Cleaning up the large number of groundwater contamination sites is a significant and complex environmental challenge. The environmental industry is continuously looking for remediation methods that are both effective and cost-efficient. Over the past 10 years there have been amazing, important developments in our understanding of key attenuation processes and technologies for evaluating natural attenuation processes, and a changing institutional perspective on when and where Monitored Natural Attenuation (MNA) may be applied. Despite these advances, restoring groundwater contaminated by anthropogenic sources to allow for unrestricted use continues to be a challenge. Because of a complex mix of physical, chemical, and biological constraints associated with active in-situ cleanup technologies, there has been a long standing focus on understanding natural processes that attenuate groundwater contaminant plumes. We will build upon basic environmental science and environmental engineering principles to discover how to best implement MNA as a viable treatment for groundwater contamination plumes. Additionally we will delve into the history behind and make predictions about the new directions for this technology. Any professional working in the environmental remediation industry will benefit from this in-depth study of MNA. We will use lectures, readings, and computational exercises to enhance our understanding and implementation of MNA. At the completion of this course, students will have updated understanding and practical tools that can be applied to all possible MNA sites.
Matrix Diffusion Sampling
Loading...
來自 Rice University 的課程
Natural Attenuation of Groundwater Contaminants: New Paradigms, Technologies, and Applications
96 個評分
從本節課中
Attenuation and Storage in Low K Zones
In these lectures, we will focus on the idea that storage of contaminants in geologic media as a potential natural attenuation mechanism. Under this scenario, certain contaminants diffuse in low permeability media such as silt, clay, and limestone and then eventually reenter the aquifer by a process called back diffusion. Like sorption, this process is not destructive, but attenuates a contaminant plume by sequestering some of the initial loading and then slowly releases it over long time scales. You will learn about diffusion vs. dispersion, see case studies in the lab and field, learn how to sample and model this process, and hear the latest research about degradation processes associated with matrix diffusion.
與講師見面
Pedro AlvarezGeorge R. Brown Professor of Civil and Environmental Engineering
Department of Civil and Environmental Engineering
Charles NewellVice President
GSI Environmental Inc.
David AdamsonSenior Environmental Engineer
GSI Environmental Inc.
Okay, well, we've talking about monitored natural attenuation and matrix diffusion
and how they're connected in terms of this storage and attenuation in these low prime
building zones, and today we're going to talk a little more about sampling.
But just to start out, one analogy that I've used today to describe this matrix
diffusion Is this idea that the physicists used that out there
when they look at the universe there's all these dark matter that's out there.
They know it's there but they don't know what it is or where it is.
And is some ways when we go to our ground water sites or
mediation sites we've been analyzing and sampling these sands.
These transmissive zones but really not looking into those clays and
those silts or those low k compartments.
And so that's our type of dark matter that's out here.
Have a big effect on how these sites, buthere's a question of some
as how can we measure the mass in this LocA compartments?
And how is this different than the fields typically done?
>> Yeah. >> So, what do we
typically done in the past?
>> Yeah, well I think this graphic lays it up pretty well.
And this is a graphic from the scene and science report for
this sort of project that we've talked about several times.
But we sort of call this first generation methods.
These are the conventional methods that a lot of people are comfortable
with going out to various sites.
They rely on a lot on groundwater samples.
So that's showing there on the left.
Basically installing these long screen wells within aqua
ferns mostly within the most transmissive points.
That's what you're basically targeting there and that averages then,
a concentration in there to get you this one number, right?
>> Yeah, and then maybe you're collecting some soil samples, but
probably not within the actual saturated zone.
And definitely probably not within those low K zones.
Most of that soil corn might be focused on the unsaturated zones.
So in this first generation sampling, no sampling at all in this low K zone.
>> Yeah, so we're still left with sort of that dark matter down there that were
not really even targeting with our investigations.
>> Okay, so that's first generation, what a second generation look like?
We're talking about a few different methods here within there.
But really, a lot of these things have this initial step, step one, where you're
doing real time profiling in order to identify the intervals of interest.
The depth intervals that you want to further look at with maybe soil sampling.
There's a lot of different ways to do this.
And there's tools, like the Hydraulic profiling tool, or the Waterloo Profiler.
And the Waterloo profiling gate is sort of shown in this graph, so that's something
that can get you, for an example, this index of hydraulic conductivity.
It's basically an estimate of how permeable those soils that you're
profiling down through.
It also actually will allow you to collect particles samples as you going
down through there so you're get contaminant concentration data.
