June 23: What’s up in the water?

Posted: Thu Jun 24, 2004 7:34 pm

Today we ran the investigation I described in yesterday’s post: a 12-hour lake experiment to determine how quickly pollutants (of one type) break down under Arctic waters (on one lake).

It has been an extremely full day, between labeling the samples, securing the samples to the lake reactor, taking samples, extracting the pollution from the samples back in the lab, transferring extractions to smaller vials for later analysis at Ohio State, etc.

Here is a picture of the loaded lake reactor.

The reactor measures about 35 cm by 50 cm (its base is a milk crate) and holds 24 quartz tubes filled with water and spiked with small amounts of PCBs (polychlorinated biphenyls). PCBs were commonly used in electrical equipment and in heat transfer devices, and as plasticizers in rubber, paints and plastics. When they find their way into the environment, they are persistent pollutants and are suspected to cause cancer, many eventually end up in Arctic waters. Here’s what the EPA has to say about this class of chemicals: http://www.epa.gov/pcb/

Taking samples is a relatively simple task.

Using a pair of wire cutters, we cut the plastic ties holding the tubes in place, store the quartz jars in a dark box and take them back to the lab. That’s all. (I said it was simple.)

Once in the lab, we have to follow a couple simple steps before we can analyze the samples:
a) we transfer the water into clean and empty vials using a clean pipette
b) add a bit of hexane (which the PCBs like MUCH more than water)
c) thoroughly mix the hexane and water (allowing as much PCB to leave the water and attach to the hexane as possible),
d) we use a second pipette to transfer the hexane (now laced with PCBs) to a smaller vial which will store the liquid until we are ready to test it on the gas chromatograph here or at Ohio State.

Important detail:
The gas chromatograph shows the quantity of chemicals present in our sample, providing us with graphs that show amount of chemical detected throughout the course of the 20-minute cycle on the gas chromatograph. The graph has peaks that we can measure and infer the amount of chemical present.

BUT we are less interested in the amount of pollution and more interested in the CONCENTRATION of pollution. How do we get from amount of chemical to concentration of chemical? Cleverly.

First, we inject several known concentrations of PCBs in the gas chromatograph and make a graph of the area of each of the peaks vs. concentration of PCBs (this graph is called the “calibration curve”). If the machine is working well, the calibration curve will be a straight line, with a relatively steep slope.

The slope of the line on the calibration curve tells us how responsive the chromatograph is to the chemical we are using. If you’re feeling mathematical, you could write the relationship between PCB concentration and peak area with a formula:
peak area = (slope on calibration curve) * (PCB concentration)

When we get the results from running our samples, we can rearrange the equation above to find the concentration of PCBs in our samples:
PCB concentration in sample = slope on calibration curve / peak area

Points for clever thinking!

(Of course, the concentration of pollution in hexane depends on how much hexane there was to begin with… that’s why we also measure the mass of the hexane we add to our samples to get the PCBs out of the water…. Well, really we measure the mass of the vial, measure the mass of the water we add to the vial, and then measure the mass of the hexane we add to the water in the vial.)

Don’t you feel smarter already?

Other curious and fun things:
This afternoon has been swampy. Rumor is that a tundra fire is contributing to a muggy haze that has enveloped the camp. Four Arctic terns and a Glaucous gull came calling. Squint just right and you can see them circling above Toolik Lake.