The Lone Ranger Strikes Out

This semester, I have the last chemistry lab I have to take before graduating: Analytical Chemistry. This branch of the science consists of two basic areas, which are qualitative analysis ("Does this sample contain chemical A?") and quantitative analysis ("How much of chemical A does this sample contain?"). Quantitative analysis is the more useful and more commonly used of the two, and in practical terms, the more difficult to perform. In some cases, you're looking for something in terms of parts per million, or less.

My class is something of a crash course in the discipline. We perform six experiments in six weeks, three of which require journal-style lab reports. This week's experiment promised to be interesting. It has been speculated in recent years that a link exists between ingested aluminum and Alzheimer's Disease. The purpose of this week's lab was to simulate the effects of cooking with aluminum pots and pans. We were to 'cook' a weak citric acid solution (similar in pH to many fruits and vegetables) in an aluminum pot for an hour, and then determine how much aluminum had been absorbed.

The procedure for all of this was a little involved. In addition to obtaining an aluminum sample from the cookware, we would also create solutions with known aluminum concentrations ranging from 10 to 100 parts per billion for comparison. We would then take each of these samples - five standards of known concentration, the sample from the pan, and an unknown provided by our TA's - and run them through a device called a spectrofluorophotometer. When compounds containing aluminum ions are mixed with a chemical called 8-hydroxyquinoline (8-HQ) in chloroform, the resulting metal ion complex becomes fluorescent when exposed to light of a certain wavelength. The higher the aluminum concentration is, the stronger the fluorescent effect is. The results from the known standards would be used to construct a calibration curve. We would then test the pan sample and the provided unknown to determine the amount of aluminum present in each. Piece of cake.

So I arrived a few minutes early to lab on Monday to begin boiling our sample and making our standards. My lab partner wasn't there yet, but since I was early, I wasn't concerned. I chatted a little with the other students as I got the necessary labware together - ten 100 ml volumetric flasks, 1 ml and 10 ml volumetric pipets, that sort of thing. About fifteen minutes after the lab period has started, my lab partner showed up and informed me that his class load was too heavy, so he had dropped the class. So I'm flying solo now, not only for this experiment, but most likely for the whole semester.

Preparing the standard solutions took almost all the lab period (12:55-5:00 p.m.) on Monday. I came to the lab today, made up the 8-HQ/chloroform solution, and began extracting the aluminum from the various samples I'd already prepared. Once the extractions were complete (about 10 minutes for each of the seven samples), I had the TA turn on the instrument - we never refer to it as a 'machine' - and got ready to run the tests.

It was a spectacular failure. My results looked like I'd pulled them at random out of the air. There was no correlation between the nominal concentrations of the solutions I'd prepared and the readings I obtained from the photometer. The TA checked to see that everything was set up properly, and it was. I ran the tests a second time, and got the same results. How exasperating! The TA was baffled; he said things had worked just fine when he had run the experiment, but he acknowledged that the extraction process "could make things dicey". He said he'd try it again over the weekend to see what kind of results he got. I asked him, with just a touch of frustration in my voice, what I could even do with numbers like the ones I'd generated. His suggestion was to talk to him again on Monday.

Great. The lab report's due on Wednesday.