T H I S W E E K I N N A N O T E C H NanoScience and NanoBusiness News from NanoApex
Dear Subscribers,
This Week in Nanotech covers research and commercialization of MEMS and nanotech from around the world, the emerging marketplace, and its many players. This Week in Nanotech is your complete weekly update on everything going on in the world of tiny tech. Get your business information from NanoInvestorNews, hosting the largest nanocompany database in existence with over 800 entries. ____________________________________ ADVERTISEMENT
Free Nanotech & Technology Related Magazine Subscriptions As a valued member of the NanoApex community, we would like to extend the opportunity to subscribe to leading industry publications. There are absolutely no hidden or trial offers, and no purchase is necessary. Publications are absolutely free to those who qualify. Browse from the extensive list of titles currently offered and be sure to check back often as we will be adding new titles over the coming weeks and months. http://nanoapex.tradepub.com
________________________________________________________________ NanoApex Corp. Announces Two New Versions of Nanocompany Database NanoInvestorNews (http://www.nanoinvestornews.com), the investment portal hosted by NanoApex Corp. (http://www.nanoapex.com) [profile], today announced the release of two downloadable versions of the popular NanoInvestorNews Nanocompany Database (NCD). The database is the largest global publicly accessible database of MEMS and nanotech companies; There are currently 808 distinct listings from all over the world with daily updates. All listings are moderated to ensure that only nanocompanies and not companies abusing the nano- prefix are included.
National Nanotechnology Initiative Workshop on Nano-electronics, -photonics, and -magnetics A National Nanotechnology Initiative Interagency Workshop on Nano-electronics, -photonics, and -magnetics, will be held Feb. 11-13, 2004, at the Holiday Inn Arlington at Ballston, Arlington, VA. Media are invited to attend this workshop where leading scientists and engineers from government, academia and industry will exchange information, research findings and ideas toward identifying needs and opportunities for applications of nanostructured materials and devices. A draft agenda is available. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4255
Purdue researchers create device that detects mass of a single virus particle WEST LAFAYETTE, Ind. Researchers at Purdue University [profile] have developed a miniature device sensitive enough to detect a single virus particle, an advancement that could have many applications, including environmental-health monitoring and homeland security. The device is a tiny "cantilever," a diving board-like beam of silicon that naturally vibrates at a specific frequency. When a virus particle weighing about one-trillionth as much as a grain of rice lands on the cantilever, it vibrates at a different frequency, which was measured by the Purdue researchers. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4254
Optoelectronic technique controls fluid flow in microdevices Improvements in optoelectronics miniaturization underpin a novel technique that uses light to control the flow of nano-size volumes of fluids over solid surfaces. This has set the stage for an advanced line of optically driven microfluidic devices capable of transferring small droplets of fluids in a reprogrammable way. This innovative optical technique uses lasers, or optical systems comparable to those in liquid crystal display (LCD) projectors, to generate complex patterns of differing light concentrations on a flat substrate. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4253
The Dark Secret of Hendrik Schn - transcript NARRATOR (JACK FORTUNE): This is the story of the man behind the most remarkable discovery. His breakthrough seemed so revolutionary it could have created an extraordinary new world. A world where disease could be destroyed before the first symptoms appear. Where nothing would be beyond the boundaries of human knowledge. But others thought it could also be a world where the darkest evil could be unleashed. Where microscopic machines would link up to destroy us all. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4252
Weighed in the nanoscale They're coming big time. Heavyweight reports with nanotechnology in their titles are hitting our bookshelves with increasing frequency. Since the last Green Futures article on this little understood technology of the seriously small [GF34], we've a pile of studies by everyone from the ETC Group and Greenpeace to the Economic and Social Research Council and the Better Regulation Taskforce. The headline Grey goo threat to world' has adorned the front page of a Sunday newspaper, and the Royal Society and Royal Academy of Engineering have set up a working group on the issue, commissioned by the UK government. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4251
Nanotech researchers see the light Researchers in the Mazur group at Harvard [profile] have found a way to make nanofibres only 50nm thick; thinner than the wavelengths of light they carry. Made from silica, the nanofibres transmit the light by acting as a guide for it to flow around, rather than through; and because they can be made smooth and of uniform diameter, the light remains coherent as it travels. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4250
