[Twigg]

I smelled some ammonia coming from my lab room after I'd used some household ammonia a little while before. I went to turn on the lights to see if everything was alright, but the light bulbs flicked on for a split second then went dark and made a popping noise. They didn't turn back on, but the next day they worked just fine. So I thought that the ammonia must've been oxidized by the current in the bulb and disrupted the flow of electricity enough to prevent it from being turned on again (because fluorescent bulbs require a lot of energy to turn on). The problem is, those bulbs are supposed to be sealed. Can oxidation occur indirectly across such a distance or should I be worrying that the bulbs are not sealed (or not anymore)?

[Borek]

I doubt bulb problems were related to ammonia. My bet is you will see such problems more and more often till the bulb will finally die.

[Twigg]

Thanks Borek!

As a follow-up question, is it a bad idea to have fluorescent light bulbs in a lab in general? After all, they've got mercury in them and they make a big current with a big energy transfer at start-up. Has anyone ever had problems with fluorescent lighting in a lab?

I googled "ammonia lightbulbs" and I found a bunch of reports of rioters throwing ammonia-filled fluorescent light bulbs as explosives. It seems controversial whether this works or not from the articles, and I can't find any scholarly articles discussing such a phenomenon. The closest phenomenon that has been studied an documented is the oxidation of ammonia using UV light, Fenton's reagents, hydrogen peroxide, and sodium hypochlorite to remove ammonia from waste water as N2. According to one source¹, the process is just 2NH3 -> 6H+ + N2 catalyzed by HOCl. In short, ammonia is oxidized by hypochlorite to produce some chloride; the chloride gets oxidized to chlorine radical and then to hypochlorite by hydroxyl radical; the hydroxyl radical is produced by the ferrous-catalyzed dissociation of hydrogen peroxide in UV light (Fenton's reaction).

There isn't any hydrogen peroxide floating around in the air, but there is plenty of water and oxygen. This is a bit far-fetched, but bear with me:

1) H2O + O2 + hv -> HO2 + *OH
2) HO2 + H2O -> *OH + H2O2
3) H2O2 -> *OH + OH-
4) *OH + NH3 -> *NH2 + H2O
And so on. Doesn't seem likely to me, but after all the light bulbs did die out in the presence of ammonia (and I haven't had this happen with them before or since).

¹ Brito, Nubia Natalia de; Jose Euclides Stipp Paterniani; Giovanni Archanjo Brota; Ronaldo Teixeira Pelegrini. "Ammonia Removal from Leachate by Photochemical Process Using H2O2".

[Borek]

is it a bad idea to have fluorescent light bulbs in a lab in general?

Never heard about any reasons to not to.

After all, they've got mercury in them and they make a big current with a big energy transfer at start-up.

Old thermometers had mercury in them, that was not a reason to not use them in labs.

after all the light bulbs did die out in the presence of ammonia

Correlation does not imply causation.

[Twigg]

Good point Borek. I held up the open bottle of ammonia to the bulbs when lit and during lighting, and nothing happened. It was just the bulbs failing. Thanks again.