REGENESIS Technical Perspectives – Part 1 – PlumeStop® Colloidal Activated Carbon and Modelling

Hello and welcome everyone. My name is Dane Menke. I am the Digital Marketing Manager here at Regenesis and Land Science. Before we get started, I have just a few administrative items to cover. Since we’re trying to keep this under an hour, today’s presentation will be conducted with the audience audio settings on mute. This will minimize unwanted background noise from the large number of participants joining us today. If the webinar or audio quality degrades, please try refreshing your browser. If that does not fix the issue, please disconnect and repeat the original login steps to rejoin the webcast.

If you have a question, we encourage you to ask it using the question feature located on the webinar panel. We’ll collect your questions and do our best to answer them at the end of the presentation. If we don’t address your question within the time permitting, we’ll make an effort to follow up with you after the webinar. We are recording this webinar and a link to the recording will be emailed to you once it is available. In order to continue to sponsor events that are of value and worthy of your time, We will be sending out a brief survey following the webinar to get your feedback.

Today’s presentation is the first in a limited educational webinar series titled Regenesis Technical Perspectives on the application of advanced modeling for in-situ groundwater treatment using colloidal activated carbon. This module in the series introduces the connection between plume stop and modeling and shows how designs can be explored and optimized. With that, I’d like to introduce our presenter for the series.

We are pleased to have with us today Dr. Jeremy Birnstingl, Vice President of Environmental Technology at Regenesis. Dr. Birnstingl serves as a Senior Regenesis Technical Resource on key remediation projects involving advanced institute technologies worldwide. He is the author of the PlumeForce software used for design and modeling of the Regenesis activated carbon-based technologies. Dr. Birnstingl received a Bachelor’s of Science in Environmental Biology from the University Essex and a PhD in Environmental Chemistry from the University of Lancaster. He is a fellow of the Royal Society of Chemistry in the United Kingdom and a chartered environmentalist. He has over 30 years experience in the commercial and academic environmental sectors, including 20 years with Regenesis. His creative and scientific insight have been recognized through three commercial patents.

All right, so that concludes our introduction and now I will hand things over to Dr. Jeremy Birnstingl to get us started.

Good, well, thank you very much, Dane, for that introduction. As Dane mentioned, this is a series of three webinars looking at the practical use of modeling and the design and management of in-situ groundwater treatment with colloidal activated carbon. This is module one, which will focus on the principles of Plume Stop colloidal activated carbon, how that segues with modeling and how that can be used in the design. Overall, this webinar series is about the use of modelling to expand our understanding of in situ activated carbon remediation projects.

Modelling can offer a window into the governing processes behind the limited snapshots available to us through the point groundwater samples with which we’d be familiar as environmental professionals. The information that we can get through this can then be leveraged to improve design, improve performance tracking and projection and to help support engineering management and project control. So within the series webinar one will introduce the connection between PlumeStop and modeling and shows how designs can be explored and optimized. Webinar two will show how modeling can provide a window into hidden processes and shows why the mechanisms behind what we see may not be as they first appear. And then webinar three, we’ll explore how modeling can be used as a support tool throughout a project, assisting in communication, data interpretation, and performance optimization.

So let’s move on with this webinar, which is webinar one. PlumeStop colloidal activated carbon and modeling. We’re going to look at the connection between PlumeStop and modeling and how designs can be explored and optimized. First some basics and I’ll keep these simple. What is PlumeStop? What is a model? And how do we use them together? Well, what is PlumeStop? PlumeStop liquid activated carbon comprises tiny particles of activated carbon one to two microns in diameter in dispersion polymers. It has a area of about a thousand square meters per gram, that’s a hundred acres per pound, and that’s just the dry weight of the carbon, not the weight of the material itself as it would be shipped. This has a very high sorptivity.

It is injectable at low pressure, it will flow like an ink and like an ink it will coat the flux channels that it passes through as a liquid and will attach to the soil surfaces. The coating that it leaves is small and it does not impede groundwater flow to any measurable extent. This then gives us a powerful tool for plume management, the clues in the name plume and stop, but beyond that there are a number of sophisticated applications for which the reagent can also be used. To give an idea of how it actually moves through the subsurface, this remains my favourite animation and what we can see is plume stop being added to the column on the left and padded activated carbon, similar quantity being added to the column on the right.