And it is also giving you some physical chemical properties
of the ground water as well as geo-chemical conditions.
>> So what's going on the picture there?
>> Yeah, that picture basically showing then on that left panel is
going down through a different soil of types.
You're seeing a response I K so you might see as you go to a low K silt.
Rob dip in that Ik and so that's telling you that there's not a lot of water that
could actually move through that unit.
>> Low K unit, low probability.
>> Low probability.
In this case we're also collecting brown water samples as yore going through, or
maybe you're following that cup of salt samples but you're seeing that
the concentrations really spike when you got to that low K unit.
So this is the case where storage really is a big deal at this particular site.
>> So on the second generations sampling, I sort of think of this first step
is looking for this interphases really, right?
>> Exactly.
>> Transmissive zones and this low print buildy zones.
Now once you find those interphases.
What do you do?
Let's go to this slide.
>> Well, in a lot of cases we're talking about them moving into these
high resolution soil coring methods,
where you're subsampling a core that you get out a really fine detail.
So some nice pictures here.
A lot of these are courtesy of the University of Guelph who we've worked with
a lot in the past in terms of doing some of these methods.
But when you're going out, you're collecting as high a quality core
as you can so that showing in aluminum tube or other type of liner.
Get that core out splitting it lengthwise and then [COUGH] wrapping it in foil this
could try to minimize in metal loss of containment that you'd see out of there.
And then your subsampling, your taking this little plug of soil and
in this case we're doing field preservation so we're using methanol in
the field in order to again minimise the amount of loss of PUCs that you'd see.
>> Quick question, you did the sort of study that this,
one of the analysis you set, how important is that Methanol preservation?
>> Yeah, it's pretty important, you see,
pretty significant losses of several times in some cases that,
of these things just getting from fields of lab without preservation.
>> But here are you sampling the sands for
the plays or both or- >> Well you want to do both but you
really want to get high resolution, lots of samples once you move into that locate
moment because as we've said before things can really change at really small scales.
And that's the really goal in all of this cases the sample at the scale that's
appropriate to what you might expect some differences there.
So lots of samples once you move into that low K so.
>> Once you go past that interface.
>> Exactly. >> On the bottom right panel.
>> Exactly.
>> Let's look at some more images.
This is actually how you get this maybe your sampling of clay.
So what's going on?
>> Yeah so if you sort of look at this as a representation of core down there at
the bottom and you're taking these samples at pretty regular intervals.
Is shown a few inches in this case but
you've got some sort of sample that's designed to do this.
This is an example of a sort of a stainless steel sample that University
of Guelph uses but a lot of people are familiar
with soil core sampling that fires a lot of same principles but again,
the ideas we're trying to get samples that can capture all of the mass that's present
there in order to sort of to give us their complete picture of what's happening.
>> To measure that dark matter that's down there.
So I think a couple of pictures that University of Guelph team
I think right now we've got Steve Chapman.
>> Yeah, we talked about a little bit about Steven in past lecture but
here he is, doing some soil sampling here and
the upper right we see these cores laid out and you're sampling it at
pretty fine intervals and >> Or last in these cases.
>> Each one of those holes is where they took that plunger and
took out that sample, right?
>> Yeah, exactly and
so the bottom right is just an example of collecting these data on the field.
>> Don't need a special rig, right?
>> No. >> Just pull that core up you can do them,
you can do it with direct push.
>> Okay so you do all this and you can sort of put together these unique data
centers, so this is some stuff you did at Jacksonville Naval center station, right?
>> Yeah, yeah exactly we were collecting a lot of data at this, the various sources
and here's an example of a former dry cleaner that they had and we were looking
at chlorinated solvents in this case >> So
on our x axis is the concentration that we're seeing in the soil of PCE, TCE and
TCE and then we're looking at depth as you go down on that y axis.
And you can see that there's this clay layer that shown
overlay the geological log on, onto this one.
>> Okay, as you go down gradient here's this next one, right?
>> Yeah and we're moving down gradient with ground water flowing here.
And we really saw a lot of mass in the low k zones of this site.
And we didn't see necessarily a lot of daughter products or
transformation until you sort of move down gradient.
We also saw a shift away from the amount of mass that was stored in that low k
unit as you move down.
>> I notice the green line is increasing on this one, OU-35, that's a,
is that a daughter product?
>> That's sis, yeah.
That's SISDC.
>> The transformation product.
>> But occurring in this case, in the sand, not the clay.
And here's this last one, right?
>> Yeah. And so basically, a lot.
a lot of mass says you were in the source zone.
