Thermal COnduction

[TalShiar]

I am looking for a material with the following characteristics:

* Highly thermally conductive (ie, useable between a transistor and a
heatsink)

* Non-evaporating (can sit around for years with ought 'toughening up')

* Is a paste-like substance or at least flexible semi-solid.

* Viscosity in the 77K-343K (-196C to +70C) temperature range.

Any ideas or know where I could look let me know.
Thanks.

 

[rgwalt]

What do you mean by "viscosity"?

 

[CTho9305]

Originally posted by: rgwalt
What do you mean by "viscosity"?

probably something that doesn't drip.

 

[TalShiar]

A paste like substance or rubberish would be great. Something that won't freeze when in those low of temps. I need to allow for expansion and contraction between the two surfaces.

 

[zsouthboy]

What exactly do you need this material for, if I might ask?

BTW..... I believe the thermal paste called Arctic Silver III fits most of those requirments lol

 

[Evadman]

Originally posted by: zsouthboy
What exactly do you need this material for, if I might ask?

BTW..... I believe the thermal paste called Arctic Silver III fits most of those requirments lol

AS2 actually. You would have to modify it for the low temp though. That is a HUGE tempeature range. Good luck.

 

[FrustratedUser]

Actually a vacuum grease called Apiezon-N is a quite good conductor, especially down to 4K.
$120 / tube.

Not sure if it's good all the way up to +70C

 

[Shalmanese]

Diamond works on all thosepoints except the paste like one.

 

[TalShiar]

Thanks FrustratedUser, I will look into that. I am making a peltier cooler for aerospace apps. which is why I need it for really low temps. Those silver compunds will not work at a low enough temp, however I am in contact with some of those companies.

 

[Daovonnaex]

Originally posted by: Shalmanese
Diamond works on all thosepoints except the paste like one.

How could diamond possibly be a good conductor? It's a covalent network solid, which means it has neither mobile electrons nor mobile ions--the two things which lead to good conductivity. Perhaps you confused it for an ionic crystal (and those only conduct well in a liquid state)?

 

[Howard]

No. Natural diamond has a thermal conductivity of 2000 W/m-K. Same goes for synthetic diamond...

 

[Oodie]

Originally posted by: Daovonnaex

Originally posted by: Shalmanese
Diamond works on all thosepoints except the paste like one.

How could diamond possibly be a good conductor? It's a covalent network solid, which means it has neither mobile electrons nor mobile ions--the two things which lead to good conductivity. Perhaps you confused it for an ionic crystal (and those only conduct well in a liquid state)?

Strange, ain't it? Never the less, it's true. Diamond has a thermal conductivity 5 times greater than silver. It also has the highest hardness, melting point, and lattice density of any substance. (Regurgitated info from http://mineral.galleries.com/minerals/elements/diamond/diamond.htm)

 

[Weyoun]

I think he confused it for electrical conduction, as opposed to thermal. All of the properties he described fit the bill pretty well...

 

[Daovonnaex]

Originally posted by: Weyoun
I think he confused it for electrical conduction, as opposed to thermal.

Check that. Oops.

 

[Cogman]

Originally posted by: Weyoun
I think he confused it for electrical conduction, as opposed to thermal. All of the properties he described fit the bill pretty well...

yah, Umm, You dont hear much about them Diamond Conductors, as it is, infact, a resistor. They're right, Diamonds have a very high Thermal Cunductivity. Saw it on the news, Thats why synthetic diamonds are exiting, because they retain the same properties as diamonds, take less time to make, and are MUCH cheaper.

 

[f95toli]

Apiezon-N is a good choice. We use that down to 20 mK without problem.
I think it will work up to quite hight temperatures but I am not sure how high. We have a cryostat that covers the range 1.7K-400K and it uses SOME type of grease, unfortunately I don't know which kind.

Another way to do it is to find some suitable vacuum grease and mix it with fine copper powder. It works quite well but it is of course a conductor so keep it away from any wires.

 

[TalShiar]

Hey, that Apiezon-N looks good. Thanks.

 

[Degenerate]

so is VERY FINE powdered diamond good?

 

[Howard]

How would you hold it in place? And if the heatsink were moved, everything would get scratched up.

 

[f95toli]

We have a "cheap" version of thermal grease that we use sometimes. It consists of copper-powed mixed with vacuum-grease (I think any mineral-grease would work). It works but unfortunately it can be electricaly conductive so one one has to be careful not to short anything.

 

[silverpig]

Originally posted by: Degenerate
so is VERY FINE powdered diamond good?

I wouldn't think so. There would be a lot of air in between the particles, and air = bad.

Besides, I think the conductivity of diamond rests on the fact that it's atoms are all packed very closely together in a regular pattern.

 

[Cogman]

Now, If you could make a heatsink out of diamonds Might use Thermal compound on it, but I dont think you would. Imagion though, Youre only limitation is going to be with the fan. Or if you us it as a radiator .