We’ve looked at the time lapse over about three minutes and we can see that the plume stop has just flooded down through the column, leaving a trail of black carbon on the soil behind it, whereas the activated carbon has plugged up at the top. So the activated carbon image on the right has kind of plugged and what we can see is a sieve with a mesh size greater than the carbon particle sizes, the carbon not going through it on the right. On the left we can see plume stop and the coating it’s left behind. This image that you can see is sand particles prior to application of plume stop. We look at the scale bar down here, we have a scale of about 50 microns, so these sand particles are perhaps 100 microns across.

After the plume stop has been applied, note the bars reduced to 20 microns. We can see the coating of activated carbon particles showing as this rough boulder field coating the top of the sand grains. That’s kind of the coating that we would leave and you can note that the macro pore spaces are hardly impacted by this. We’ve just got a coating on the surface and that’s why the impact to groundwater flow is negligible. How do we use Well, let’s look at a typical barrier configuration and I’m going to keep this very simple for first base. This is the first webinar after all.

We have an aquifer. Let’s contaminate it with some dirty water. We’re going to install a plume stop barrier. There it is. Same coloration as we just saw in the sand column. That will behave more or less like a carbon filter, a brita filter in the subsurface. so that when contaminated water passes through it, the water that comes out of the other side is clean. If we have biodegradable contaminants, then bioregeneration of adsorption sites, or indeed ferro regeneration of adsorption sites, if we use zero valent iron and solvent, will extend the barrier longevity. The carbon quantity that we need for this is tiny.

The groundwater flow is not appreciably affected. The capture efficiency is very high, which means the contaminant infection, the contaminant movement in groundwater, is slowed significantly. The contaminants are slowed, the groundwater isn’t. The groundwater may therefore flow through the barrier in days, but the contaminants may take years to pass through. And if they are amenable to degradation or zero-valent ion treatment, then applying reagents to support that into the barrier will extend its performance. And if the degradation rates are fast enough, the extension will be indefinite. What is a model?

Well, a model can be thought of as rather like a flight simulator. It’s a tool which will perform a complex set of interacting calculations for us so that we can see how the system behaves when different inputs are adjusted. For example, we can use it to explore the impact of different reagent doses, different reagent mixes, or different reagent placement arrangements. And the software can then generate quantitative visualizations of the predicted outcomes of any of the scenarios that we’ve chosen to explore. Overall, therefore, this helps us with the three principles of understanding better, designing better, and communicating better.

Plume stop and models, how do we use them together? Plume stop lends itself very well to modelling. In simple terms, plume stop slows or stops a plume, plume and stop. Slowing a plume is known as retardation in the modelling vernacular, and retardation is common to all fate and transport models that may be available for looking at the fate and transport of contaminants groundwater. Modelling can therefore provide insights into what plume stop will do. So let’s look at the core principles of how this works and how this connection between plume stop and modelling takes place.

The first thing to consider is that although plume stop is often considered a means of contaminant capture and is often described in those terms, this descriptor may be a good starting point, but more accurately what Plum Stop is doing is changing the equilibrium between phases. By the phases I mean the soil phase and the groundwater phase principally. So in the image that we can see here, this is a schematic looking at dissolved phase contaminants in groundwater and salt phase contaminants on the soil. And the image that we can see on the left has what we would call a low Kd. A Kd is the equilibrium of soil concentration divided by water concentration.

When the system is in balance and the rate of mass desorbing is the same as the mass sorbing or in equilibrium, then the ratio of the concentrations we see in soil and groundwater is the Kd. So we divide the soil concentration by the water concentration. So if this is our water concentration, high number, and this is our concentration, low number, low number divided by a big number gives us a very low Kd and vice versa. If we have a low dissolved concentration and a high sorb concentration, we’re going to have a high Kd. I’m going to come back to the phrase Kd again and again through this talk. Just recall that we’re talking about the balance of mass, balance of concentrations on the dissolved or the sorb phases. The equilibrium, the KD, is what will determine how much this plume is slowed.

In other words, it’s retardation. So, how does a barrier work? Plume stock barriers will slow the migration of a plume. Retardation is defined as the slowing of the plume relative to the flow of groundwater. And it occurs when some of the advancing contaminant attaches to the soil. So, example, if one part in two of the advancing plume attaches, then the plume velocity is reduced by half. Because retardation is caused by absorption of mobile dissolved contaminants onto the immobile soil particles, equilibration retardation is therefore directly linked to the Kd. And so the higher the Kd, the greater the retardation.

