. This is a specific form of clustering. 01:53:57.000 --> 01:54:03.000 Called hierarchical clustering. Where at the base of this tree, so to speak. 01:54:03.000 --> 01:54:13.000 These are the nodes and those represent individual samples. And as you kind of look at that level of granularity at the bottom, everything's different. 01:54:13.000 --> 01:54:22.000 But as you kind of zoom out, certain things are more similar to each other than others. And that's what forms the branches that form cluster three, two, and one. 01:54:22.000 --> 01:54:27.000 At some point, everything is similar. If you look at it with a broad enough brush. 01:54:27.000 --> 01:54:51.000 But it's up to the analyst to determine at what level is it relevant, at what level is the cluster structure relevant so here three are fairly distinct. Now, it's important to keep in mind too that just because a clustering output gives you a number. It doesn't necessarily mean that those are really well-defined clusters. It just means that those samples are more similar to each other than other things. 01:54:51.000 --> 01:54:56.000 So it's important to again Check your work, do your homework. 01:54:56.000 --> 01:55:01.000 Make sure you have a good diagnostics in place. 01:55:01.000 --> 01:55:06.000 To kind of vet your answers. And then when you tie that to the PFAS data to actual meaning. 01:55:06.000 --> 01:55:28.000 It's important to understand that these aren't necessarily different sources you could have a lower concentration group that's just like diluted samples you could have a non-point source. You could have um You could have transformations, et cetera. These aren' Each cluster isn't necessarily a source. 01:55:28.000 --> 01:55:33.000 So you have to then, again, interpret carefully. 01:55:33.000 --> 01:55:42.000 Factor analysis, also known as receptor modeling in the literature. Is similar to principal components analysis. 01:55:42.000 --> 01:55:50.000 But unlike principal components analysis it doesn't throw orthogonal spheres through your data to preserve that structure. 01:55:50.000 --> 01:55:57.000 Goes through the data at angles. This is more of an exemplar. 01:55:57.000 --> 01:56:03.000 Identification effort where you're finding the unique signatures in the data. 01:56:03.000 --> 01:56:08.000 So to do this, you have to estimate how many you think there are. 01:56:08.000 --> 01:56:16.000 And then what the method will do is it will extract the profiles of those end members, so to speak. 01:56:16.000 --> 01:56:20.000 And also the contributions of each end member to each sample. 01:56:20.000 --> 01:56:24.000 Again, you know, it's up to the analyst to determine how many. 01:56:24.000 --> 01:56:41.000 And the contributions to those samples aren't necessarily source contributions. It could be an ambient signature. It could be a fate and transport process or a precursor transformation. So again, interpreting carefully. 01:56:41.000 --> 01:56:49.000 So again, each of these methods are limited. That's not to say don't use them, but be careful when you use them and use them well. 01:56:49.000 --> 01:56:55.000 Factor analysis and clustering require the user to specify a guess of how many. 01:56:55.000 --> 01:56:59.000 There's diagnostic ways to do this, but beyond the scope of this presentation. 01:56:59.000 --> 01:57:04.000 The analyst also has to iterate a few times and just check to make sure that's a good answer. 01:57:04.000 --> 01:57:11.000 And then at the end of the day, this is not a method. These are methods that will tell you the answer per se. They will give you a result. 01:57:11.000 --> 01:57:20.000 That you must interpret in light of the CSM. And the chemistry. 01:57:20.000 --> 01:57:38.000 So, you know, there's conditions that will affect these signatures, you know, again, so we're identifying signatures here and then the trick is true sources out of those signatures versus things that are not sources per se there could be diffuse or ambient signatures. 01:57:38.000 --> 01:57:46.000 In samples, it could be fractionation due to transport. You could have precursor transformations or other reactions going on. 01:57:46.000 --> 01:57:50.000 And you could be fingerprinting those things rather than a true source. 01:57:50.000 --> 01:57:56.000 And again, one of the baselines of forensics is use it as a line of evidence, but use other things to go with it. 01:57:56.000 --> 01:58:02.000 Other lab analyses branched or linear isomers. 01:58:02.000 --> 01:58:06.000 Again, putting it on a map, put it on a map and see where these things are. 01:58:06.000 --> 01:58:20.000 See how things change over time. And compared to your site history records those are very strong ways to leverage the CSM to kind of check your work and interpret the data. 01:58:20.000 --> 01:58:33.000 So looking at physical and chemical processes you know again a source or a sample in a source may reflect the same thing as a down gradient sample. It just may have fractionated along the way. 01:58:33.000 --> 01:58:44.000 There's things that you also measure typically like the ph the redox soil parameters. 