[paul.fr]

why do halogen light bulbs carry a warning not to touch them. not only is it quite difficult not to touch the bulb when changing it, but the warning is not specific as to any reason why it should not be touched.

is it the coating that may contain something harmful if you touch it then put your fingers in your mouth? or that it may have some effect on the life of the bulb?

[another_someone]

The problem is that halogen bulbs can get very hot when they are operating, but they need to heat evenly, and if you leave traces of grease on the bulbs from your fingerprints, it can cause uneven heating when the bulb is in use.

[eric l]

Strictly from memory.  Halogen bulbs are not made from common glass but from a more temperature resistant quartz-type glass.  That is because a halogen bulb reaches much higher temperatures than a "common" bulb would.
If you touch this quartz type glass, some of the fat and protein from your fingers will stick on it, and start to etch it at this high temperatures.  This causes weak spots and eventually a shorter lifetime for the bulb.
Of course you should never touch a burned out bulb before it has really cooled off.
I also have the experience with the tube type halogen bulbs that they should work in a horizontal position, otherwise you shorten the lifetime.  This may be due to uneven temperature distribution if the bulb is not working in a horizontal position -unless someone comes up with a better explanation.

[another_someone]

I too recollect that the glass is indeed high temperature quartz (hence why they are sometimes referred to as quartz-iodine bulbs), but I have never come across a case where there has been a failure of the outer envelope.

The idea is halogen bulbs (again, from memory), is that ordinary tungsten bulbs, the hot filament will evaporate and then condense on the inner surface of the glass envelope.  As the filament evaporates, it shortens the life of the bulb, and in order to extend the life of the bulb, the temperature of the filament is limited.

With halogen bulbs, a small amount of halide is introduced into the bulb, and this reacts with the tungsten that has condensed on the inner surface of the glass envelope, and allows it to be recycled back to the filament (but, ofcourse, it is important that the process is even, otherwise some parts of the filament will receive more recycled tungsten at the expense of another part of the filament).

In order for this recycling to happen, the temperature must be kept high; but because of the recycling process, the temperature may be kept higher than would be normal for an ordinary tungsten bulb (although the high temperature is also maintained not only by having a hotter filament, but because the envelope is smaller, so the glass is closer to the filament, and thus kept hotter).

Again, as I say, this is all from memory (and it is too early in the morning for me to remember anything right now).

[eric l]

I suppose you are right, that temperature differences due to spots on the glass are more likely than failure of the glass due to etching.  But touching a new or cold bulb with your bare fingers is bad for the bulb.
Occasionally I have to change bulbs in theatre projectors (e.g. switching from a 1000 W to a 1500 W).  I use a new plastic bag I put over to bulb to retire it from the fitting (and put it back in storage).  Of course this can only be done on cold bulbs.
Anyway, working on a projector with a hot bulb is a bad idea as hot filaments are more likely to break.

[lightarrow]

Maybe it can also depend on the fact that the organic film on the glass surface absorbs light more than glass, increasing the glass temperature and so reducing its life.

Another consequence should be the fact shorter wavelenghts are absorbed from the film, reducing the whiteness of the light emitted from the lamp.

[anthony]

A friend of mine recently had this experience with a car-bulb that had been fitted with a finger print on. The bulb cracked after a week's use, the fingerprint having "burnt" itself to a black carbon mark. I was unaware that car bulbs were now halogen, it's possible that this bulb wasn't of course, in which case it's a common feature.

I think the failure occurs due to an uneven temperature at the glass surface, as the burn conducts and radiates heat differntly. The glass is very thin and the stress induced due to different expansion due to temperature cracks the glass. Rather like the hot water in the crystal tumbler chestnut.

My memory, and I don't want to contradict for contradictions sake because it's subtle, was that the reaction between the tungsten and the halogen occurs in the gas-phase, the resulting compouind is preferentially deposited onto the filament.

It's also worth noting why halogen bulbs need to be hotter. It's because this allows them to give out much more blue light. A normal bulb gives out whacking amounts of red light but it drops off substantially through the green to yellow. Halogen bulbs drop off less, but still give out very little light by violet. That's why a halogen bulb will always give a whiter light than a normal bulb - check it out on a sheet of white paper.

If you were really clever you could probably use the effect to see a blue colour around the normal light area. The blue wouldn't be there of course, it would be a trick of perception.

[that mad man]

Another_someone has summed it up correct.

Any fingerprint pollution can cause localised hotspots and uneven temperatures inside.
This can cause the filament to burn unevenly, it rarely causes the quartz glass envelope to fail.

Bee