So in this image, we can see that most of the contaminant is in the dissolved phase and moves forward and as it moves forward only a small fraction attaches to the soil. The greater body moves forward with the groundwater and so the slowing by loss onto the sorb phase is quite minimal. On the contrary if we have a relatively small amount of mass in the groundwater compared to the amount that’s going to sorb to the soil then it’s going to take an awful lot of plume flooding in to actually advance forwards if most of that mass is just nose diving onto the soil itself as we go. The higher the Kd, the more attaches to the soil, the less moves forward and the greater the retardation or the slowing of the plume. The degree of retardation is described by the retardation factor.

Let’s take a moment to look at retardation factors and complete the foundation to what I want to talk about in the rest of this presentation. A retardation factor determines, as I mentioned, how fast contaminants move relative to the groundwater. So if the retardation factor is one, it’s going to move at the groundwater velocity. So if this is a travel distance per unit time, the groundwater might move this far per unit time. Something with a retardation factor of two is going to move at half the groundwater velocity. Something with a retardation factor of 10 is going to move at a tenth of the groundwater velocity. So when the groundwater has moved this far, depending on the retardation factor, the amount of movement is slowed significantly.

What are some typical values that we might see in our industry? Well, for VOCs, chlorinated ethanes in soil, then we might typically have retardation factors in the range of about one vinyl chloride say to three TCE say. That’s on natural soil though. If we were to put a small amount of plume stop in that soil 0.1 of a percent and we have let’s say 100 micrograms per litre of TCE coming in we’re not going to have a retardation factor of three anymore, we’ve changed the Kd, we are going to have a retardation factor more like 600. And if that was a Pifoa, for example, and let’s say we’ve got the same amount of plume stop, 0.1%, and let’s now say 100 nanograms per litre of Pifoa, which would be more typical for mid-range Pifoa plume, we might have a retardation factor of about 12 ,600. That means that the Pifoa is going to move at one 12 ,600th the distance that a groundwater plume will move in a given amount of time.

This is for illustration only, I’ve used single species for these calculations but you can get an idea of the scale of difference that we can make. Plume stop and modelling. Plume stop lends itself to modelling because retardation is a function of the solved dissolved equilibrium, the Kd. Kd is fundamental to all fate and transport models. So therefore, if we can adjust the Kd in any model, we can then explore what Plume Stop will do. Well, let’s move on to an example. And for this example, I’m going to use the Remclaw MD, Remclaw Matrix Diffusion model.

So REMCLAW stands for Remediation Evaluation Model for Chlorinated Solvents. The first version was produced by Ron Falter of Clemson University in South Carolina in about 2007 and it’s recently been extended and updated with between Ron GSI and Kira Lynch of the EPA to accommodate Matrix Diffusion, which is a helpful addition. I’ve chosen Remclaw because it’s relatively recently produced. It’s freely available. You can download it from the GSI’s website, www.gsinv.com. And in my opinion, it’s a nice piece of software. So fan mail to Sharla, Ron, and the developers for the production of this. Thank you.

In common with other off-the-shelf fate transport software. It is not, I repeat not, designed for modeling activated carbon applications but also in common with other faith and transport software it can be used or perhaps I should say abused for this purpose. The outputs are not perfect but nevertheless they still do offer powerful insights into what an activated carbon application, what a plume stop application will So let’s have a look at an example. So this is an example PFAS hack of Remclaw MD. And Remclaw is not set up for working with PFAS, but it can be hacked to do that. And the way I’ve done this is to set up the model for the subject aquifer as standard. You don’t need to be able to read the details in the box, but these were some of the setup parameters that I used.

But after this setup, I have then put PFAS parameters, sorption parameters, et cetera, into the slots that were really designed for the chlorinated solvents. I’ve therefore turned biodegradation off because PFAS don’t biodegrade. And then I’ve entered the Kd values manually. You can do this in this little box down here, R in remclor MD. The Kd values to be put in can be calculated from the contaminant isotherm for activated carbon from whatever species that we’re looking at and the isotherms you can generally find from the literature. What is an isotherm we may ask? An isotherm describes how the adsorb concentration of a contaminant changes relative to the dissolved concentration of that solute equilibrium. It’s a mathematical description of that process.

