Nanoclast

IBM has created a graphene-based transistor capable of operating at 100 gigahertz, which is more than twice as fast as silicon chips with speeds of 40 GHz using the same gate length. But unlike other high-speed transistors which are made from expensive semiconducting materials like indium phosphide, the graphene transistor will not require the same cooling.

IBM developed the graphene radio frequency transistors for the Defense Advanced Research Project Agency under its Carbon Electronics for RF Applications (CERA) program. The goal of the CERA program and of the IBM researchers is to get the speed of the graphene transistor to 1 THz.

The graphene transistor demonstrated in the research reported in Science had a gate length of 240 nanometers and by applying currently available lithographic techniques it will be possible to reduce that gate length to 35 nanomaters. This reduction in gate length should bring the 1 THz goal within reach.

As reported last week with IBM’s announcement of creating a band gap for graphene, Phaedon Avouris and his team of researchers at the IBM T.J. Watson Research Center in Yorktown Heights, New York, used a metal top-gate architecture employing a high-k dielectric oxide insulated from the graphene layer by a polymer for this high speed RF transistor.

It would seem with these two announcements from IBM within a week of one another that graphene could move more quickly into commercial electronic applications than its carbon cousin, carbon nanotubes.

You have to hand it to Bayer Material Science in their latest announced opening of the world’s largest carbon nanotube plant. Without any real indication that any or even one of carbon nanotubes’ (CNTs) applications are spiking in demand, they just added 200 metric tons to the world’s annual capacity.

Keep in mind Bayer’s pilot plant won’t be number one for long as France-based Nanocyl will be bringing on line a new reactor capable of 400 metric tons annually in July of this year.

In just 2007, Bayer made it known in an article in Chemical & Engineering News that its production capacity for CNTs was 60 metric tons. That means Bayer will be increasing its capacity by 300% in a little over two years to meet what they define as a 25% increase in demand. 

Along with Nanocyl’s production increase this year, it would seem there’s going to be a lot of carbon nanotubes flooding the market without much in the way of demand to use them. Maybe we will hear soon about the “killer app” for CNTs that is going to change that dynamic.

Ever since graphene was first produced in a lab at the University of Manchester in 2004, researchers around the world have been fascinated with its potential in electronics applications.

Graphene possessed all the benefits of carbon nanotubes (CNTs), namely its charged-carrier mobility, but it didn’t have any of the down sides, such as CNTs’ need for different processing techniques than silicon and the intrinsic difficulty of creating interconnects for CNTs.

But all was not easy for applying graphene to electronics applications. One of the fundamental problems for graphene was its lack of a band gap, which left it with a very low on-off ratio measured at about 10 as compared to in the 100s for silicon.

Now this fundamental hurdle has been overcome. Based on research led by Phaedon Avouris at IBM’s IBM T.J. Watson Research Center, Yorktown Heights, New York, IBM is reporting that they have created a significant band gap in graphene.

The research as reported in NANO Letters relates how the researchers were able to create a band gap and a resulting large current switching on/off ratio at room temperature.

IBM researchers achieved this by using a bi-layer (two-layer) designed graphene field-effect-transistor with a polymer insulating the high-k gate dielectric from the graphene transistor channel. The results in experiments have demonstrated that 130 meV can be opened by a strong vertical electric field and an on-off current ratio of 100 is achieved at room temperature and a few thousand at lower temperatures.

Due to the great potential that this research promises for using graphene in applications including digital electronics, tunable terahertz technology, and infrared nanophotonics, IBM has given us the opportunity to do a Q&A with Phaedon Avouris, which we have below.

In addition, in the following days we will also run a Q&A with another representative of IBM research on the overall strategy in pursuing nanoelectronics, specifically with graphene and carbon nanotubes.

Answers by Phaedon Avouris, IBM Fellow and Manager, Nanometer Scale Science & Technology

With the solution that you have come up with for creating a band gap in graphene, what, if any, are the tradeoffs for graphene’s electronic properties?

Using bilayer graphene instead of monolayer would, in principle, lead to a lower intrinsic carrier mobility. However, the mobilities of both materials are so high, much above what is technologically needed, that in reality it makes no difference which one you use. In fact, technologically, bilayer has advantages because electrical noise is lower and there is an intrinsic screening of the influence of trapped charges in the gate dielectric which affect very much the mobility in the monolayer.

