Yesterday, I had a curious afternoon weighing a sphere. The sphere in question is a rather perfectly made object manufactured out of two hemispheres so that when they are bolted together the internal surface encloses a highly perfect spherical shape. We are hoping to determine the average internal radius of the sphere by filling it with water and weighing it. However we had severe problems with the water leaking out of gaps through which it should not have leaked. We think we have it sorted now and earlier in the week we tested it with a helium leak tester.
To do this we filled the sphere with helium gas. Helium gas has light and mobile molecules and they can be guaranteed to find the tiniest of leaks. We then use a ‘sniffer’ outside connected to a mass spectrometer which can very sensitively detect helium exiting the sphere. Actually its all a bit trickier than this – but that is basically what happens. We left the helium in the sphere until just before the weighing when it occurred to me that the helium might affect the weighing, and so I quickly decided to flush it out. However for some reason I chose to flush it out with argon gas rather than compressed air.
After flushing it out for 10 minutes we took the sphere to the rather tasty weighing machine at NWML. As my colleagues set up the balance, they noticed the sphere – which weighs 9 000 g was 300 milligrams (0.3 grams) heavier than the last time weighed it. That may not sound like much – but we need a measurement uncertainty below 0.000 1 grams. After a moment the penny dropped – it was still full of argon. Argon molecules have a relative atomic mass of 40 compared with the average mass of an air molecule of about 28.8. So the 1 litre of argon in the sphere weighed about 1.67 grams rather than 1.2 grams that 1 litre of air would weigh. The difference (0.47 grams or 470 mg) was plenty large enough to explain the observed discrepency if we imagined it also contained a bit of air.
So first I tried using a ‘puffer’ to put air into the sphere, and after re-weighing the mass had fallen by 30 mg. Eventually I had to hold the sphere upside down and shake it in order to empty out the argon while my colleague used the puffer to flush air into the sphere. Eventually we seemed to get rid of most of the argon and the weight of the sphere returned to within a few milligrams of its previous value.
Anyway. I just thought I would mention this because (a) it was a curious moment and (b) it really brought to light very graphically the heaviness of gases in a way in which we are not usually aware – by simply weighing them!