How Many Materials Scientists Does It Take to Change a Light Bulb?

The answer is one, but only if she can wrestle the lamp out of the glove box, shove it into an antechamber that’s barely large enough to accommodate it, bring it out into the open, and hopefully not inhale any trace amounts of arsenic powder that are certainly on the lamp.

Then, she has to be able to unscrew all of the screws on the back panel that are too tight (because that’s what screws do to you when you need them out the most), pull out the dead bulb, put in the new one, realign the panel and re-screw the screws, shove the whole assembly into an antechamber that’s still too snug, and wrestle it back into its original position in the glove box.

Let me back up a bit, though.

For five days, I had been keeping my realgar samples under a lamp to degrade them. At 4 pm each day, I would remove one of my samples (prepared with Essie top coat and my French pepper grinder) from the light and place it inside a slide holder sealed with a binder clip. (Bear in mind this is all with my hands shoved into four pairs of gloves at once.)


On the day of the removal of my fifth and final sample, I strolled into the lab to finish off the first full trial of my experiment.

The lamp wasn’t shining.

I hadn’t touched the lamp in 120 hours, so why wasn’t it on?!

I immediately turned around and power walked back to the office. “David, the lamp isn’t on and I don’t know why.”

He follows me downstairs back to the lab and, based on my description of the scene (the lamp’s power supply was still on but read 0.0 volts, instead of the 11.5 volts we’d been keeping it at), he concluded that the tungsten-halogen bulb had burned out.


Tungsten-halogen bulbs are similar to regular incandescent lightbulbs, which also use a tungsten filament. Current passes through the filament, heating it up until it glows. This is part of the reason why tungsten, which has the highest melting point of all elements, is used.

But, because incandescent bulbs get so hot, the tungsten slowly evaporates over time, weakening the filament and coating the inside of the lamp. (This is why bulbs can gradually blacken.) Tungsten-halogen bulbs combat this evaporation with a little chemistry.

The inside of a typical incandescent bulb is filled with an inert gas such as argon or krypton. Tungsten-halogen bulbs include a little bit of a halogen gas, like iodine or bromine. When the tungsten evaporates, it reacts with the halogen instead of depositing on the inside of the bulb’s surface. When it gets hot enough (such as when it settles to the hottest part of the bulb, which happens to be the thinnest, weakest part of the filament) this resulting tungsten halide can dissociate, reapplying the tungsten to the filament and freeing the halogen to repeat this process.

Halogen cycle

Because of the high temperatures required for this tungsten-halogen reaction to occur, tungsten-halogen bulbs operate at much higher temperatures than regular incandescent bulbs (typically by several hundred degrees). As a result, tungsten-halogen bulbs also burn much more brightly. This makes them ideal for my particular experiment, which needs a powerful light source to induce realgar degradation.

But alas, just like in any incandescent bulb, the tungsten filaments in tungsten-halogen lamps are susceptible to failure. This seems to happen by a few different mechanisms.

One is the presence of some defect in the filament that causes the temperature to be slightly higher at that spot; the tungsten evaporates from this “hot spot” more quickly, thereby weakening that portion of the filament, further increasing its temperature relative to the rest of the filament, and so on. Eventually, the temperature of the hot spot reaches tungsten’s melting point, and the filament fails.

Melted tungsten filament

Melted tungsten gobbed up on the end of a failed filament

Another reason is that the grain boundaries (see below) can slide past each other under the influence of stress (such as that caused by the filament’s own weight or the high currents passing through). Movement along grain boundaries can also eventually lead to failure.

(Quick explanation: most samples of metals are polycrystalline, or made up of many small crystallites, and the borders between these crystallites are called grain boundaries)

Grain boundaries

Grain boundaries in a tungsten wire (above) and in a nice diagram (below)

Grain boundary filament

Cracking along a grain boundary in a tungsten filament

Some of this is presumably what happened to my bulb, which had been in my lamp since last year (aka, for a while).

If you read the first two paragraphs of this post, you know what happened next. If you want a recap, here’s one in GIF form:

David fixing bulbs upright

Several times during the whole process, other people from the group would walk into the lab. Then they’d see the masks over David’s and my noses and the thing that was once in the glove box no longer in the glove box. Then they’d decide that maybe it would be best to come back later.

After the lamp was safely fixed and back in its rightful place, David and I decided that it would be a good time to do some glove box renovation.

We laid down a bunch of sheets of clean room sticky mats (think flattened-out lint rollers) to help collect any fallen realgar powder and mitigate the amount floating around everywhere in the box. We also laid some down immediately outside the glove box’s antechamber to make transferring realgar slides (or possibly lamps) in and out of the glove box a bit safer.


David’s been talking about installing sticky mats since I first got here a month ago, and it finally happened.

Maybe some good can come of having to change a light bulb contaminated with arsenic.


Microscopic tungsten filament images courtesy of

O. Horacsek, “Properties and failure modes of incandescent tungsten filaments,” in IEE Proceedings A – Physical Science, Measurement and Instrumentation, Management and Education – Reviews, vol. 127, no. 3, pp. 134-141, April 1980. doi: 10.1049/ip-a-1.1980.0023

Peet, M. J. (n.d.). Tungsten Filament Lamps: A Case Study (Rep.). Retrieved July 26, 2016, from

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s