Electron Superhighway

[Rudy Toody]

Current issue of Science News has major article on new technology.

 

[silverpig]

It's cool stuff for sure.


<-- does research on graphene

 

[CycloWizard]

Bob Prud'homme's group at Princeton claims to have found a way to process graphite into graphene sheets in an automated fashion. Their method is proprietary, unfortunately, and they started a company to capitalize on it. The basic method he described was similar to breaking up layers of clay for nanocomposite material synthesis, but he declined to elaborate on the details.

 

[Born2bwire]

I saw a presentation on this at URSI back in July. Can't remember who gave it though but I think that it could have great potential.

 

[BrownTown]

Meh, people try to act like carbon nanotubes are all the shit, and I don't know why, whats so good about carbon nanotubes that we should use them for everything except that they have the word "nano" in them which is a big buzzword these days. I will personally consider them useless until someone makes something worthwhile out of the suckers.

 

[CycloWizard]

Originally posted by: BrownTown
Meh, people try to act like carbon nanotubes are all the shit, and I don't know why, whats so good about carbon nanotubes that we should use them for everything except that they have the word "nano" in them which is a big buzzword these days. I will personally consider them useless until someone makes something worthwhile out of the suckers.

There are a lot of reasons why they are so hyped. Many properties, both mechanical and electrical, are dominated by "flaws" in the material, such as inhomogeneities, imperfect or unpredictable bonding patterns, et cetera. Carbon nanotubes have carbon completely bonded to other carbon atoms, which implies a flaw-free structure. I'm not too familiar with the electrical properties, but the mechanical properties are absolutely amazing.

Elastic modulus: 1200 GPa (steel: 208 GPa)
Tensile strength: 150 GPa (steel: 0.4 GPa)
Density: 2.6 g/cm^3 (steel: 7.8 g/cm^3)

These properties occur because to deform the material, you must actually bend or break a carbon-carbon single bond, which is extremely strong. The reason they are not in widespread use for much as yet is because it is extremely difficult to manufacture something with such a complex structure in significant quantities. Until fairly recently (probably 10 years or so), the global production rate of carbon nanotubes was on the order of grams per year (i.e. less than a pound annually worldwide). Newer production techniques have arisen recently and they are much more readily available now, but they are still very expensive. They will soon be appearing in many diverse applications, everything from drug delivery devices to structural components in new USAF planes.

 

[f95toli]

Originally posted by: BrownTown
Meh, people try to act like carbon nanotubes are all the shit, and I don't know why, whats so good about carbon nanotubes that we should use them for everything except that they have the word "nano" in them which is a big buzzword these days. I will personally consider them useless until someone makes something worthwhile out of the suckers.

Nanotubes are already being used commercially to make displays. Nanotubes can also be used to make e.g. very fast transistors (as demonstrated by e.g. IBM years ago).
The main problem is that we have yet to figure out how to actually grow them where we want them; at present most devices are made by first finding the tubes by an AFM and then depositing the electrodes. This method is for obvious reasons not suitable for massproduction.

Anyway, the article is about graphene, not nanotubes. Graphene is even less useful than nanotubes at the moment, but on the other hand it has only been around for 3 years.

 

[BrownTown]

I regard to CycloWizard's numbers, those are some pretty cherrypicked numbers and don't necessarily say anything useful concerning the uses of carbon nanotubes. Comparing a metal like steel to a network covalent material like carbon nanotubes and using the elastic modulus and tensile strength is only giving half the story. Anyone with an understand of material science should be able to tell you than any network covalent material will DESTROY a metal in these categories. However in any real world application you are going to be concerned with a whole host of properties and metals will destroy network covalent materials in many of these. If we are talking for example of the amount of energy to break something then you are talking about the toughness. Carbon nanotubes will likely still beat steel in this category, but not by nearly so much. And that having been said, steal is freaking dirt cheap, nanotubes are not. Its always nice to point out some unobtainable product which has incredible properties, but its not gonna be all that usefull unless it can be much much cheaper. I mean the reason silicon is used is because its the most readily available and easily worked material, not because it has the best properties, so proving that carbon nanotubes are way better at electrical properties than silicon, or way better at mechanical properties than steel doesn't really say anything that isn't already known, the reason these materials dominate is their cost, not their overwhelming properties. I mean I look at all the buildings being built around here, and elsewhere in the world, all poured concrete, concrete has TERRIBLE properties compared to pretty much anything you can come up with, but its dirt cheap so thats what everything is made out of. I'm sure a carbon nanotube building could be 400 stories tall and pwn the new 200 story concrete and steel building, too bad it would cost 1000x as much.

 

[irishScott]

Regardless of how fast it is, optical circuits/chips will own when they come out.

