There is always something left open in my mind.  I hope it’s not a real hole.

I don’t remember what made me think of this lately, but I’m still wondering about something.  In a gas mixture, who wins out:  gravity or diffusion?  Or possibly, do they each have their portion of the win?

Hey!  I’m not a physicist.  I’m an engineer.  This is actually a real problem.  Somebody must have the answer, or do they?

Over thirty years ago, this question actually showed up in my work as an engineer.  I had a clever answer for it, but now I would like as real an answer as possible — just out of curiosity.

Way back many years ago, I was working on the design of an oxygen generator that is now on U.S. submarines.  It was a technology that we bought in the middle of its development from another company.  It involved water electrolysis to produce separate streams of the two components of water:  oxygen and hydrogen.  In that sense it was nothing new, but the company that was developing it for the Navy seemed in the Navy’s eyes to be a bit slow in bringing the design out of the development stage and into production.

The new generator was a replacement for the previous design then in use on submarines, one developed by another company.  (By the way, I’m leaving out company names because there is no bad guy in this story.  And, it is not relevant in this context anyway.)

The new design operated at 3,000 psi using distilled water with no electrolyte.  And then it was noticed that there was another market for commercial low pressure generators as well, separate from the Navy contract.  And that is where the problem arose.  In the design of the Navy generator, I had taken great care to make sure the design was safe to use.  I didn’t want the thing to blow up.  Why would it do that?  Well, if you allow oxygen gas and hydrogen gas to somehow contact each other (after they have been separated) at the design pressure of 3,000 psi, there would be an explosion…no flame or spark required.  That danger decreases with decreasing pressure, but it is never far away.  In the case of the Navy generator, one of the design requirements that I insisted on was that the worst possible explosion had to be contained within the equipment itself.

And then one day, sometime after I was through with the design of the Navy unit, a different design group had developed a low pressure generator for a commercial purpose.  Unfortunately, there had been a slight lack of safety concerns in that design.  The design allowed hydrogen and oxygen to contact each other.  First, a single failure of a single barrier allowed the contact, and second, a catalyst was used in the system and had spread pretty much everywhere.  The presence of the catalyst caused the mixture to explode even though the pressure was only slightly above normal atmospheric pressure.  The third flaw was that the commercial system was not built to withstand an internal explosion.

I don’t remember how I got involved in the investigation process, but there I was.  One side of the argument was that leakage of hydrogen  was OK once we solved the catalyst problem, which was doable.  The problem that still existed was the possibility of leakage of hydrogen out of the system into the room.  To some, that didn’t seem to be a problem worth talking about.  After all, hydrogen is lighter than air, and it would simple rise to the top of the room and slowly dissipate given enough time.

And thus the question arises, is gravity the main player here, or is it diffusion?  Obviously, most of the people in the room thought that gravity would overcome diffusion, and the hydrogen would rise to the highest point in the room and not be a hazard…but would it really?  (For now we’ll ignore the question of the hazard it might be  at ceiling level.)

So, just considering the question of gravity overcoming diffusion, I wasn’t about to pin the safety of the system on something we obviously did not understand, but how to convince others?

Several of the attendees were smokers (people smoked in the office back then).  So I proposed that we run an experiment.  Everyone was to put out their cigarettes, pipes, and cigars.  I would go get a bottle of hydrogen.  I would open it on the table in front of us and let some out (avoiding suffocation with too much hydrogen).  We would then give the hydrogen time to collect at the ceiling level.  AND FINALLY, light’m up boys!  If they were willing to do that, I would go along with their opinions on the safety of hydrogen leaks.

Wisely, they caved.

And here I sit many years later, still wondering who won:  gravity, diffusion, or just a clever argument that assumed that in then end, nothing is what it looks like.  I’ve done a little searching on the web, but to no avail.  It seems to be that there is a consensus that in a 1 g gravitional field, gravity loses big time.

The right answer appears to be more like:  Why risk it?   Put your money, or in this case “your life,” where your mouth is!

And here is the joke of it all — I mean you have to ask —

So, does hot air really rise?