It can be plotted out and in simple terms the Kd is then the ratio of the axis values of that isotherm plot at the groundwater concentration being modeled. So, for example, if I was dealing with a concentration of whatever this point is for PFAS species number one, then I can read from the isotherm or calculate from the term, what the sorbed equivalent sorbed mass will be on the carbon at that point. And if I look at the ratio of those concentrations and do a bit of extra math, then I can work out what the actual Kd is really just correcting for the quantity of activated carbon in the system. So it’s a reasonably simple approach. This calculation has to be done manually. It’s not a part of Remclaw MD.

When I do it though, this is the type of thing I get. So, this is an output from Remclor MD of my PFAS species after about 10 years. The graph is saying PCE, TCE, CIS, but the parameters I’ve put in are for PFOS, PFOA, and PFBS for this exercise. There’s no activated carbon in there, there’s just the natural fraction of organic carbon, the FOC, at about 0.1% of the soil mass. And we can see the concentrations of PFAS over this hundred foot distance declining. They’re not declining because of degradation, they don’t degrade. This I believe is a phenomenon of matrix diffusion. Types of KDs and retardation factors that we’ve got are the values that we can see here.

If, however, and this is where the fun starts, I adjust the KDs to match those that we get with plume stop in the system, engineered KDs, and I’m using a fraction of activated carbon plume stop of about 0.2, a fifth of 1%. Then you can see that the KDs and retardation factors are very different. Typically, we’re gonna be increasing them by 100 to 1 ,000 times. That’s two to three orders of magnitude. And we can see in the image that the distance the plume has moved is very, very different. The plume is advanced, maybe 250 to 1800 feet under the natural setting, whereas with this plume stop addition, it’s advanced less than 20 feet, in fact, less than about five feet for Bifoa and PFOS. And this will tell us that maybe a 20-foot plume stop barrier at this application would contain these concentrations under that kind of seepage velocity, 220 feet per year, for about a 20-year period.

It’s a very simple exploration. It’ll give an idea of scale. We can play around with it too. If I double the dose, we can see now that the fraction of activated carbon is about 004 rather than 002. Then the retardation factors are going to increase further and the distance that the plume moves over the 10-year period is going to reduce. So at that higher dose, we could have a more compressed barrier. Perhaps we don’t have space for a 20-foot barrier. Or indeed, by the same token, if we were to double the dose, we could, for a 20-year period rather than a 10-year period, contain that plume stop. So we can play tunes with this.

Double the dose, double the time, or double the dose, half the amount of distance the barrier has to cover for a given time. I trust you get a general feel of the types of play around we can do even with some hacked software. Remclaw MD is just one example of what can be used. Similar hacks are possible with any model. The entry point might be different, but the principle is the same. The approach is well suited, I would say, to explorations with an existing model setup. If we perhaps have a model that’s already been set up for a site that we’re working on for some purpose or another, most of the setup work has been done and the additional input adjustments, be it MT3D or whatever we’re looking at, the additional input adjustments for plume stop are very small and yet the output changes that we might see can be very significant.

There are some fascinating explorations to be had there. There are limitations. Explorations using KD hacks to models are approximate only. The models were not designed for this, but the exercise is still informative. By all means, use it for aha moments to see how things behave, and indeed for semi-quantitative explorations such as the ones that we looked at just a moment ago. Do not use the approach, however, for formal designs or for placing bets. The limitations are significant, but not so much that it makes the exercise worthless. The principal limitation that we have is that most existing models do not have a consideration for competitive sorption. And this makes a big difference.

There are other limitations too, but competitive sorption is the most numerically significant. For this purpose, the PlumeForce modelling software was developed really to accommodate competitive sorption and to overcome some of the limitations of using off-the-shelf models for plume stop explorations and to get to the point where modelling could be used for formal design. What is the PlumeForce modelling software? Well, it’s proprietary modeling software developed by Regenesis and operated by Regenesis. I designed it and I started on it in perhaps 2016. That’s what I looked like when I started, so you can get an idea about what the process will do to an individual.

Plume Force is a multi-phase finite difference model for the modelers among us that will accommodate dynamic desorption, destruction and competitive interactions between target species themselves and between target species and non-target species. Practically, it will predict groundwater concentrations of the contaminants at any point in space and time in the modelled domain. And this valuable uses at different stages of a project. It can be used in the early stages for conceptual site model refinement. It can help expose gaps or find out what the drivers are to what we might be seeing in a given plume or a given system.