 

[CycloWizard]

Originally posted by: BrownTown
I regard to CycloWizard's numbers, those are some pretty cherrypicked numbers and don't necessarily say anything useful concerning the uses of carbon nanotubes. Comparing a metal like steel to a network covalent material like carbon nanotubes and using the elastic modulus and tensile strength is only giving half the story. Anyone with an understand of material science should be able to tell you than any network covalent material will DESTROY a metal in these categories. However in any real world application you are going to be concerned with a whole host of properties and metals will destroy network covalent materials in many of these. If we are talking for example of the amount of energy to break something then you are talking about the toughness. Carbon nanotubes will likely still beat steel in this category, but not by nearly so much. And that having been said, steal is freaking dirt cheap, nanotubes are not. Its always nice to point out some unobtainable product which has incredible properties, but its not gonna be all that usefull unless it can be much much cheaper. I mean the reason silicon is used is because its the most readily available and easily worked material, not because it has the best properties, so proving that carbon nanotubes are way better at electrical properties than silicon, or way better at mechanical properties than steel doesn't really say anything that isn't already known, the reason these materials dominate is their cost, not their overwhelming properties. I mean I look at all the buildings being built around here, and elsewhere in the world, all poured concrete, concrete has TERRIBLE properties compared to pretty much anything you can come up with, but its dirt cheap so thats what everything is made out of. I'm sure a carbon nanotube building could be 400 stories tall and pwn the new 200 story concrete and steel building, too bad it would cost 1000x as much.

The head of the US Air Force Research Laboratory's materials lab thought it was a pretty big deal about six years ago when he taught me about them in his composites class. You're right that nanotubes will probably not be used to build a regular house any time soon. They are a specialty material. As far as toughness and other material properties go, I already said why nanotubes are so superior to traditional materials: because they are flawless. Material properties are largely dominated by flaws. Much of materials processing is dedicated to removing these flaws or at least minimizing their effects. However, carbon fiber's properties are still controlled by processing because flaws will always appear when things are manufactured in the top-down approach. Scaling laws have been developed showing exactly how this will happen and it is very well understood. However, nanotubes are a bottom-up process, so they can be produced completely flaw-free to (theoretically) any length scale. Practically, nanotubes have been produces up to a few inches in length at this point, but scale-up of nanoscale processes is becoming much better understood all the time. There is no reasonable expectation that the size of these tubes must remain small, and the properties do not depend on the length scale of the final product. Even if you only consider small scale nanotubes as fillers in composite systems, they lend extraordinary properties to them. The cost of nanotubes has also decreased by orders of magnitude in the last couple years with no price floor in sight. In short, I don't know why you are so intent on disbelieving such a promising technology unless you just don't get what the big deal is.

 

[Born2bwire]

Nanotubes can also provide a construction material on the nanoscale. I saw a presentation on a paper last week about using nanotubes in improving the efficiency of solar cells (whoo! Two buzzwords in one paper). The solar cell used gold or platinum atoms to basically replicate a process similar to photosynthesis. Normally, these atoms are deposited onto the surface of a sheet that pulls in the electrons. But electrons may not go down into the sheet but instead travel along the atoms, losing energy and being reabsorbed before reaching the conductor. Nanotubes were used to build columns projecting up from the sheet. The atoms were deposited along the surface of the nanotubes. That way, the electrons would flow from the atoms into the nanotubes and down onto the sheet. In addition, the chirality of the tubes did not matter so that freed up some of the problems that occur during manufacturing. In the end, it helped to greatly increase the efficiency of the solar cell.

 

[silverpig]

Originally posted by: f95toli

Originally posted by: BrownTown
Meh, people try to act like carbon nanotubes are all the shit, and I don't know why, whats so good about carbon nanotubes that we should use them for everything except that they have the word "nano" in them which is a big buzzword these days. I will personally consider them useless until someone makes something worthwhile out of the suckers.

Nanotubes are already being used commercially to make displays. Nanotubes can also be used to make e.g. very fast transistors (as demonstrated by e.g. IBM years ago).
The main problem is that we have yet to figure out how to actually grow them where we want them; at present most devices are made by first finding the tubes by an AFM and then depositing the electrodes. This method is for obvious reasons not suitable for massproduction.

Anyway, the article is about graphene, not nanotubes. Graphene is even less useful than nanotubes at the moment, but on the other hand it has only been around for 3 years.

I visited a group at Stanford a few months ago and I'm pretty sure they deposited pads on a substrate, then grew nanotubes from pad to pad. It definitely wasn't perfect because they had to go to a probe station to test hundreds of sets of pads to see which sets had a nanotube grown between them. But yeah, they're not easy to work with yet.

Graphene basically needs to be grown in order to be useful. I've heard of some proposed research using silicon carbide wafers and then somehow getting the top few layers of the wafer to change somehow and squeeze out a layer of graphene, but nothing useful has been done AFAIK. That would be the best way to go I'd say... that way you'd have your graphene layer on top of a standard wafer already.

 

[miniMUNCH]

Graphene in be 'made' in situ on SiC by basically heating the wafer up to sublime Si leaving a Carbon enriched surface which will then restructure to form graphene.

Very recently scientist have opened, and somewhat figured out out how to control, a bandgap in graphene... paper just came out in Nature Materials.