Molecular Manufacturing feedstock- nanotechnology

[qazwsxedc]

From what I understand of nanotechnology, a lot of it is based around assembling individual molecules one by one to make superior materials such as graphene and nanotubes. If this is correct, will most materials be made out of carbon in the future (assuming that they are manufactured with nano-tech), because of its strong bonding? Basically will they just throw a bunch of graphite into a machine and come up with a product, or will it be many different types of molecules (ie the feedstock is composed of many complex molecule chains that differ in each application)?

 

[The Wister]

Carbon will probably take a much more prominent role as costs come down, but for all it's properties, it still fails catastrophically, so metals will be needed where toughness is required vs. cost. Processing costs need to come down too. Gold is one of the best conductors, but we still use copper. Most of it is about price really.

I think the one area carbon falls short is lack of thermal conductivity, so heat sinks and heat exchangers or anywhere you don't want thermal isolation, you cannot use carbon.

 

[MATLABdude]

[...]I think the one area carbon falls short is lack of thermal conductivity, so heat sinks and heat exchangers or anywhere you don't want thermal isolation, you cannot use carbon.

Eh? Depending on the allotrope, they've got an order of magnitude higher thermal conductivity!
http://en.wikipedia.org/wiki/Thermal_conductivity#Experimental_values

To the OP, it's not necessary to assemble graphenes atom-by-atom:
http://technologyreview.com/computing/20558/?a=f

Nanotechnology is great at making really small things, often very quickly (but usually at a hefty price tag, depending on how many of these you're making). Making big things (like, say, something even the size of a drinking cup) is not done using these techniques since the existing machining and materials technologies are so much more economical, time-efficient, and, well, easy to do.

Unless great strides are made to address these (or the whole paradigm of mass production of manufactured goods falls apart), you're not likely to see the majority of everyday items manufactured in such a fashion.

 

[The Wister]

Eh? Depending on the allotrope, they've got an order of magnitude higher thermal conductivity!
http://en.wikipedia.org/wiki/Thermal_conductivity#Experimental_values

How would you propose making a radiator out of graphite? It's physical and strength characteristics do not lend to easily exploiting this property.

 

[MATLABdude]

How would you propose making a radiator out of graphite? It's physical and strength characteristics do not lend to easily exploiting this property.

Graphite is actually one of the softest materials out there, and readily machinable. You can probably 'machine' yourself a rough heatsink from a block of this stuff with a pin-file, if you were so inclined. I simply sought to address your original point about carbon's lack of thermal conductivity, which is inaccurate (though I'm not sure if you perhaps mistyped).

Owing to the low density of graphite (especially compared to copper), it actually *is* being used as a heatsink (or at least, a heat spreader, or heat pipe) in certain applications, and if CPU transistor count and densities keep on increasing the way they have been (along with the power dissipation), you may very well see places like extremeoverclocking.com soon reviewing and singing the praises of (and possibly, giving a convenient link to buy from a 'trusted partner') the first generation of (presumably expensive) graphite heat sinks:
http://powerelectronics.com/mag/power_graphite_heatsinks_copper/

I don't mean to get into a pissing match with you over this matter, but the members of PhysicsForums strive (speaking at least in the hard science and academic areas) have a duty to give accurate information to the best of their abilities.

 

[alxm]

Matlabdude: You are aware that there's a huge difference between graphene and graphite?

One simply cannot compare properties of a single molecule to the bulk properties of a material. And for all intents and purposes, graphene is a single molecule. To boot, the intermolecular bonds between the graphene sheets in graphite are very weak, since they're only held together by London forces. For that reason, graphite is not a particularly good conductor of either heat or electricity (often an order of magnitude smaller than for metals, depending on the graphite).

Not only is it a generally a silly idea to extrapolate properties of an atomic monolayer to a "3d" bulk property, but in the case of graphene it's additionally bad: Its conductive properties are dependent on the sp2-hybridization of the carbon atoms. Additional bonds or distortion from the planar will directly reduce its conductivity. (the extreme example being the fully sp3-hybridized diamond allotrope, which on the contrary is one of the best insulators out there).