High significant is matrix diffusion, degradation, whatever else it may be. We can use it for design optimization. That’s one of its main uses to identify which reagents are best for a given setting, what type of dose we need and what might we expect from a given application scenario. And then we can use it as a performance yardstick to help with the tracking of a project once the PlumeStop application has been made and we’re looking at data coming in on month one, month two, year one, year two and so on. How is performance tracking against the expectations? Is there any divergence? What does that divergence look like?

We can use the modelling to help understand what that might be and how to get the project on track if there is divergence. That’ll be the third webinar. Development of PlumeForce was specifically to support PlumeStop projects. The first version, as I mentioned, was developed in 2016. That accounted for sorption and biodegradation only, but there has been six, seven years of ongoing refinement and functionality and scope since that time. The key driver for its first development was to accommodate competitive sorption. This is not a feature of off-the-shelf fate and transport models, but it is a critical feature for activated carbon modelling. So let’s take a look at what I mean by competitive sorption. This is pronounced on activated carbon, plume stop, but it is not pronounced on natural organic carbon, FOC.

And for this reason, it’s not in most fate and transport models, which are based around retardation on FOC, the fraction of organic carbon. And it is important because the sorption extent, the Kd of each species will change as the contaminant mix and the relative concentrations of contaminants in a mix change. These changes will occur through biological or abiotic isca transformations and or the chromatic separation of contaminants within a barrier owing to differential retardation. Well, what on earth do I mean by chromatic separation or chromatographic separation? Well, for any analytical chemist among us, it’ll almost be immediately apparent.

The principle is the same as we would see in chromatography in an analytical laboratory. The process is something like this. Let’s say we’ve got groundwater flowing in this direction. We have a continuous source of contaminants and we have a mix of species. Species A is high sorbing, has a high Kd. It therefore has a high retardation factor, as we’ve seen, and therefore it will have a low velocity v relative to the groundwater. Species B is medium sorbing, medium KD. Its retardation is therefore medium and its velocity will then be medium. It’s moving faster than species A and likewise species C. Much lower sorbing, lower retardation, higher velocity. And so what we can immediately see is that naturally the faster species are going to start to move ahead of the slower species.

They may all start out together, like marathon runners, the starting gun, but because they’re all going at different speeds or different velocities, some way down the marathon course, then some are going to be arriving long before the others are going to arrive. And that means that whereas they might be bunched up in the beginning, and there’s lots of sort of competition on the carbon within the barrier for them, as they progress through the barrier, that competition for the faster species reduces and reduces and therefore the Kd’s are going to change and therefore the velocities are going to change and therefore the modeling outcome is going to be inaccurate if we don’t take account of that. How significant is the change? Perhaps you’re asking.

Well, let’s take an example with some Ptas species and I’m going to arbitrarily choose three species. 6 ,2-FTS as my high-solving species, PFOA and PFBS as my species B and C, and what we can see here is the retardation factors that we might have for a plinth stop application of, let’s say, 0.2% of the soil mass, all species at 50 ,000 nanograms per litre, that’s quite hot. The retardation factors are going to be these that we see here. What’s most significant is that as species B and C move away from the competition, you can see that their retardation factors increase significantly as they start to free themselves of the competition of other species. And you can see the lighter species, for example, at each step, its retardation is increasing by about a third.

And so if we were to try and apply these concentrations that we see in a mix for the entire duration the treatment, we would massively underestimate the retardation of the lighter species and our modeling would be way off. That’s the type of thing that would happen if we were to just measure isotherms with the source mix and then try and apply those isotherms to what’s going to happen throughout the barrier. We’re going to massively underestimate what’s happening, especially for the faster moving species, which typically would be the critical ones for the breakthrough times the barrier. Now this is an example with just three species.
A site’s PFAS soup may contain thousands more than this, so the principle gets extended. The core plume force functionality at the present time therefore includes the complexities of multiple reagents that we can add, different spatial arrangements and timing of reagent additions, reagent consumption and longevity, but also the interplay of multiple competing species, whether they’re the initial mixtures, whether they’re parent-daughter cascades, or indeed the natural and hidden competitors, maybe not just the contaminant species that we’re looking at, but the other stuff that may be in the mix as well.