Virtual Nanotech Modeling materials one atom at a time It's hard enough to thread a needle. Imagine trying to manipulate threads and needles miniaturized to one-millionth the normal size. Now, you're thinking like the emerging group of nanotechnologists whose growing dexterity at fashioning new materials and devices may eventually improve every arena of technology, from aerospace to drug development. While many researchers focus on developing tools for working on nanoscale materials, others are pursuing a virtual pathway toward nanotechnology applications. As ever-more powerful computers have become ever more affordable, computational nanoscientists can readily simulate materials atom by atom. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4249
Industries await dawning of nanotechnology age "It will be a ubiquitous technology," said George Stephanopoulos, professor of chemical engineering at the Massachusetts Institute of Technology. He echoes other nanotech supporters who say industrial countries are already sliding toward its use in every aspect of manufacturing. But with such a huge gap between what is and what might be, it remains a difficult realm for investors, who cannot yet be confident that the global market will reach $1 trillion by 2015, as the U.S. government predicts. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4248
Is this the end of the world? Many would have trouble spelling nanotechnology, let alone defining it. But, as Richard Jones and Stephen Wood write, it is here and it is going to be driving the economy well into the 21st century. Nanotechnology is currently thought by many to be the innovation that will drive the economy and the stock market for the next 50 years, changing all aspects of life for the better. But opponents foresee dire consequences environmental degradation, a widening of the gulf between the rich and the poor, even the eventual extinction of the human race. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4246
Nanotechnology Information Devices Workshop A workshop looking at Nanotechnology Information Devices (NID) will take place at the National Centre for Scientific Research "Demokritos"(1) in Athens (Greece) from 04 to 06 February 2004. During the 13th NID workshop, a joint "Greek/PHANTOMS" Symposium on Nanotechnology will be held in order to provide a research overview currently performed on this country. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4245
NSF to invest heavily in nanotechnology The National Science Foundation is seeking $305 million to fund research into nanotechnology, which is the research and development of technologies at the atomic, molecular and macromolecular level. NSF Director Rita Colwell said yesterday the 2005 budget request represents a 20 percent increase over fiscal 2004 levels and is the foundation's "largest priority area investment." http://news.nanoapex.com/modules.php?name=News&file=article&sid=4244
U. New Mexico plans for system upgrade University of New Mexico [profile] and the entire state of New Mexico will soon be on the cutting edge of computer technology if one group of University staff members has its way. Bill Adkins, director of Computer and Information Resources and Technology, said the University is looking to take advantage of the collapse of the telecommunications industry two years ago by purchasing a stockpile of reduced-price fiber optics cable. UNM would use that fiber optic cable to link the state to one network to collaborate on research, Adkins said. http://news.nanoapex.com/modules.php?name=News&file=article&sid=4243
Nanotechnology center plans in the works at U. Massachusetts Charlena Seymour, provost and senior vice chancellor for academic affairs, announced that the University of Massachusetts Amherst [profile] plans to establish a nanotechnology research and development center to be called MassNanoTech.
...
On 9 Feb 2004 17:09:40 GMT, a...@nanoapex.com (Aryavarta Kumar) wrote:
>The Dark Secret of Hendrik Schn - transcript >NARRATOR (JACK FORTUNE): This is the story of the man behind the most >remarkable discovery. His breakthrough seemed so revolutionary it >could have created an extraordinary new world. A world where disease >could be destroyed before the first symptoms appear. Where nothing >would be beyond the boundaries of human knowledge. But others thought >it could also be a world where the darkest evil could be unleashed. >Where microscopic machines would link up to destroy us all. >http://news.nanoapex.com/modules.php?name=News&file=article&sid=4252
The referenced transcript open a very interesting issue:
====================
Prof MICHIO KAKU: We could be facing economic stagnation because computers are simply not capable of evolving to the next step if they are based on silicone. As power levels off the wealth of nations, the productivity of workers, the prosperity of societies could be endangered because of the stagnation of computer power.
NARRATOR: A whole engine of our economic growth could stall. No more growth means no more profits. We could be plunged in to a depression. This is the fear implied by Moores Law. Today billions are spent trying to squeeze more out of silicone, but the worry is that we will eventually get to the stage where we can push it no further. Some think it is a problem in desperate need of a solution.
==================
Fear? Desperate? Depression? Strong words.
Of course, silicon (not "silicone") LSI will hit the wall fairly soon. Current bleeding-edge processors, memory, and ASICs are fabbed with 90 nm minimum features and clock around 4 GHz. If we assume, conservatively, that silicon technology will bottom out at, say, 30 nm and 15 GHz, in maybe 2015 or so, will civilization in fact collapse?