So consistent with the fact that,
a significant portion of the mass at this case, was associated with that clay unit.
>> Okay. This is coming from the SERDA project,
say the science support, I think that's ER174 40, right?
And then you actually put together sort of a protocol based on some of the work
that Gwelf had done and sort of some ideas.
If you're going to go to an unconsolidated site what would you do so?
>> Yeah.
>> A lot on this graph.
>> There's a lot on and we'll just sort of touch on this again.
sort of this step wise procedure and you have various options available to you that
are sort of on the top in terms of looking at screening level ways of
gathering data so you might use membrane interphase probe.
Or you might use the hydraulic profiling tool or
the Waterloo APS that Waterloo profile or
sample to get you real time data in order to make decisions about that.
And you'd follow up with detailed soil coring.
So that's shown in the upper right hand corner,
in order to get lots of depth discrete data about where that contaminant mass is.
>> Okay, at the end of the day you end up with the stuff in the middle,
these type of data sets that you can apply >> Really improve your conceptual
model, right?
>> Yeah, and you can maybe go onto the next slide then two with sort of laid out
a little bit of flowchart and again you can look for this information if you look
at the reports but various different ways that it can take you in terms of
gathering this data and the important element of this type of investigation.
>> Right, and I think there's a companion one done by Dr.
Parker at for fractured oxides.
>> Yeah, we model this off the work that she'd already done for
investigating fractured oxides.
>> And there is so many things you can do with that knowledge.
One benefit is, you can see this dark matter.
So here's this, again from this paper by Chapman and Parker Water Resources 2005.
2005 they've taken cores into their clays and they're looking at this dark matter.
So on the left we see here.
>> So that's depth going into that low K unit and in this case it was an aquatard,
so you're going down as you move on the y axis, from top to bottom.
>> And they're Canadian, so that's in meters?
Is that [CROSSTALK] >> That is in meters, correct.
And then you've got TC concentration, the soil concentration on the x axis.
So in this case, at these wells, they were gathering this data, so
each of those is individual data points, is a soil corn measurement that they
collected and then they were able to fit it to a fairly simple 1D model.
>> Based on Fick's law, right?
>> Based on Fick's law and
got a pretty good fit in this case with a diffusion based sort of model.
>> So this panels with DNAPL and they're really showing you the concentrations in
the gray zone in that >> In there, right?
>> Exactly. >> And
then they move out beyond the building way on this transect about 300 meters
down gradian.
Looks something like this and you have this shark fin.
Same axis here that depth into the clay out of this aquatard three meters and
then this Porewater TCE concentrations.
Concentrations about lower, right?
>> Yeah, yeah. And
again let's remember that there's a source here.
The sources upgrading has been isolated and so now we're seeing what's happening.
>> Down gradient and select different sort of profile of concentration versus
depth and this depth location on the right is one where
backed diffusion has been occurring in sustaining the plume above.
That's why we have the shark fin and either a lot anything is if you have that
shark fin shape we did a project with ESTCP to let you.
Allow to reconstruct a source history, what was the dissolved phase concentration
above any clay, if you know the shape of that curve.
And basically this shows that the concentrations in that transmissive zone
have been going down.
So this back diffusion's occurring on the right.
Dave, what's the source history on the left?
>> On the left are be consistent with sort of a more constant source history
where you're concentration gradient is such that they're still high concentration
that transmissive zone pushing through things down into that clay.
And so you see more of a consistent decreasing with depth into the clay.
>> Got it. And so
there's actually a tool we built spreadsheet tool you put in this data.
Dr. Shahla Farhat, you worked on this.
That you can then reconstruct your source history.
So lets just go, I think to do our key points here in terms of.
What we're looking at, that's sampling for
matrix diffusion really requires different strategies, different tools.
>> In step 1 in these cases is a lot of ways, hunt for interfaces,
find those areas where there's a lot of geological heterogeneity.
>> Okay, and then step 2 of that, once you find those interfaces.
Then, do high resolution sampling of this low probability zones to determine
the presence and concentration of contaminants in those low-k-zones.
And then, that data is going to give you a concentration versus
depths sort of profile in that information that you get within those low-k-zones
you can use to tell you a story about what's happened at that site.
>> So this lecture we talk about how you can sample these low permeability
zones to study matrix diffusion.
In our next lecture, we'll bring in Dr. Sheila Farhalf from GSI environmental, and
we'll talk about how you can model, using computer models and
equations, how you can model these matrix diffusion processes.
