Environmental Water Analysis

This is an appropriate lab since many of my students are environmental science majors. In this part of Pennsylvania, we have the misfortune that much of our surface water is polluted from coal mining. The problem is a classical phenomenon known as acid mine drainage. The Tioga River, which runs through Mansfield, enters Tioga County some twenty miles upstream as pristine as rainwater. In the first ten miles of its course through the county, it traverses an area that was strip-mined as recently as the 1980s. By the time it reaches Blossburg, it is so saturated with iron that iron oxide deposits heavily in the river bed; the pH is below 4. Tributaries bringing mine drainage to the river have pHs ranging from 2.8 to 3.4.

Dr. Arnold George, now retired from my department, is a wealth of information about the problem, and has made many suggestions about how to incorporate study of this type of pollution into the lab curriculum. I have approached this lab exercise from different points of view in the past, but I think I have settled on a format that will be instructive about the problem, how it may be corrected, and the chemistry behind both.

If you are fortunate enough not to have access to polluted water, this exercise may be of little use to you. You could, of course, provide students with "synthetic polluted water" (not an approach that I really like, but sometimes necessary). I don't know the exact concentrations you would need to use, but calcium and ferric iron and sulfuric acid would give you all you need to see the results we get.

The chemical tests we use vary across the spectrum from qualitative to quantitative and are as follows:

Conductance: Students are told this is a nonspecific test for ions. Various devices can be used. The best situation is to use a truly quantitative conductance meter, but these are not very common and are relatively expensive. On the simpler side, students can use a device like that suggested in the lab manual for Chemistry in Context (series circuit with 9-V battery, 1 kilohm 1/4 watt resistor, LED) to compare the light intensity from water samples with that seen with distilled water and with 0.1 M NaCl.

I sort of compromised by designing and constructing a device that can be dipped into a sample in a small test tube and lights up to ten LEDs depending on the sample's conductance [see J. Chem. Ed. 77, 1606-1608 (2000)].

pH: Students are reminded this is a quantitative test for acidity. I have used both a pH meter and a pH indicator mixture selected to cover the pH range from 2 to 8. I prefer the pH meter, but I find that the most common pH electrodes can't be used with typical test tubes, since they are too wide (1/2-inch) to fit into 13 x 100 mm tubes and too short to reach the bottom of 16 x 150 mm tubes. If you are setting up and have money to spend, get pH electrodes of 3/8-inch diameter. With the devices I use, students can sequentially determine pH and conductance on the same 2-mL samples in the same tubes.

The combined pH indicator I have used contains 0.04% each of bromcresol green, phenol red, and methyl orange in 50% ethyl alcohol. One drop is added to a 2-mL sample. Standards are prepared by the instructor using a solution of citrate and phosphate with pHs ranging from 2.5 to 8 in 0.5 unit increments. Over most of the range, except from 5-6, pH discrimination to the nearest 0.5 unit is possible.

Hardness: Students are reminded this is a test for mineral ions that give precipitates with soap. When I first did this exercise, I used a simple, semiquantitative method. Students added soap dropwise to a 2-mL water sample, with vigorous shaking, until a stable foam forms. The soap is the same 10% Ivory in 70% isopropyl alcohol that we use in the "Surfactants in Hard Water" lab. Formation of curd may not be very fast, so students should wait a few minutes before concluding that suds did indeed form in any sample.

More recently, I have used a simplification of the complexometric method typically used in Analysis courses that is quite similar to the procedure used in the lab manual that accompanies Chemistry In Context. Students place 2 mL of buffered (see below) sample or buffered calcium standard (5.0 mM calcium chloride, equivalent to 500 ppm calcium carbonate) in test tubes with 2 drops of indicator (0.5% Eriochrome Black T in 95% ethanol). The color should be wine-red. EDTA solution (10 mM disodium EDTA in water) is then added a drop at a time with mixing until the color turns abruptly blue, and the number of drops is recorded. I tell them the concentration of calcium in the standard and they are to calculate the equivalent hardness of the unknown samples from their data.