CMOS chips already have a billion transistors, and designs (except memory) are becoming limited by their sheer logic complexity and design/verification/mask costs, not by the billion transistor limit. At 30 nm, we'll have maybe 10 billion transistors on a chip. Will the world economy be trashed because we can't have, say, 100 billion?
This is being typed on a 700 MHz Dell PC; Agent would work about as well on a 486DX. Except for severe simulation applications, I don't really see a need for nanometer-scale logic, nor do I see why the flattening of the Moore curve is an economic threat. Today's silicon mostly serves to churn cell-phone minutes and to transport and display porn.
Do we really need all that much computing horsepower? Ideas welcome.
> On 9 Feb 2004 17:09:40 GMT, a...@nanoapex.com (Aryavarta Kumar) wrote:
> >The Dark Secret of Hendrik Schn - transcript > >NARRATOR (JACK FORTUNE): This is the story of the man behind the most > >remarkable discovery. His breakthrough seemed so revolutionary it > >could have created an extraordinary new world. A world where disease > >could be destroyed before the first symptoms appear. Where nothing > >would be beyond the boundaries of human knowledge. But others thought > >it could also be a world where the darkest evil could be unleashed. > >Where microscopic machines would link up to destroy us all. > >http://news.nanoapex.com/modules.php?name=News&file=article&sid=4252
> The referenced transcript open a very interesting issue:
> ====================
> Prof MICHIO KAKU: We could be facing economic stagnation because > computers are simply not capable of evolving to the next step if they > are based on silicone. As power levels off the wealth of nations, the > productivity of workers, the prosperity of societies could be > endangered because of the stagnation of computer power.
> NARRATOR: A whole engine of our economic growth could stall. No more > growth means no more profits. We could be plunged in to a depression. > This is the fear implied by Moores Law. Today billions are spent > trying to squeeze more out of silicone, but the worry is that we will > eventually get to the stage where we can push it no further. Some > think it is a problem in desperate need of a solution.
> ==================
> Fear? Desperate? Depression? Strong words.
> Of course, silicon (not "silicone") LSI will hit the wall fairly soon. > Current bleeding-edge processors, memory, and ASICs are fabbed with 90 > nm minimum features and clock around 4 GHz. If we assume, > conservatively, that silicon technology will bottom out at, say, 30 nm > and 15 GHz, in maybe 2015 or so, will civilization in fact collapse?
> CMOS chips already have a billion transistors, and designs (except > memory) are becoming limited by their sheer logic complexity and > design/verification/mask costs, not by the billion transistor limit. > At 30 nm, we'll have maybe 10 billion transistors on a chip. Will the > world economy be trashed because we can't have, say, 100 billion?
> This is being typed on a 700 MHz Dell PC; Agent would work about as > well on a 486DX. Except for severe simulation applications, I don't > really see a need for nanometer-scale logic, nor do I see why the > flattening of the Moore curve is an economic threat. Today's silicon > mostly serves to churn cell-phone minutes and to transport and display > porn.
> Do we really need all that much computing horsepower? Ideas welcome.
> John
John, I would agree with you if that was all that there was. Alot of the processing power will probably go into "behind the scene's" processing such as voice/visual recognition and AI functions for the OS. These will probably take up alot of the processing capabilities of advanced (>1 billion transistor)processors. I don't agree that it will be the end of the world if we don't have 100 Billion transistor chips, but it would probably slow down advances in capabilities for the PC.
> John, > I would agree with you if that was all that there was. Alot of the > processing power will probably go into "behind the scene's" processing > such as voice/visual recognition and AI functions for the OS. These > will probably take up alot of the processing capabilities of advanced > (>1 billion transistor)processors. I don't agree that it will be the > end of the world if we don't have 100 Billion transistor chips, but it > would probably slow down advances in capabilities for the PC.
> Greg
With in 10 years we will be using diamond in computers anyways, not silicon. There are already at less one company that growing perfect diamond wafers right now. They have wafers that are 76mm in diameter right now, will have 150mm in a couple of years, and will have 300mm in a couple of years after that. Diamond semiconductors can run over 5 times hotter then silicon ones can, and are better in every other way too. So we should have CPU that will go over 60ghz by then. Also who know what ether types of improvement will be thought of by then too. It would not surprise me if PC do top 100ghz with in 10 to 15 years from now. Also Intel and AMD are both talking about multi-core CPU coming on lines some were at the end of the 90nm run or the beginning of the next shrink 6.5mm, I think.