We can look at the interplay of multiple phases within the software as well. The mast that’s in the high K transport units, by K I mean high conductivity transport units, mass that’s in the lower K storage units, the silts and the clays rather than the sands or the gravels, mass that’s sorbed to the natural organic carbon, fraction of organic carbon, and mass that’s sorbed to the plume stop carbon. Let’s see how some of this works then in application. And for the rest of the slides, I’d like to show you over the following 10 minutes, we’re going to look at how applications can be explored at the design stage using PlumeForce.

The first example that I’m going to look at is simple retardation, and I’m going to use TCE as my sample species. I’m going to keep it to one species to begin with and simple, and I’m simply going to look at the dose between retardation between the Plumestop dose and the longevity that we’ve got here. There’ll be no degradation to begin with in this first example. So we’re looking at TCE in an abiotic setting, maybe it’s an aerobic aquifer and plume stop only. So here are my settings in the box. Importantly, groundwater is about 225 feet per year. There’s no degradation, it’s a bit of matrix diffusion and there’s no degradation. The ambient biodegradation is turned off.

We’ve got TCE coming in at about a milligram per litre and it’s flatlining over the distance that we’re looking at because there’s no interruption of the flow. We’ve got a constant source and therefore the concentration is constant. On the vertical axis we have concentration and on the horizontal axis we have distance. So this is like a longitudinal section along the centerline of a plume sources somewhere to the left, and the plume is flowing off to the right through 10, 20, 30, 40 feet, and so on. Let’s add a plume stop barrier, and we’re going to make it an arbitrary 20-foot thickness. So in the model setup, it’s from about the 10-foot point to about the 30-foot point, 20 feet overall in the direction of flow.

The fraction of colloidal activated carbon, the plume stop application to the soil is, in this case, a tiny amount, about a hundredth of a percent of the plume stop, 0003 as a CAC. I deliberately made this small for the purpose of this exercise. That would be a very low dose for a real-world application. What we’ll immediately see, this is time zero, what we’ll immediately see after applying it is that the concentration in the application zone drops pretty much to non-detect even a small amount of carbon like this will be enough to solve the TCE probably down to detection limits or below so we’ve immediately reduced the TCE in the application zone to non-detect.

Down gradient though there is initially no change to the concentration because it takes time for the clean water to flow through the treatment zone and to flow down gradient and the runtime is zero. So let’s start to increase the runtime. A quarter of a year, three months, the clean waterfront has now started to move down gradient and so we can see there’s a bit of a clean space starting up here but the concentration decline is not instant because we’re going to be getting some matrix back diffusion and desorption from the soil organics holding it up. As time moves forward, half a year, we can see the clean water zone starts to increase, back diffusion starts to tail off a bit, the concentrations drop, but we’re starting to see a front of contaminated water beginning to make its way through the barrier. It’s not out yet.

We extend the time, 0.75 years, one year, all the time this front is moving forward as the water front is reducing. One and a quarter years, 18 months, 1.75 years and we’ve got clean water beyond the barrier but now the mass is broken through at this tiny dose. So we’ve got in this setting less than two years longevity at a tiny dose. That’s a small dose. What if we were to increase it? Well let’s try that. We can increase the longevity by increasing the plume stock dose. Let’s do that. Initial application rate. These are the types of KDs and retardation factors that we’ve got within the barrier. This is out before the barrier. This is inside the barrier. We’ve got a retardation factor that’s perhaps 13 times higher than background, but 13 times isn’t really enough to bias anything significant.

So let’s systematically increase the plume stop dose and see what happens to the longevity. We’ll increase it bit by bit. So up to 0.2, we’ve now got 3.1 years longevity. We can see the KD and the retailization factor increasing. 004, up a bit more, six years, 006, about nine years, 008, about 12 years to break through, 001, almost 15 years, 0012, 18 years, 14, 20 years. That’s enough for this design. So a fraction of colloidal activated carbon of 0014 gives us 20 years of sorptive longevity. The retardation factor is now 435, which means the plume is affecting at about 1 435th of the groundwater velocity. The plume has taken 20 years to advance 20 feet.

Without PlumeStop, it would have taken 40 days. It’s been slowed by a factor of something like 200. That was sorption only, but TCE can degrade. So what if we combine that sorption with biodegradation? Let’s take a look at some of the synergies of reagent applications. We’re going to look at retardation coupled with biodegradation. So for example, we might be using Plume Stop, that’s Plume Stop, with a compatible electron donor that we can co-apply with it, Aquafix, a long-lasting colloidal electron donor that’s recently been released by Regenesis to work simultaneously with Plume Stop. I’m going to take a look at biodegradation.