Even though iron contributes significantly to the hardness of our samples, the above titration doesn't give a good endpoint if iron (and probably other base-insoluble minerals) is present; in the buffered samples, iron gives gelatinous precipitates which adsorb the dye. Iron and other interfering minerals are cleanly removed by filtering prebuffered samples. One volume of buffer is mixed with 25 volumes of sample and the solution is allowed to sit for several days. Iron appears to precipitate fairly quickly, but in some samples I found precipitate still forming after some hours or even days. Buffered samples are finally filtered before use. Buffer is prepared by dissolving 6.4 g of ammonium chloride, 58 mL of concentrated (28% w/w) ammonia, and magnesium chloride and EDTA at concentrations of 5 mM in a final volume of 100 mL.

Iron: To a 1.5-mL water sample, students add one drop of concentrated nitric acid (or see next paragraph) and 1.5 mL of 0.1 M ammonium thiocyanate. Our most iron-rich tributary samples give a medium to dark red-orange color. Other tributaries and the river give faint but visible orange colors.

These water samples, by the way, are not especially stable, and iron precipitation is visible within a day or two of collecting, so they must either be fresh or have been preacidified with nitric acid. I use 0.03 M final concentration, which is obtained by adding 7 mL concentrated nitric acid per gallon. When this is done, it isn't necessary for students to add the concentrated nitric acid.

The intensity of the orange-red complex can be quantified on a rough basis by eyeball, and that is how I had students do it for a couple years. Now they actually use a Spectronic 20 (at 460 nm) to get quantitative absorbance values which they compare with known values on a standard curve. I could have them construct a standard curve from actual known samples, but so far I've saved them some time by giving them approximate data as follows: 0 ppm iron, 0.000; 5 ppm, 0.298; 10 ppm, 0.601; 15 ppm, 0.897; 20 ppm, 1.188; 25 ppm, 1.491; 30 ppm, 1.793.

Sulfate: To a 2-mL water sample, students add one drop of concentrated nitric acid and 6 drops of 0.1 M barium chloride. Precipitates of barium sulfate are readily apparent. (As with the iron, preacidified samples don't require the students to add nitric acid.) For this test, I've had students simply report an eyeball estimate of the turbidity, which ranges from absolutely clear for pristine samples to quite turbid for the most contaminated samples. If a more quantitative determination is desired, the barium sulfate can of course be determined gravimetrically.

Remediation of this kind of pollution is challenging. While nothing has yet been done to clean up the Tioga River, steps have been taken to improve the water quality of another acid-impacted stream. Babbs Creek, some ten miles south of Wellsboro, is similarly polluted by mine drainage, but is part of a totally separate watershed, feeding Pine Creek, long a favorite trout stream whose trout populations were stressed by the acid inflow. At several locations in the Babbs Creek watershed, mine outflow is channeled through diversion wells, large concrete cisterns full of crushed limestone. This treatment raises the water's pH and removes iron both by precipitation directly on the limestone and immediate precipitation in the wells' outflow due to the higher pH. A newspaper report from around 1996 indicated that trout were being seen in Babbs Creek, which had previously been devoid of large animal life. This process serves as the basis of the second part of this lab exercise.

The remediation part of this lab consists of two parts. First, I bring in water samples from before and after the limestone wells for students to evaluate their effect on pollutant content using the above tests. Second, students mimic the limestone treatment in the lab by treating a 10-mL sample of our worst Tioga River tributary with two scoops of powdered calcium carbonate. After letting the solid settle, students determine pH, conductance, iron, and sulfate as above and compare with untreated sample. Although I haven't done this, iron should be recoverable from the solid residue after dissolution with acid. All in all, this is a pretty instructive exercise, with its connections to environmental remediation and basic inorganic chemistry.

In their write-up, students are asked to compare overall water quality among the samples they tested and to comment on which indicators of water quality improved in the samples treated either on-site or in the laboratory. I have also recently added a question illustrating a practical application of stoichiometric calculations: given a limestone load of, say, 800 kg, how many liters of mine outflow could be neutralized if the sulfuric acid concentration is, say, 0.0005 M?

Revised 9/1/06