A number of other approaches have been identified or implemented for creating band gaps in graphene. What is better about your approach, i.e. bigger band gap, easier to produce?

The lack of band-gap in graphene is based on the fact that the unit cell of graphene has two atoms which are equivalent (have the same potential). To generate a band gap one has to make the two atomic sites have different potential. This is difficult given they are 0.15 nanometers apart. Another approach is to reduce the dimensionality of graphene from 2D to 1D, i.e. make a nanoribbon. This opens a bandgap through the 1D confinement of the electrons. However, to open a usable bandgap for digital electronic applications one needs nanoribbon widths of the order of 2nm. This is beyond the current limits of controlled fabrication techniques and the resulting edge roughness drastically reduces the mobility.

In 2006 E. McCann theoretically predicted that applying a strong vertical electric field to a graphene bilayer would open up a bandgap. This bandgap could be tuned with increasing E-field. This is, in principle, a simpler approach that we have chosen to investigate.

You have indicated that the on-off current ratio with the new graphene-based FET is around 100 at room temperature. How much do you think it will be possible to improve that ratio at room temperature through scaling down the thickness of the insulating layers?

The on/off ratio depends not only on the bandgap, but also on the carrier scattering in the graphene. We are currently working to increase the electric field working with thinner and higher k dielectrics and also to make purer graphene. The experiments would tell us what is technologically feasible. However, having a tunable bandgap in graphene of the size that we have achieved already opens up the possibility of applications in many areas such as graphene optoelectronics, e.g. IR and THz detectors and emitters.

Does your bilayer design and insulated high-k gate dielectric present itself for easy manufacturability?

Yes, it is manufacturable. However, we are still exploring even better dielectric stacks and improved device architectures.

Creating a band gap in graphene has been identified as a critical step in using the material in electronics. With the band gap issue addressed, at least with this most recent device, what major obstacles still stand in the way now for graphene in electronics applications? 

The bandgap issue has been addressed, but not fully resolved yet. We are still exploring how far we can go. In general graphene large scale fabrication has made great progress, but homogeneity issues are still present and are intensively pursued.

This month’s Spectrum Magazine has an article available online written by Edward H. Sargent that discusses his work at the University of Toronto in applying quantum dots to the field of optoelectronics. Specifically Sargent discusses the work his group has done in using quantum dots for photovoltaics, infrared sensors and infrared modulators.

While the application of photovoltaics for quantum dots leads off the article, it is in terms of chronology likely to be the last of these three that will see commercial development.

Sargent rightfully bemoans the current state of photovoltaics in which we either have high efficiency and a high price or low efficiency and a low price.As far as which way to go, cheap or efficient, I like one of the comments on this blog that suggested we follow the McDonald’s model: “Make 'em cheap, make 'em fast, make 'em consistent, and have 'em ready when I'm hungry.” (Thank you, David Alexander).

And that is more or less what Sargent offers us with his painted on quantum dots for making photovoltaics. Now, I don’t want to argue with someone like Professor Sargent on this issue, but when he ventures out into the economics of such devices he is depending on the information provided by others.

They have warned him that organic polymer photovoltaic cells will need to achieve at least a 10-percent efficiency if they are ever to take off commercially. This may be correct. Even this blog has trotted out various per kilowatt numbers as thresholds for commercial acceptance, but I’m beginning to feel a little tug in the back of my mind that it may not be about the numbers.

At any rate, I hope that Sargent is correct and the fundamental science research is at the point where we can start looking at the engineering for some of these optolectronic devices because the persistent refrain of “…in five to 10 years” will soon lead us into the time frame suggested by George Crabtree and Nathan Lewis for crystalline materials meeting the requirements of efficiency and cost for photovoltaics.

The Internet has been buzzing this week with the story of researchers in France developing a “hybrid nanoparticle-organic transistor that can mimic the main functionalities of a synapse.”

You may have seen a number of different versions of this story this week, but I recommend Nanowerk’s coverage of it, which I quoted above. Nanowerk manages to bring some context to the research and how nanotechnologies have been used to try and duplicate some of the functions of the human brain and apply them to computer science, most notably DARPA’s SyNAPSE project. 

In addition, Nanowerk has an interview with the lead researcher, Dominique Vuillaume, a research director at CNRS and head of the Molecular Nanostructures & Devices group at the Institute for Electronics Microelectronics and Nanotechnology (IEMN) that reveals it as a far more significant piece of research that I had originally thought when I read the first news accounts.

