We tested different application methods of activated carbon for sediment remediation

This blog post is based on our recently accepted publication in Water Research (Vol. 114, p. 104-112; http://dx.doi.org/10.1016/j.watres.2017.02.025). It is available free of charge until 15. April 2017 under this link.

Activated carbon is a sorbent with the capability of strongly binding pretty much any organic substance to its surface (a process called adsorption). Since a large share of pollution in aquatic ecosystems concerns such organic substances (for example PCB’s), it would hence be a suitable sorbent to remediate them. Once a pollutant is adsorbed to the activated carbon, it is bound so strong that it is no longer available to organisms. This includes even the case where they eat the activated carbon particles “coated” with the pollutant. The organisms would just pass it through their digestive system, pooping it out unaltered. So, long story short, the idea is to render pollutants harmless to the environment, rather than having to remove them (which additionally leaves the question where to put the removed pollutant).

Unfortunately it has been shown that activated carbon itself can actually be quite harmful to certain animals. Therefore, it is necessary to not focus solely on developing these novel remediation methods to be as effective as possible, but to ensure that they are also safe to apply in the environment. After all, what does it help us if we treat the pollution in a place, but at the same time wreck its ecosystem?

In the paper this post is based on, we mainly examined several different methods of applying activated carbon to polluted sediments (which is where the major share of pollutants in aquatic ecosystems are). You basically have two general options: the more laborious one of mixing the sorbent into the sediment actively, or the more “crude” way of thin layer capping. In the latter method you just cover the polluted sediment with the activated carbon (see picture 1). In the field that would mean all you need to do is to take a shovel and spread the carbon. So, while we did know that thin layer capping would be the easier method to execute, what we aimed to find out in our tests was how it compares in matters of effectiveness and safety.

Picture 1: The setup of our test vessels (only thin layer cap tests shown). The activated carbon is applied as a thin layer on top of PCB polluted sediment. Underneath, the burrows of the test organism (Lumbriculus variegatus – a worm living in the sediment) are visible.

We simulated the two application methods in the laboratory in test vessels containing sediment from a PCB-polluted site (Lake Kernaalanjärvi, southern Finland). As a test organism we used Lumbriculus variegatus, small worms that burrow through the sediment. The amount of PCB’s that the worms take up from the test sediments told us how well the different treatments work for remediation, while their biological responses (things like their change in body mass) were used as parameters to measure the adverse effects of the sorbent material itself.

Picture 2: A quick graphical overview on the results of our results: Adverse effects can be comparably high, but remediation efficiency (meaning the reduction of the uptake of a pollutant (here: PCB) is best when activated carbon is mixed into the sediment.

The major results published in this paper were both promising and worrying at the same time (picture 2). We found out that both methods are effective in general. Worms living in sediment under a thin layer cap took up ~50% less PCB’s from the sediment than from the untreated, “raw” sediment. When the activated carbon was mixed into the sediment, the uptake of PCB’s was prevented almost completely. So, while thin layer capping is a method that is a lot easier to use (and hence cheaper), it is not quite as effective as mixing the sorbent into the sediment. Nevertheless, one has to also keep in mind, that animals dwelling in the sediment (and the thin layer cap) can mix the activated carbon with the underlying sediment. This process is called bioturbation and it was actually even visible in our laboratory test vessels (picture 1). It’s just a lot slower than mixing sediment and sorbent right away upon application. In addition, mixing via bioturbation of course requires animals to stay on the treated site and not to flee the site in panic when the activated carbon is applied.

And that’s exactly where our more worrying results come in: the adverse effects of the sorbent itself. With both application methods it became quite apparent that the worms did not really like our miniature-scale remediation works. They lost their appetite almost completely, stopped feeding and hence lost a lot of weight. While that may sound like a desirable achievement to some humans, for our worms that could be a serious issue.

A possible explanation for this sudden loss in appetite was found on electron microscope images that we took (picture 3). It looked like the activated carbon had quite some detrimental effect to the worms’ gut walls. Their microvilli, which are responsible for nutrient absorption from the gut content, were damaged severely in most worms exposed to sediment treated with the sorbent – no matter with what application method. The exact mechanism on how activated carbon causes this kind of damage remain obscure; one suggestion for example is mechanical abrasion (the carbon particles are quite sharp), but also the strong sorption capacity of the material might be involved.

