Tungsten Filament becoming brittle 1

[Mattcs]

I have been winding 0.25mm tungsten wire to make heating filaments. The wire is of course malluable before use (which is why I can wind it) and highly elastic.

After passing current through the wire (in a vacuum) the wire becomes very brittle and breaks very easily.

Why is this so? what processes are involved?

The filament holds a crucible so the filament breaking means that I lose my crucible as well.

I am trying to figure out how to prevent or reduce the risk of the wire breaking.

thanks

 

[JimMetalsCeramics]

The heat causes grain growth in the filament. At some point the grain boundaries can split (usually during a thermal expansion event -- i. e. turning the filament on). Fine MgO added to tungsten is supposed to pin the grain boundaries, but I don't know whose making that. Thoriated tungsten might suit you better.

 

[hacksaw]

embrittlement depends on how the wire was drawn.

typically it is the result of iron contamination in the drawing process.

the fact that you are using a mandrel to form your filament may be an additional source of contamination.

what is the cruciable material and what temperature are you trying to achieve?

 

[unclesyd]

I have worked with a process that uses small (0.009-0.015) Tungsten wire in air in which it got quite hot,900°F,for extended peroids. A rash of failures, at the 80% level promted a metallurgical analysis of the failures. The problem was with the manufacturer of the wire, they had forgot to anneal the wire after forming(drawing). After wire was annealed the failures stopped. I checked and the failure rate of the process is now and has been around 2%, most due operator error, for the last 5 years.

 

[hacksaw]

believe the application involves temps considerably greater than 900 oF and is being used in a relatively low stress application, but you've raised a good point

 

[Modey3]

I remember learning about this in my materials charaterization class. The heat causes the grains to coarsen and align so that the grain boundaries are perpendicular to the axis of the wire. Of course this gives terrible mechanical properties.
The driving force for grain growth is surface-tension forces between grains. JimMetalsCeramics was right in recommending using MgO inclusions to pin the grain boundaries against the sufrace tension forces. I hope this helps.

 

[cocobolorose]

Pure tungsten will recrystallize to an equiaxed structure @1200C, and in this condition is extremely brittle at RT. Lamp grade tungsten is doped with potassium (~80ppm) which impedes GB growth, raising the re-x temp to @ 2000C, and imparting ductility when cool. However, Fe, Ni, and Al on the surface can cause equiaxed morphology to develop, and brittleness, so keep the wire clean. Also, make sure there are no residual carbon-containing gases in the furnace to prevent carbide growth. If you notice that the windings have sagged, check for poor vac (i.e. oxygen contamination)
good luck

 

[SMF1964]

Addition to Modey3's note: tungsten with equiaxed grains is brittle. What makes tungsten filament so flexible is that it is not equiaxed in structure; the grains are far far longer than they are wide. If you then heat the filament to an incandescent temperature, you are recrystallizing the tungsten back to an equiaxed structure and a very brittle material. There are alloying additions to the tungsten that improve this by preventing grain boundary movement (essentially, the additives form oxides at the tungsten grain boundaries that pin the boundaries in place - or at least they did in 1987 when I did this term paper).

How does the crucible fit into the filament? When does the filament break? Is it in use or during cooldown (differential expansion of the crucible and filament)?

 

[chrisakel]

Rhenium is sometimes added to the tungsten to solve this problem. It has worked in light bulbs and receiving tubes.