The researchers have dubbed their device NOMFET (Nanoparticle Organic Memory Field-Effect Transistor), and as the name indicates it is an organic transistor, based on pentacene and gold nanoparticles.

The details of how the transistor work as described in the Nanowerk article explains the ingenious way that the researchers were able to mimic the “short-term plasticity” of a synapse.

While this is a remarkable development it would not be quite as significant to the development of neuron-inspired computer systems if it were not for its big improvement on silicon-based CMOS chips in duplicating neural networks.

As quoted in the article, Vuillaume says “…even if silicon CMOS chips have been designed and fabricated to emulate the brain behaviors, this approach is limited because it takes several – at least seven – silicon transistors to build an electronic synapse. Here, we did the same job with a single device."

When one considers that there are 10,000 synapses for every neuron in the human brain, reducing size becomes increasingly important if your aim is duplicating the architecture of the human brain.

The quote in the above headline comes from a banner used by protestors recently in Marseilles, France. Apparently the protestors have been disrupting public engagement meetings throughout France over the last couple of months.

Once again the fears that lead to these types of outbursts have little to nothing to do with nanotechnology and instead revolve around distrust of big business and the supposedly complicit governments that support the immoral aims of business.

In coming to the defense of the protestors, the Friends of the Earth, who in the past have boycotted public engagement meetings on the assertion that the “primary purpose ((in this case) of the Australian Office of Nanotechnology) in this area is to promote uncritical public acceptance of nanotechnology,” have offered some new reasons to hate public engagement on the subject of nanotechnology.

This time the Friends of the Earth have claimed that the public dialogues are ignoring nanoparticle toxicity. Well, that’s surely problematic. I haven’t seen the agendas for these meetings but if that issue isn’t addressed, it certainly should be. But the next two issues really get to the heart of their concerns and the banner slogans. You see, nanotechnologies encroach on our private life and nanotechnology is somehow tied up with war.

This is not the first time I have noted this line of thinking, which to some extent has been fostered by organizations that may not have been entirely aware what the fallout might be from their musings.

But hold on a minute, nanotechnology has not compromised our privacy IT and telecommunication technologies have already taken most of it away already.

And war? Really? I can only guess that the Friends of the Earth’s concerns are informed by websites such as this. 

These protests and their justifications from the NGOs really do indicate how necessary these public engagements are because neither of them can distinguish between nanotechnology and science fiction, and as a result they link nanotechnology to their worst fears.

A recent story  on how Saudi Arabia plans to use solar energy to power its water desalination plants is short on details and just plain confusing in some places with sentences such as: “The new nanotechnology for using solar energy to operate desalination plants was developed by KACST in association with IBM.”

I suspect what is meant by the above sentence is that they intend to use photovoltaics somehow enabled by nanotechnology to power their desalination plants. Despite the rather awkward syntax, it is clear that Saudi Arabia is intent on using nanotechnology to both help them meet their energy needs and provide their fresh water.

Two of the most critical shortages mankind faces today are water and energy. Nano-enabled processes have demonstrated some promise in easing the water issue and nanotechnology and energy seems to be the proverbial carrot to relieve our dependence on fossil fuels. However, only a handful of companies have had any success in bringing these solutions to market.

With the Kingdom of Saudi Arabia sitting on top of the largest easily accessible oil reserves in the world, it has been one of the only countries that could actually afford the enormous cost of water desalination, which ranges between $0.5 to $0.85 per cubic meter. So it is with some irony that this blogger notes that the Kingdom appears to be one of the few countries to seriously approach the use of solar energy to address their water shortage.

With the Ogallala Aquifer continuing its headlong course towards depletion maybe water shortages, and even resulting food shortages, will be another impetus in addition to expensive energy costs for applying the technologies out there, including nanotechnology, that sit unused and undeveloped as the status quo is meticulously maintained.

Andrew Maynard is in Davos this week meeting with the World Economic Forum folks who depend on plugging the term “technology” into their cure-alls for the world. Maybe while he’s there trying to get these folks to understand that technological innovation can’t be plucked from a tree he can add that maybe someone should actually try using the technology we already have. It seems as though Saudi Arabia is doing it.