Picture 3: Activated carbon can damage the gut walls (specifically the microvilli) of Lumbriculus variegatus. Image as seen through an electron microscope at a 6000x magnification.

One interesting thing we saw was that thin layer capping with activated carbon can have quite a devastating effect on Lumbriculus variegatus. This is not too surprising, since the organisms are exposed to a high dose of pure activated carbon at the sediment-water interface. However, when we mixed the activated carbon with clay before applying (thus creating a thin layer cap that resembles natural sediment that is enriched with the sorbent), the adverse effects were a lot less severe.  This doesn’t mean there were no more adverse effects, but rather that they were at a comparable level to our other tested application method of mixing the activated carbon into the sediment.

Picture 4: The transmission electron microscopy images (which you saw above) in the making. Photo: Inna Nybom.

From the results seen in this study we were able to draw some conclusions and implications for future field applications. To sum up, both methods are effective. What the thin layer capping method lacks in immediate effectiveness, it makes up for with its easier application and lower costs. When it comes to the adverse effects, we showed that neither one of the methods has a significant advantage over the other – if certain precautions, like avoiding to apply pure activated carbon, are made. So when deciding on a method, the important factors are mostly the available budget and equipment. Thin layer capping is a better option for sediment remediation in cases where special equipment required for other methods cannot be brought in easily (remote areas) or simply in cases where funds are limited. However, before deciding whether or not to utilize activated carbon in general (and big scale), we will have to make sure that its own adverse effects to the environment are not worse than the pollution effects!

Lastly – if you check our blog post on the first field trial of activated carbon based sediment remediation in Finland, you will probably spot some of these implications already “in action”!

Text: Sebastian Abel

Pictures: Sebastian Abel, Inna Nybom

Sediment field clean-up trial, Part 2: setting up and monitoring the actual field trial

In Part 1 of this Blog post we took a look at the on-site sediment remediation with activated carbon. Now we will gain a small insight to the first field trial of the method in Finland, which was started by our group in August 2015.

The test site lies in Lake Kernaalanjärvi, which was contaminated with PCBs between 1956 and 1984. There was a steady, unnoticed discharge of the chemicals from a paper mill upstream one of the lake’s feeding rivers (Tervajoki). Since no one noticed that leak for so long, quite an amount was released to the river and ended up in the lake eventually. The fact that this is still a problem nowadays, even though the leak was shut down over 30 years ago, gives you a hint on the persistency of PCBs in the environment.

Lake Kernaalanjärvi, the site of the field test in winter (Plot marked). The feeding Tervajoki River – the source of the pollutants (PCBs) in the lake – is nicely visible.
Lake Kernaalanjärvi, the site of the field test in winter (Plot marked). The feeding Tervajoki River – the source of the pollutants (PCBs) in the lake – is nicely visible.

Since the lab trials had not only shown the high efficiency of activated carbon, but also potential risks of the sorbent particles themselves, we applied it only to a small plot of 300 m2 within the lake. This way we don’t mess up a whole lake, if the side effects are bad, but we also don’t waste too much money, if the clean-up potential is not as good as seen in the lab. The plot lies in the south end of the lake – the most contaminated area. This is right where the contaminated feeding river enters into the lake (see satellite image). With it come the PCBs, usually attached to suspended particles that settle as sediment once the water flow speed gets low enough.

For the remediation works, we ordered about 1000 kg of pressed pellets consisting of activated carbon and clay (Sedimite™). The latter increases the density of the pellets, making them sink faster to the bottom of the lake. This fast-sinking property makes handling and applying the activated carbon really easy. You can basically just shovel them out of a boat onto the water surface and they sink straight down onto the sediment. With pure activated carbon – usually a powder – that would be unthinkable. In part 1 of this post you could see a picture of the mess we can easily create already in the lab when we handle activated carbon powder. Add a bit of wind or rain (which we rarely see in the lab) to that and you might not have the greatest work day of your life. Even if the powder finally reaches the water, most of it would just get suspended in the water column and settle after days at wherever the water flow brought it.

The boat is loaded with activated carbon pellets (Sedimite™) – we’re ready to get going!
The boat is loaded with activated carbon pellets (Sedimite™) – we’re ready to get going!

