We have been asked for another example of a chemical analysis being performed on the ship. One of the most famous measurements is the determination of dissolved oxygen in seawater. The CTD has an electrochemical sensor that measures oxygen continuously as it is lowered into the ocean. However, the electrochemical sensor drifts with time and all CTD measurements must be calibrated against lab measurements. Samples are taken from the Niskin bottles on the rosette for the calibration of the oxygen sensor. The lab measurements are based on the classic Winkler titration of oxygen first developed by Lajos Winkler in 1888.
The Winkler analysis of dissolved oxygen has three steps designed to turn odorless, colorless dissolved oxygen into something that we can see.
Samples are collected in Winkler, or BOD, bottles that are specifically designed with a conical top to help exclude bubbles.
A basic solution of Mn(II), MnCl2 (aq), is added to the sample bottle. Under alkaline conditions dissolved oxygen will oxidize manganese(II) ions to manganese(IV), MnO(OH)2. This reaction is fast and stoichiometric so that each mole of oxygen produces two moles of Mn(IV).
2 Mn2+ (aq) + O2 (aq) + 4OH- → 2 MnO(OH)2(s) (1)
Excess Mn(II) and base are added in step one so that oxygen is the limiting reagent (see the blog on limiting reagents). Sodium iodide is also added during step one. In base, iodide does not react, but it is added to be ready for the next step of the reaction. You can get an idea of the amount of oxygen in the sample by observing the brown MnO(OH)2 precipitate that forms in the bottle. In this step we are converting a dissolved gas to a solid. The amount of solid is proportional to the amount of oxygen that was in the bottle.
The sample bottle must remain tightly capped during the first step to prevent oxygen from the air from reacting with Mn(II). Recall from the CTD video that all the air bubbles were carefully removed from the bottle during sampling and the Bottle Cop makes certain that the samples for dissolved oxygen are taken first. The capped bottles are allowed to sit for over 30 minutes to completely react with all the oxygen.
Next, excess acid is added to the bottle and the MnO(OH)2 (s) formed in step one reacts stoichiometrically with iodide (I-) to form a yellow, triiodide (I3-) solution in two steps. The cool part of this reaction is that iodide reacts in acid, but oxygen does not. This means that the samples will no longer react with oxygen from the air. Why is this important?
MnO(OH)2(s) + 2 I-(aq) + 4H+ → Mn2+(aq) + I2(aq) + 3H2O (2)
I- (aq) + I2 (aq) → I3- (aq) (3)
Again, iodide and acid are added in excess. The limiting reagent in each step is highlighted in red. Triiodide is yellow so we now have a solution that we can see. The more oxygen in the original sample the more yellow the solution. On the Melville, Melissa uses an automated titration system measure the triiodide by the stoichiometric reaction of triiodide with thiosulfate, S2O32-..
2 S2O32-(aq) + I3- (aq) → S4O62-(aq) + 3 I- (aq) (4)
When the solution returns to a constant color all of the triiodide has been converted back to I-. In the video the black box on the right monitors the color of the reaction. The tube entering the top of the bottle adds thiosulfate. It may seem easier to simply measure the intensity of the yellow solution. This would probably work for really clear water samples, but it doesn’t work for water samples that are murky or have some other source of color.
Notice that the samples can be handled in the air because step four occurs in acid. Melissa can process twelve oxygen samples in about an hour. She has processed over 600 samples during the cruise. That’s a lot of titrations!