Since I at times play the role of a channel surfing, football watching male, I am curious to know more about anything that proposes a way to keep beer cold. So when I saw the quote, "We've managed to cool six cans of beer"  on Twitter, I had to investigate.

Indeed two researchers from the University of Technology, Sydney, Professor Geoff Smith and Dr Angus Gentle, have reported in this month’s online edition of Nano Letters on a nanocoating that not only manages to keep beers cold but presents an ingenious way of ensuring that heat is emitted on a wavelength that allows it to escape the earth’s atmosphere.

By using a mixture of silicon carbide and silicon dioxide nanoparticles they discovered that they could get heat to emanate at wavelengths of between 7.9 and 13 micrometers allowing the heat to escape the earth’s atmosphere.

Smith is quoted in the story above as believing that the nanocoating could be used as a sort of “reverse solar collector” in which air or water could pass below a plate coated with the nanonparticles and be cooled. The water or air could then be circulated through a building and serve as a type of air conditioning.

The coating appears to be a remarkably simple way to cut on energy costs and maybe even find a way to reduce global warming. But the true genius, as I think we can all agree, was testing it on cans of beer. Well done. 

I imagine like most other governments intent on figuring out how to make the most out of the promise of nanotechnology, the UK government requests a number of reports to research how it and business can position itself to best exploit the so-called “nanotechnology industry”.

A couple of the latest of these UK reports have come under scrutiny by nano-focused blogs. In one, our old friend TNT Log, highlighted in yesterday’s post, takes a critical look at a study entitled: Nanotechnology: a UK Industry View.

In the other, a new blog that focuses on nanotechnology called 10minus9 takes on another report that appears on the UK government’s Science: So What? website that presents what the future jobs might be.

While TNT Log takes apart the thread-worn recommendations trotted out for seemingly the millionth time of the UK Industry Review report, at least that report was based on the underlying science of nanotechnology. The report under 10minus9’s scrutiny doesn’t appear to burden itself with picky details like science.

The jobs-of-the-future report was produced by consultants at Fast Future Research and comes up with one job in the future called a nano-medic that may have excited the noted intellectual Stephen Fry to exclaim that this was the job he would like to have, but unfortunately had a rather strange description that included “creating sub-atomic devices”. 

The 10minus9 blog does a good job of simply explaining how we are not likely to be making sub-atomic devices soon, or ever, and that the consultants apparently based their sloppy work on a rather slim foundation.

But after all the hullabaloo that this report seemed to have generated, including favorable reviews from Prime Minister Gordon Brown and Science Minister Lord Drayson, you have to wonder, as does 10minus9, if anyone bothers actually to read these reports.

As evidenced by the blog links to the right of this post two of my favorite blogs that deal with nanotechnology are TNT Log and 20/20 Science. The two authors behind these blogs share some commonalities. They are both scientists, are both handy with the written word and are both from merry ole England.

It is the latter trait that appears to have led to them to be involved in an odd coincidence: both of them within a week of each other trying to deconstruct the science reporting of your typical British tabloid. For 20/20 Science the object of its attention is the Daily Mail  and for TNT Log it’s the same rag.

I mainly know the British tabloids by reputation, which is that of sensationalistic headlines and stories that don’t always burden themselves with facts. So, we could expect that the two science-trained scribes would make pretty short work of the publication in question.

Well, TNT Log takes the example of a piece written not by a trained journalist but by a representative of the Soil Association, which two years ago decided to not certify any products as “organic” that used nanotechnology additives even while they admitted the ban would impact zero products. Can you say “Grandstanding”? Needless to say, the half-truths and the impulse to extrapolate into fictions are quickly dismissed.

However, 20/20 Science started out to write a critical piece on the Daily Mail’s science reporting but ended up finding some value in it for at least providing some useful information despite the sensational headline.

The Daily Mail comes under pretty regular criticism for its coverage of science-related stories and for good cause.

But I was a little curious as to why they had set their sites on nanotech of late, and I think the answer is in the frequency of the criticism from blogs like Bad Science.

One of the fears of your common tabloid reader is that science is doing some terrible thing to them without them even knowing about it. It could be just about any scientific discipline, it just so happens nanotech is a pretty attractive target at the moment with its colorful doomsday terms like “grey goo’. 

I am sure we will see one day in one of the tabloids all the ways nanotech causes cancer and the next all the ways it cures cancer and not even a nod to the irony.