In other words, we can state as fact that there's absolutely no way to turn graphene into a bulk material with properties that are at all like that of a single sheet.

 

[The Wister]

Owing to the low density of graphite (especially compared to copper), it actually *is* being used as a heatsink (or at least, a heat spreader, or heat pipe) in certain applications, and if CPU transistor count and densities keep on increasing the way they have been (along with the power dissipation), you may very well see places like extremeoverclocking.com soon reviewing and singing the praises of (and possibly, giving a convenient link to buy from a 'trusted partner') the first generation of (presumably expensive) graphite heat sinks:
http://powerelectronics.com/mag/power_graphite_heatsinks_copper/

I don't mean to get into a pissing match with you over this matter, but the members of PhysicsForums strive (speaking at least in the hard science and academic areas) have a duty to give accurate information to the best of their abilities.

My b, I stand corrected. Graphene would make a good heat spreader and is a good heat conductor. Still, I don't think it will come to be one of carbon's defining qualities, which is ultimately the point I was trying to make.

I'm not trying to get into a pissing match either, but let's be realistic, when designing a product, you can't focus on one property and ignore the rest. I think the radiator example summed it up best, because it just wouldn't hold up. In the majority of applications where thermal conductivity is required, graphite is a poor choice because it does not have the combination of properties--thermal conductivity AND modulus, UTS, stiffness--to construct a robust product. In computers the parts are isolated and not typically subjected to shock loads, but this is just a small area. In most other situations (where more than just a single layer coating is needed) this is not the case.

Providing accurate information is paramount, but it shouldn't be at the expense of ignoring the OP's original question, which was, essentially, 'Will everything in the future be made of carbon.' I was simply trying to state that we still can't replace metal in all applications.

 

[MATLABdude]

Matlabdude: You are aware that there's a huge difference between graphene and graphite?

One simply cannot compare properties of a single molecule to the bulk properties of a material. And for all intents and purposes, graphene is a single molecule. To boot, the intermolecular bonds between the graphene sheets in graphite are very weak, since they're only held together by London forces. For that reason, graphite is not a particularly good conductor of either heat or electricity (often an order of magnitude smaller than for metals, depending on the graphite).

Not only is it a generally a silly idea to extrapolate properties of an atomic monolayer to a "3d" bulk property, but in the case of graphene it's additionally bad: Its conductive properties are dependent on the sp2-hybridization of the carbon atoms. Additional bonds or distortion from the planar will directly reduce its conductivity. (the extreme example being the fully sp3-hybridized diamond allotrope, which on the contrary is one of the best insulators out there).

In other words, we can state as fact that there's absolutely no way to turn graphene into a bulk material with properties that are at all like that of a single sheet.

SA to SA, I do not mean to be testy. While I am not a materials guy per se I am indeed aware of the difference. Graphite, even in a (albeit, specialty) binder is still sufficiently thermally conductive such that it can be reasonably considered for use in heat sinks and similar applications.

EDIT: ...And perhaps that's where I failed, as the table specifically lists graphene as a great thermal conductor, and their particular graphite as a poor conductor (though that was at an earlier point in the conversation).

Three blind mice went up to an elephant. One felt the tusk, and said, "The elephant is smooth and cool!" One felt the leg, and said "It is rough and warm" One had the unfortunate luck to stick their hand in the eye and be trampled. We've all answered different questions, and I most certainly have not helped in this matter.

Perhaps to the OP, what were you asking again? We seem to have lost track of the forest for the trees.

 

[qazwsxedc]

I know its been a while but I just came back to this, and if anyone is still interested the original question is:

Will molecular feedstock(the specific molecules they dump into a machine in order to rearrange them) used for molecular manufacturing be mostly carbon to form nanotubes, graphene, other carbon lattices, ect. or will it be a bunch of different atoms and if so which ones predominately?