My glove box experience should probably be measured in man-years and based on this I will call the proposal to wear a lab coat while working in a glove box counterproductive and bothersome.  If you use a glove box for hours your arms get disgustingly sweaty, particularly in the summer months, and the last thing you need is some non-form fitting and insulating sleeves adding to your misery.  The added discomfort can no doubt lead to or compound rash decision making within the box that leads to an accident.

Other PPE is, of course, required. Safety glasses in particular.

And now I will present a real life example that makes both sides of the argument.  Spoiler alert: risk assessment.

I previously discussed a laboratory explosion that fully engulfed a double fume hood after someone poured ethanol into an unlabelled chromic acid waste bottle. The two victims besides the person who did the pouring were splashed with chromic acid.  One was walking out the door to go home at the time. He was not wearing glasses since he was on the way out, but it happened behind him so the back of his leather jacket took the bulk of it.  The other was working in a glove box that was perpendicular to the hood so he had chromic acid splashes to the upper back and side of his neck. I cut his shirt/sweater off him, a trick I learned in a previous incident.  Would a lab coat have made any difference for the glove box worker?  No.  His back was essentially unaffected despite being splashed, but his neck and the side of his face had some pretty good small burns.  Would it have made a difference for the worker who was leaving? Absolutely, if he wasn't already wearing a leather jacket.  Or for another bystander had there been one. This is why we wear PPE such as lab coats and safety glasses in the lab even if we aren't "doing anything".

By that principle, shouldn't the glove box worker also wear a lab coat?  I suppose the magnitude of the splash could have been greater which means a lab coat could slow down the absorption of the material enough to make a difference.  But one has to ask what the probability of something like that is versus the concerns I listed in the first paragraph.  A lab coat would not have covered any additional skin area on this worker. I believe the probability that a similar incident with enough material that the added layer of lab coat could make a difference is is not commensurate with the tradeoffs discussed in the first paragraph. In this scenario, my risk assessment disfavors a coat-required policy.  I find that experimental screwup from irksome discomfort is much more likely.

However, in another lab with other hazards or greater quantities, the risk assessment may differ, of course.  If fewer hazards or less material are present, the risk assessment more strongly favors a no-coat required policy for the glove box, and if there are more, then perhaps the risk assessment asks why are we trying to deal with a splash on a worker when we could prevent the exposure in the first place by changing our processes/materials or by installing a splash shield or divider between the box and the hazard. 

Rob Toreki