Well, this is the same starting point that we had before, so no degradation, no interruption, and I’m going to start this exploration by looking at biodegradation on its own. So here is my electron donor application zone in a theoretic sense in plume force. We’ll model a 20-foot application zone and we’ll allow in the software a 20-foot downgradient redox halo of donor influence, where the treatment zone set up kind of segues back to ambient background. And with rates taken directly from the literature, I’ve used Aronson and Howard from 1997, maybe slightly faster in the case of vinyl chloride and ethene. This is the type of degradation that we might see.

We can see the TCE flowing into the barrier. And when it hits the barrier, starts to degrade, cis-DCE and vinyl chloride start to appear as daughter products, the cis breaks down into vinyl chloride, the vinyl chloride breaks down into ethene and this process is not completed within the degradation zone and so we’re seeing these some concentrations of the daughter products start to come out to the other side. Cis-DCE, vinyl chloride were not present to begin with their daughter products. I’ve set ambient degradation to zero which is why these are flat lining outside. Under natural settings there might be some slow degradation but for illustration we’ll keep to the donor alone. So three-quarter reduction of TCE, okay, but now we’ve got a new 300 micrograms per litre of DCE and 75 micrograms per litre of vinyl chloride, not good enough.

Well maybe these rates were too slow, maybe we were a little bit cautious with these. What if the rates actually double? Well, let’s double them down to half of what they were. This is what we see. Better, but we’ve still got quite substantial amounts of daughter products coming out of the barrier. Well, what if we extend the barrier zone? Let’s double it from 20 to 40 feet. This is what we see. We’ve got now a 40-foot active biozone plus 20 feet of barrier. Getting better but still not there. Okay well maybe the rates are a bit faster or maybe that’s a bit ambitious. Could I tweak them all? Yes but I believe it I’m not sure. So let’s see what else we can do.

Let’s add some plume stop and this is what we get. Here’s some plume stop going in. It’s a 20 foot barrier in this case And what we can see is that the TCE and all the daughter products are entirely consumed within the first three feet of the barrier. And that’s happening because we’ve suddenly bought an awful lot more retention time in the barrier to the flow of those contaminants. Rather than flow through over a few days, the TCE would have taken years to go through. And that slowing has bought a lot more time for the degradation and so within the first three feet or so of the barrier, we’ve got full completion.

Well, that’s great, Jeremy, but you didn’t change the degradation rates. Are you sure that the same degradation rates would apply in a plume stock barrier? Well, let’s have a look. Let’s do a sensitivity test using the software of those degradation rates. What happens if we slow them by an order of magnitude let’s say. This by the way we would have data on from a variety of studies but I won’t go into that in detail in this particular setting. So let’s just do a sensitivity test and reduce the rates by 10. So we’ve got half lives of 300 days for TCE and DCE and 70 days or 35 days a month or two months for vinyl chloride and ethene. And still we can see that contaminants of concern are consumed within the first 10 feet of the barrier. So that’s pretty good. What about maybe not quite a slowed rate? Let’s try that five times.

We’re slowed down. These are perhaps more reasonable rates to put in. We can see that we’ve still consumed our contaminants of concern within the first five feet of the barrier. We could therefore safely reduce the barrier dimensions to, say 10 feet and still contain our plume. What that’s done now is compress the treatment zone by greater than 75 percent compared to bio on its own and there’s still space in the barrier. Maybe we could relax the plume stop dose. Well, let’s have a go at that. This is the dose we’ve got at the moment. If we relax it a little more, we’ll see that the wave that we’ve got within the barrier can start to reduce a little bit as the plume stop dose is relaxed. 0012, it’s extended a bit. 001, it’s extended a bit more. 008, ethene is almost coming out of the barrier. 006, whoops, ethene’s coming out.

Vinyl chloride is almost on the cusp of out. That’s probably a bit tight. So that’s slightly relaxed too far, but it’s given us an idea of the feel of things. We’ve got a better feeling of the dose-response relationship. Perhaps a fraction of colloidal activated carbon of 001 was about right. Let’s put that back in. So there’s our treatment zone at that dose, still compressed by greater than 75 percent compared bio on its own. The footprint is now a lot smaller. It will fit into the space perhaps between a building and the site boundary. And the contaminant destruction is greater, even with a generous safety margin on the degradation rates. So what we’ve seen here is that the plume stop dose can be altered to increase the barrier longevity or reduce the required barrier footprint.

