Electromigration in soldering iron heaters

[ajb]

Looking around at some of the soldering station projects out there has me wondering about electromigration in heating elements when using DC supplies.  Theoretically, any direct current tends to cause material to migrate along the direction of current flow which could cause premature failure, and this is exacerbated by higher operating temperatures.  This naturally would seem to be a substantial concern for soldering irons given that the heaters will be operating in the neighborhood of 300C, and this would seem to be born out by the fact that Hakko, JBC, et al are still using AC to drive their irons (naturally price plays into that as well, I'm sure).

I'm curious about the extent to which this is a real concern, though, and in particular if it represents a reasonable risk of shortening the life of a heater/cartridge.  Most of the detailed discussions I've found refer to IC interconnects, where the small feature size amplifies the problem (higher current density and there's less material to be lost), but I haven't found much that seems applicable to the larger scales and higher temperatures of a soldering iron.  Clearly the construction and composition of the heater will affect it's susceptibility to failure.  I took a stab at using Black's Equation with little better than wild-ass guesses at some values and got an answer of ~9000 hours, but both the validity of the equation at larger scales and the reasonableness of my assumptions are suspect.

My gut feeling is that it's probably not likely to be a problem, but I'd be curious if anyone is sufficiently familiar with the physics and the construction of soldering iron heaters to provide more informed commentary.

[NiHaoMike]

It's negligible even in automotive light bulbs that operate at much higher temperatures.

The high end JBCs and Metcals are induction heated which only works with high frequency AC.

[Dago]

It's negligible even in automotive light bulbs that operate at much higher temperatures.

The high end JBCs and Metcals are induction heated which only works with high frequency AC.

JBCs are not induction heated, they just have a traditional heater which is integrated right at the tip with the thermocouple being a part of the heating element.

[Gyro]

I think the reason they use AC to drive their (Not RF induction, obviously) heaters is that it's cheaper. Why generate a high current DC supply when you can just use a Triac.

[ajb]

It's negligible even in automotive light bulbs that operate at much higher temperatures.

An excellent point, but it glosses over differences in construction and materials between tungsten filaments and soldering iron heaters that may or may not make difference.  I don't know what sort of resistance materials are used in the latter, or how they're assembled, which is a big part of why I asked.

Black's equation incorporates a term for activation energy, which Wikipedia specifies as "grain boundary diffusion" activation energy.  Those terms don't seem to lead to much useful comparative information for different materials, so I don't know what range of values we might be concerned with, but the term is in the exponent, so it has a substantial impact on the result.  Assuming, again, that Black's equation is valid in this realm.

As I said, I don't really expect that it's a concern, but I'm interested in hearing a bit more of an authoritative take on the subject.