>"10of100" <Greg_Bal...@hotmail.com> wrote in message >news:c0dnr906jd@enews1.newsguy.com... >> John, >> I would agree with you if that was all that there was. Alot of the >> processing power will probably go into "behind the scene's" processing >> such as voice/visual recognition and AI functions for the OS. These >> will probably take up alot of the processing capabilities of advanced >> (>1 billion transistor)processors. I don't agree that it will be the >> end of the world if we don't have 100 Billion transistor chips, but it >> would probably slow down advances in capabilities for the PC.
>> Greg
>With in 10 years we will be using diamond in computers anyways, not silicon. >There are already at less one company that growing perfect diamond wafers >right now. They have wafers that are 76mm in diameter right now, will have >150mm in a couple of years, and will have 300mm in a couple of years after >that. Diamond semiconductors can run over 5 times hotter then silicon ones >can, and are better in every other way too. So we should have CPU that will >go over 60ghz by then. Also who know what ether types of improvement will be >thought of by then too. It would not surprise me if PC do top 100ghz with in >10 to 15 years from now. Also Intel and AMD are both talking about >multi-core CPU coming on lines some were at the end of the 90nm run or the >beginning of the next shrink 6.5mm, I think.
Who is growing the diamond chips? Have they managed to dope them to get transistor action? Diamond is nice... low dielectric constant and something like 20x the thermal conductivity of silicon.
Yes, multiple CPU cores are a logical progression, especially when immense amounts of cache are available. That would be good for simulation (of nanotechnology!) and could eliminate multithreaded operating systems, the majority of which are awful. No context switching!
Nanotech and/or organics seem to me to have more interesting (and more probable) potential for dense, nonvolatile data storage than for superfast logic.
Robert V Hill wrote: > With in 10 years we will be using diamond in computers anyways, not > silicon. There are already at less one company that growing perfect > diamond wafers right now.
But that won't push out the limits very far. Tolerating higher temperature means we can increase the clockspeed, assuming the power-drain is acceptable, but assuming diamond behaves similar to silicon, where the heat produced typically scales with something like the cube of the frequency, then tolerating 5 times the temperature don't amount to *that* much more clockability.
> So we should have CPU that will go over 60ghz by then.
Very optmisitc. Not only because it's a strange idea that tolerating 5 times the heat should let you clock 10 times higher, but also for other fundamental limits. Such as for example the speed of ligth. In vacuum ligth goes like 300.000 km/s, which means that if you run your processor at 1Ghz, parts that are up to 30cm apart from eachothers could theoretically communicate. Multiply that by 60, and you see that ligth in vacuum would only manage 2cm.
Still sounds doable, until you start to consider that actually, we're not talking ligth in vacuum, but electrons in half-conductors. And actually, we're not talking straigth paths, but winded conductor-paths.
There's some things that *would* radically help for this problem though, such as a switch to a 3-d layout rather than basically paths on a surface as today. Dealing with the heat would be a problem though.
Mechanical nanotech-computers would also likely be limited by heat. You can pack an enormous amount of transistor-equivalent mechanics into a cube cm of volume, but if you operate them at a high enough frequency, and each operation vastes just a little bit of energy in the form of heat, you're going to have problems preventing that cubic cm from melting. (in practice it'd offcourse stop functioning before melting)
>On 9 Feb 2004 17:09:40 GMT, a...@nanoapex.com (Aryavarta Kumar) wrote:
>>The Dark Secret of Hendrik Schn - transcript >>NARRATOR (JACK FORTUNE): This is the story of the man behind the most >>remarkable discovery. His breakthrough seemed so revolutionary it >>could have created an extraordinary new world. A world where disease >>could be destroyed before the first symptoms appear. Where nothing >>would be beyond the boundaries of human knowledge. But others thought >>it could also be a world where the darkest evil could be unleashed. >>Where microscopic machines would link up to destroy us all. >>http://news.nanoapex.com/modules.php?name=News&file=article&sid=4252
Unfortunately I found this to be one of the worst Horizon programs ever. The link between Hendrik Schon's falsification of data and Nanotech was pretty forced. The whole program seemed to be sensationalising the downsides of Nanotech with dire warnings about Grey Goo.
To quote from the final lines of the transcript
" But there is some good news. One route to the world of grey goo has faded. That future is as far off as it has ever been. It means we can now rest a little more securely. "
Definitely the sort of TV coverage Nanotech can do without.