For the application of activated carbon we had to first of all change our “lab rat” attitude to field-trial-mode: In the lab, we usually work with precisely measured doses, carefully applied in controlled environments. In the field, we had to take more of a “rough estimate” approach. We started by measuring a 10 x 30 m field on the lake, marking it with buoys and ropes. For better orientation and to achieve a more even layer of activated carbon, we diverted the plot into 5 x 5 m intersections, which were handled one at a time. After we had applied (read: shoveled) all of the pellets onto the test site, we took some sediment core samples of the freshly covered site. Luckily we could see that the pellets had actually worked as intended and we achieved a quite good layer of activated carbon on top of the sediment.

Applying the pellets to the marked plot on an average Finnish summer day.
Applying the pellets to the marked plot on an average Finnish summer day.

Now – about one year later – we checked in to see how the field looks like. We took core samples on the same spots again and unsurprisingly, the field looks a lot different. Wind and waves have affected the plot heavily:  a lot the sorbent has been swept away. In addition, a thick layer of new sediment has covered what was left on site.

Sediment core samples showing the applied layer of activated carbon one day after the plot setup and 10 months later.
Sediment core samples showing the applied layer of activated carbon one day after the plot setup and 10 months later.

How this is affecting the remediation potential and the adverse effects of activated carbon, we plan to find out in the near future. We have scheduled a lot of monitoring works, such as surveys on the condition of the local sediment fauna and changes in the PCB uptake by the organisms living on our plot.

Important part of every exhausting field trip: Beer and Sauna.
Important part of every exhausting field trip: Beer and Sauna.

Text by Sebastian Abel, photos by Sebastian Abel and Jarkko Akkanen

The first Finnish field trial of on-site sediment cleanup with activated carbon

Part 1: the method and preceding lab work

What you might think of when hearing about sediment clean-up (remediation) is the conventional method of dredging the contaminated material and depositing it somewhere else (off-site methods). But did you ever try grabbing a fistful of mud from under your feet when you’re standing in the water? Not so easy! You usually manage to get some to the surface, but what about all that slurry that stays suspended in the water? In sediment remediation, this can easily cause even more trouble, since it leads to increased dispersal of contaminated material over the water body, as well as increased exposure to everything that has to swim through the water-sediment suspension. Besides that, an excavator vessel is not the cheapest thing to rent either.

Activated carbon -based “on-site” remediation has been proposed as an alternative method. The basic idea is to add the activated carbon as a sorbent straight to a contaminated site, where it binds the contaminant so strongly, that it becomes unavailable for organisms to assimilate and accumulate. So while the pollutant is still in the sediment, it is rendered mostly harmless. It works pretty much the same way as medical activated carbon: The poison that you accidentally ate is bound and thus prevented from entering your bloodstream, from where it could cause havoc. The only difference in sediment remediation is that this sequestration of contaminants happens already before they are taken up by an organism. A more detailed description of the method and its mode of action you can find here.

Testing activated carbon for sediment remediation in the lab.
Testing activated carbon for sediment remediation in the lab.

In our current research we are focusing on the use of activated carbon to clean up sediments polluted with PCBs.  This group of chemicals that is found in the environment of most parts of the world. Listing all the uses and potential dangers of these PCBs in the environment would probably fill another blog post. In brief: it was seen as harmful enough for a worldwide (!) ban of production and use in 2001. One of the biggest problems with PCBs in the environment is their persistency and the fact that they accumulate easily in organisms that are exposed to it.

This is where activated carbon enters the stage: many researchers, including our own group, found that already small doses of activated carbon suffice to prevent almost any of this accumulation of PCBs. So you might say: “Great! It sounds like a great alternative to the messy and laborious dredging operations”. But as Bernard Shaw once said “Science never solved a problem without creating ten more” – we also found that activated carbon itself might have negative side effects to certain organisms. Our job is now to find out if the new problems we create are actually worse than the original one, or if they are a minor trade-off. Our lab studies showed a relatively “balanced” situation, showing both high remediation efficiency accompanied by strong adverse effects. However, lab studies are always limited in their meaningfulness, because we are bound to exclude a lot of parameters that make up a natural environment.

Working with activated carbon powder in the lab can be pretty messy.
Working with activated carbon powder in the lab can be pretty messy.

Therefore the next logical step was to bring the tests of activated carbon based sediment remediation to the field. So in August 2015 our research group has set up the first ever field trial in Finland aimed at investigating the potential and the risks of this method. How this looked like and worked in detail, you can find out in the second part of this blog post.

Text by Sebastian Abel, photos by Sebastian Abel, Jarkko Akkanen and Inna Nybom