Question For You Chemists

[Protobyte]

Hello, so I have recently jumped into this wickedly cool hobby you guys do. I've taking it upom myself to try, and put together a 445nm diode that can output 7W. Being overly cautious and overall just curious what would graphene powder heavily suspended in a low yielding resin prove to be useful for a shelling? I am trying to understand thermal conductivity, and I am realizing that these particles need to be touching each other for the best possible outcome. Is this an overkill process which would end up being a waste of money and time, for it would be suspended within resin molecules thus preventing contact needed? That is my vague understanding through wiki research. At any case would love to hear your opinions on why this won't work, or if it would. If it is left as an unknown I may shell out some money just for the sake of science, but would like to know if it is insanely obsurd before doing so.

 

[diachi]

Shelling?

 

[Protobyte]

I apologise, Im talking about the initial housing/casing for a diode to be soldered into, and provide as a heatsink.

 

[diachi]

I apologise, Im talking about the initial housing/casing for a diode to be soldered into, and provide as a heatsink.

The diodes aren't soldered into the housings, they are press fit. But that aside, chances are you'll be better off with copper. Graphene powder suspended in some other material may work well as a thermal interface material (similar to Arctic Silver or IC Diamond), but it'd likely be inferior to a solid metal module in terms of both mechanical and thermal properties.

Feel free to experiment if you can find/make such a material, but I doubt the results will be all that good.

 

[Anthony P]

Graphine has amazing thermal conductivity, but it seems to me a great deal would be lost to the resin. As diachi said, Cu is probably your best bet. If you are not trying for portable, ther are active cooling options: TEC, fan and fins, water cooled.

 

[Immo1282]

I'm pretty sure that with well prepared surfaces, the close press-fit that a diode into a metal module gives is going to do a better job. An interesting suggestion - but not one that seems all that practical unfortunately...

 

[Protobyte]

Thank you all for the responses, and has taught me a lot. Been doing some research on it, and excited to see what advancements can be avaliable on a consumer level with carbon super conductive composites. To add a note to this I ended up purchasing a 99.999% high density graphite electrode just to play around with. Ill keep yoy all updated, and worse comes to worse I can use it for chemistry projects .

 

[theDUDE77]

As you stated, the graphite/graphene would have to be in close contact but the resin would act as an insulator so heat would not dissipate through it and maybe even build up in the resin. I stated graphite/graphene because the only difference is graphene is one layer thick of graphite and there has been naming confusion about graphene powder being sold is actually graphite. However, you are on to the fact that both graphite and graphene have higher thermal conductivity (W/mK) than copper. The higher the thermal conductivity value indicates a better heat transfer. Copper at around 400 W/mK is only slightly worse than silver at 420 W/mK which are two of the best at transferring heat. However, graphite/graphene formulations can be on the order of 2 to 4 times higher (in the 800 W/mK to 1600 W/mK range.)
I found this website that sells graphite sheets https://www.mouser.com/new/panasonic/panasonicthermalgraphite/ and they claim 2 to 5 times better than copper but they are small (4.5" x 3.5" x <1/16" thick) @ $22.31 per sheet. You could line something like a plastic case, resins, etc that do not conduct heat very well with this and have a different type of laser container or you could line copper with it to make it an even better heat transfer medium. Electronic companies have started using these in place of copper or under heatsinks to help better facilitate heat transfer in small electronics. Good luck and hope this helps.

 

[Lifetime17]

Hi, Maybe try a solid graphite round bar, I have one put aside might some day make a sink out of it. should be a real messy job then might have to seal the surface with a light coat of urethane .
Rich

 

[Immo1282]

I'd like to see this - though you'd have to be very careful about the orientation of the bar - as the thermal conductivity of graphite is around 400W/m⋅K (bloody incredible) but it's around 20W/m⋅K in the other direction - which is significantly worse than copper or aluminium...

 

[smallfreak]

No need to mess around with graphite or graphene (if you can get enough of it). Just take a diamond of proper size and all your thermal problems are gone. With 2200W/mK it easily beats even graphene hands down.

And your pointer will get a very precious look with the diode mounted in a clear diamond instead of greasy black graphite.

 

[Anthony P]

Just a thought... Shouldn't specific heat capacity of material be taken into consideration. A material that heats up too quickly will result in very short duty cycle.

 

[theDUDE77]

Just a thought... Shouldn't specific heat capacity of material be taken into consideration. A material that heats up too quickly will result in very short duty cycle.

Yes. So ideally you want a material that has a low specific heat and a high thermal conductivity. This means the material would take a lot of the internal heat from the diode to heat up (low specific heat) and transfer that heat to the surroundings quickly (high thermal conductivity). So when looking at copper and iron, both their specific heats of 0.39 kJ/kg*K vs. 0.45 kJ/kg*K are very close. However, copper has a much higher thermal conductivity than iron (400 W/mK vs. 94 W/mK). Under the same conditions, iron would get hotter than copper so copper would be the better material to use.

 

[kecked]

Don’t bother use copper or aluminum. The amount of heat removed has been shown through actual use to be just fine. Don’t make it more complex than it needs to be. It you want to get fancy addin a little indium foil. Truly don’t bother/

 

[Anthony P]

IMO, if you want something tried and true use Al or Cu. However, like most this is an experimental science. If you have the resources try it! Let us know how it goes... good or bad. That way everyone learns something.

 

[kecked]

Try silver