Preparing an inert atmosphere reaction vessel with only an input?

One time, the graduate student that I work with was showing me how to prepare an inert atmosphere in a double necked round bottom flask. After drying it in the oven, she used a torch to drive out all the water vapor. Then she put septums over both openings, and inserted a needle connected to the nitrogen input through one of them, but with no outlet! I asked her to explain how any gas could escape in this set-up, and she tried but I was still very confused.

2 Answers

Her setup was wrong. There needs to be an outlet, the simplest solution in this case being a second needle with nothing connected to it, puncturing either septum to act as a vent to the atmosphere. After a while, depending on the volume of the flask and amount of nitrogen flow, most of the atmospheric air is removed.

This procedure is, however, quite inefficient, and likely unsuitable for particularly sensitive reactions. While most oxygen and moisture is quickly removed, it becomes exponentially slower to remove the last remnants. A much quicker and more reliable method is to perform vacuum/nitrogen backfill cycles. Three or four cycles will fully replace the interior atmosphere with nitrogen gas of the same purity as the input. This can be performed using a single needle and a Schlenk line (or any system connected to nitrogen and a vacuum pump with a intermediating stopcock), which essentially switches the same needle between a gas inlet and a gas outlet.

Just to give a practical example to further illustrate Nicolau’s point: My lab typically relies on an argon balloon at the end of the Schlenk line. When a fresh balloon is used and all the joints are properly greased (and any rubber tubes in the apparatus are still fine) and a flask is connected to the line with septa covering any outlets, the balloon typically stays filled overnight. If it has a good day, it’ll stay filled for two nights.

In a typical evacuation/backfill cycle, you can actually see the balloon getting slightly smaller when backfilling (assuming you have a larger flask than $25~\mathrm{ml}$. This is about as much as one looses to minute holes overnight. So it takes about 18 hours for the entire line to loose as much gas volume as a single flask has. No way that her reaction was water- and oxygen-free.