01:58:44.000 --> 01:58:50.000 That you can use to help you understand not only CSM, but also your PFAS data. 01:58:50.000 --> 01:59:01.000 And then there's things about the compounds themselves, like the chain link, functional groups, the structure that can give you clues to help unpack these analyses as well. 01:59:01.000 --> 01:59:08.000 We've got a few literature references here for later review. If you want to see some of these methods used in action. 01:59:08.000 --> 01:59:16.000 And there's definitely a number of other references as well in the ITRC guidance document. 01:59:16.000 --> 01:59:25.000 So with that, I will pass it back over for review and wrap up. 01:59:25.000 --> 01:59:37.000 All right. Thank you, Skylar. Quite a lot of information there so So yeah, just to review, we covered in this module. 01:59:37.000 --> 01:59:43.000 Site characterization, patent transport and source identification and forensics. 01:59:43.000 --> 01:59:51.000 We hope that we've given you a better idea of fate and transport processes that are unique to PFAS. 01:59:51.000 --> 01:59:55.000 How their physical and chemical properties play a role in that. 01:59:55.000 --> 02:00:02.000 And how to use that information to help plan your site characterization effort. 02:00:02.000 --> 02:00:08.000 And really target the appropriate sample locations and types of samples as well. 02:00:08.000 --> 02:00:25.000 And then we've given you an overview of how to interpret some of those data for different different purposes, including potentially looking at forensics And distinguishing between different sources. 02:00:25.000 --> 02:00:41.000 So again. This is one resource of many that ITRC PFAS team has developed We build these trainings off of our online guidance document. 02:00:41.000 --> 02:00:55.000 Which you can find at the link on this slide. You can also find introductory training for folks who would like to go back and get some of the basics. There's an archived introductory training. We give that live occasionally as well. 02:00:55.000 --> 02:01:14.000 And other video resources as well. And all those other resources that are available. So if you need a little bit more to help your understanding after getting this training, check out all those resources they're uh We're really proud of them. 02:01:14.000 --> 02:01:21.000 And just as a reminder, this is one of five modules in this Beyond the Basics training. 02:01:21.000 --> 02:01:29.000 So feel free to jump in to the ITRC website. And, um. 02:01:29.000 --> 02:01:44.000 Find those other archive trainings. Also, if you'd like to give us any feedback, there's a link there to a feedback form and certificate for the training so please feel free to check that out. 02:01:44.000 --> 02:01:50.000 That I'll hand it back to Taylor. 02:01:50.000 --> 02:01:57.000 All right. Thank you, Robert. And thank you, Skylar. We've now reached the end of the training and the last Q&A portion. 02:01:57.000 --> 02:02:02.000 I do realize that we are a bit over time, so I'm just going to ask Skylar one question before I wrap it up today. 02:02:02.000 --> 02:02:09.000 For factor analysis, can you give an example of what the factors represent? 02:02:09.000 --> 02:02:29.000 Yeah, so factor analysis, kind of an exemplar analysis the most unique signatures in your data. And again, these could be anything from sources to non-point or ambient contributions or fate and transport. So I wouldn't be surprised If you have a source or two. 02:02:29.000 --> 02:02:39.000 That could be represented by factors, but then you've got to unpack too which of these profiles could be due to fate and transport? Where are they located? 02:02:39.000 --> 02:02:44.000 What's around them, what other context can you add? And then the ever-present question of like. 02:02:44.000 --> 02:02:52.000 What what is an on-point source and and where How do I tease that out of the results as well? But yeah, you could be all of those things. 02:02:52.000 --> 02:03:00.000 And especially within a single site, you could have multiple sources as well. 02:03:00.000 --> 02:03:07.000 All right. Thank you, Skylar. And thank you to all the trainers for being here today. And thank you all for attending the training. 02:03:07.000 --> 02:03:16.000 We would like to hear back from you, so please be sure to fill out the online feedback form that is on this last slide and posted in the chat. 02:03:16.000 --> 02:03:23.000 Filling out the feedback form and certifying that you participated will allow you to receive a certificate of completion by email. 02:03:23.000 --> 02:03:35.000 If you need further clarification on the answers or would like to ask more questions, feel free to email us at itrc at itrcweb.org and we will follow up with our trainers to get your questions answered. 02:03:35.000 --> 02:04:05.000 Thank you all.