Originally Posted by SynchronicBoost
When Z says one direction, I think you mean the very narrow definition of direction as angular vector right? The definition of a sphere is an infinite number of points equidistant from a point in 3 dimensions. If you drop one graphene molecule, it can be pointed in a vector in any one of 360^2 degrees. Assume though that angle of inclusion is pretty tight for vectors to bleed to the side once the heatsink is on top of the THS.
Is that too narrow? So is Graphene like a thermal diode? Could you use LN2 to realign molecules? Perhaps a magnetic field, etc?
technogiant explained it better than I could have
Originally Posted by technogiant
I guess you almost answered the question there....if graphene is 10X greater than copper then in a random orientation I suppose the worst case scenario would be that the 50% of the particles would have the wrong orientation and so drop it to 5X that of copper.
My other concern is that the pressure squeezing the the heatsink onto the cpu in a thin layer may force the orientation of the graphene particles into a more lateral and unfavorable orientation.
I suppose it all depends on the shape of the graphene particles.....if they are small flat plates as we appear to be assuming then this would be a consideration....but it may be that the graphene is in the form of "Buckyball" graphene spheres which I guess would transfer heat in all directions equally.
I've been enquiring about panasonic pyrolitic graphite sheet and had some information back which may be of relevance to this..... the lateral plane it has a thermal conductivity of 1700 w/mk while in the perpendicular plane it is only 10 w/mk......so a considerable difference.
(on a side note, Graphene can have a thermal conductivity of up to 2300w/mk according to some sources. It really depends on the purity of the test sample.
This is what graphenes molecular design looks like, perhaps it will make it easier to understand, for anybody who doesn't get the "heat moving in a single direction" concept.
caveman59847 already has confirmed it has MUCH better results than normal thermal paste, however I'm curious to see how it stands up to top end TIMs like Indigo Extreme, or other liquid metal TIMsEdited by ZytheEKS - 5/23/13 at 4:31pm