We’ve seen a synergy between bioremediation. It could have equally been ferroremediation, ISCA, ZVI. Plume stop will shrink the treatment footprint for the reagent use, and the reagents will reduce the plume stop requirements for that footprint and extend the treatment longevity. And we’ve seen how modeling can assist in the design process to help optimize the reagent combinations and to assist with the sensitivity testing of input parameters to work out what would be an appropriate but still appropriately conservative design for a given site. So these were examples using just two processes, sorption and bio, and sorption and bio in combination.

As I mentioned, zero valent iron alone or in combination could equally have been explored. What we did I think enough for this talk. The permutations, as you might have fathomed, are endless, hence the benefit of modelling. Plume, slot, bio, zvi, any combination, different doses, dimensions, placement arrangements, separate, overlapping, partially overlapping, combined, put in at different times, all of it can be modelled. This enables what-if scenarios can be explored and optimal design scenarios can refined. But it’s still only a simulation, perhaps we’re thinking. How do we know it’s right? Well, that’s really dealt with through validation and calibration, which I’ll look at very briefly so that we can keep to time here.

This is done through comparing the predictions that we’ve made with the software to the observations that we get in the field post-application. Regenesis is very to see a lot of field performance which enables us to calibrate the modeling to field performance rather than just trying to do what we can with test tubes and columns etc. So what we do is we look at predictions and we compare them to what’s actually happened in real-world application. This provides validation and enables calibration of the model for that particular site and so once we’ve the calibration and we can see how numbers are actually tracking, then the estimated parameters in the model can be adjusted to match the actual field observations and the model predictions for that site can then be refined.

So give us a better idea of the longevity of the performance or whether any adjustments might be necessary. This can be done normally within the first few months by post application. The growing data set of observations that we get through doing this site after site and even on the same site over time but site after site gives us a data bank of information that we can use to help improve future design predictions and the assessment of the model predictions against field data also helps inform me what other refinements might be appropriate to improve the model and go for its continuous utility enhancement. So we’re on about 50 minutes. Let’s wrap up.

Plume stop impacts can be modelled. This can be powerfully informative. Any model can be used for this up to a point. Activated carbon modelling requires consideration of competitive sorption. Modelling can support the remediation design processes, the exploration, the optimisation and the communication of what we might see. Regenesis has a range of products that are compatible with Plume Stop that do not mess with desorption. We have S-micro-ZDI, an injectable colloidal sulfidated zero-valent ion. Details on www.regenesis.com. We have Aquifix, an injectable colloidal long-lasting solid electron donor. We have ORC Advanced, slow-release dissolved oxygen source. It may be that the target contaminants are biodegradable aerobically. Perhaps we’d use another product, Petrofix, for that. But sometimes the target contaminants might not be degradable, but competing species might be. We might be dealing with, let’s say, PFAS in a fire training area.

Well, Rick McGregor of Insitu Remediation Services used ORC Advanced with Plumestop on the first PFAS plume stop treatment to be formally written up in 2016. Do go and look at some of Rick McGregor’s articles on his field experience with plume stop and PFAS. SMCVI and Aquafix can be co-applied with plume stop. This reduces the field work. Each has a powerful reagent in its own right, but compatibility allows the combined strengths to be leveraged and provides powerful synergistic opportunities benefiting overall project efficiency and performance. And the ability to model this in Plume Force is essential to design and dose optimization. It also facilitates communication of the projected performance outcomes of the different design alternatives.

So the Plume Force software helps us understand better, design better, and communicate better. With that, I’ll close the first webinar, and the next webinar in this series, we’re going to go a little deeper. We’re going to go into the matrix and start to look at some emergent phenomena. There’ll be some spooky surprises, we’ll use modeling for X-ray vision, and we’ll look at some calibrated examples of the utility of this. More formally, we’re going to be looking at processes in isolation versus in combination, emergent phenomena, kinetic equilibria when things look like they’ve stalled but they certainly haven’t stalled.

We’re going to look at the impacts of hidden mass and how we manage invisible things and we’re going to look at windows into hidden dynamics and how these impact project design. So with that I’ll close. My contact details are here. Thank you for tuning into this webinar And at this point, I’d be happy to hand back to Dane to take a look at any of the questions that perhaps have come in through the course of this webinar.