Converting Seawater into Oil
This is an example of an experiment that we let run out of control - literally. Initially we set out to create a small laboratory process to create deuterium oxide (D2O or heavy water) using electrolysis to use in a neutron moderation experiment, and we decided to try out a set of carbon electrodes instead of the platinum ones we normally use (actually, we broke one of the platinum electrodes and had no choice but to use the carbon ones).
We used a standard Hoffman electrolysis apparatus to refine 100 mL of 'crude' D2O we had previously manufactured. To speed up the initial electrolysis process, we added 10 grams of NaCl and 2 mL of 10 mol HCL and filled the rest of the machine up with city tap water. Our power source was two 18 Ah sealed 12 volt cells connected in series (e.g., the potential difference between the carbon electrodes was 24 volts).
The test setup and experiment progress is shown in the photo slideshow below.
What we found after a day was an oily-scum forming on the negative (-) or hydrogen-generating side of the apparatus. We cracked open both the hydrogen (-) and oxygen (+) purge valves slightly to continuously bleed off excess gas and decided that this was far more interesting than making straight D2O, which we could do anytime.
After about two weeks, we had fully depleted the oxygen (+) electrode and had formed a thin, water soluble, oil-like substance which we tested in our mass spectrometer and found trace amounts of light hydrocarbons. We had in fact turned what amounted to tainted salt water into a light crude oil.
We filtered the substance several times using a vacuum Buchner funnel and tested samples in a centrifuge until we got a fairly precipitate-free sample. We then burned off the residual water using a 500 mL retort and a bunsen burner set to a low flame. The results as you can see is not unlike a light machine oil, although you probably would not want to use the stuff as a lubricant.