>The referenced transcript open a very interesting issue:
>====================
>Prof MICHIO KAKU: We could be facing economic stagnation because >computers are simply not capable of evolving to the next step if they >are based on silicone. As power levels off the wealth of nations, the >productivity of workers, the prosperity of societies could be >endangered because of the stagnation of computer power.
>NARRATOR: A whole engine of our economic growth could stall. No more >growth means no more profits. We could be plunged in to a depression. >This is the fear implied by Moores Law. Today billions are spent >trying to squeeze more out of silicone, but the worry is that we will >eventually get to the stage where we can push it no further. Some >think it is a problem in desperate need of a solution.
>==================
>Fear? Desperate? Depression? Strong words.
>Of course, silicon (not "silicone") LSI will hit the wall fairly soon. >Current bleeding-edge processors, memory, and ASICs are fabbed with 90 >nm minimum features and clock around 4 GHz. If we assume, >conservatively, that silicon technology will bottom out at, say, 30 nm >and 15 GHz, in maybe 2015 or so, will civilization in fact collapse?
>CMOS chips already have a billion transistors, and designs (except >memory) are becoming limited by their sheer logic complexity and >design/verification/mask costs, not by the billion transistor limit. >At 30 nm, we'll have maybe 10 billion transistors on a chip. Will the >world economy be trashed because we can't have, say, 100 billion?
>This is being typed on a 700 MHz Dell PC; Agent would work about as >well on a 486DX. Except for severe simulation applications, I don't >really see a need for nanometer-scale logic, nor do I see why the >flattening of the Moore curve is an economic threat. Today's silicon >mostly serves to churn cell-phone minutes and to transport and display >porn.
>Do we really need all that much computing horsepower? Ideas welcome.
>> With in 10 years we will be using diamond in computers anyways, not >> silicon. There are already at less one company that growing perfect >> diamond wafers right now.
>But that won't push out the limits very far. Tolerating higher >temperature means we can increase the clockspeed, assuming the >power-drain is acceptable, but assuming diamond behaves similar to >silicon, where the heat produced typically scales with something like >the cube of the frequency, then tolerating 5 times the temperature >don't amount to *that* much more clockability.
>> So we should have CPU that will go over 60ghz by then.
>Very optmisitc. Not only because it's a strange idea that tolerating 5 >times the heat should let you clock 10 times higher, but also for other >fundamental limits. Such as for example the speed of ligth. In vacuum >ligth goes like 300.000 km/s, which means that if you run your >processor at 1Ghz, parts that are up to 30cm apart from eachothers >could theoretically communicate. Multiply that by 60, and you see that >ligth in vacuum would only manage 2cm.
It's even worse than that. The interconnects on most ICs are terrible transmission lines, more like distributed R-C networks than nice TEM wires. Effective velocities are far below the speed of light. Most VLSI chips speeds are dominated now by interconnect delays, not basic transistor switching speed. Increased density doesn't help all that much: the interconnects get shorter, but they get thinner and skinnier too. If the chip complexity goes up and the chips stay the same size, the lengths don't even go down.
>Still sounds doable, until you start to consider that actually, we're >not talking ligth in vacuum, but electrons in half-conductors. And >actually, we're not talking straigth paths, but winded conductor-paths.
>There's some things that *would* radically help for this problem though, >such as a switch to a 3-d layout rather than basically paths on a >surface as today. Dealing with the heat would be a problem though.
High-end chips already have 8 or 10 copper interconnect layers.
>Mechanical nanotech-computers would also likely be limited by heat. You >can pack an enormous amount of transistor-equivalent mechanics into a >cube cm of volume, but if you operate them at a high enough frequency, >and each operation vastes just a little bit of energy in the form of >heat, you're going to have problems preventing that cubic cm from >melting. (in practice it'd offcourse stop functioning before melting)
Any non-silicon nanotech logic is going to have the same physical problems as silicon: speed of light, resistive interconnect losses, power dissipation.
> > With in 10 years we will be using diamond in computers anyways, not > > silicon. There are already at less one company that growing perfect > > diamond wafers right now.
> But that won't push out the limits very far. Tolerating higher > temperature means we can increase the clockspeed, assuming the > power-drain is acceptable, but assuming diamond behaves similar to > silicon, where the heat produced typically scales with something like > the cube of the frequency, then tolerating 5 times the temperature > don't amount to *that* much more clockability.
> > So we should have CPU that will go over 60ghz by then.
> Very optmisitc. Not only because it's a strange idea that tolerating 5 > times the heat should let you clock 10 times higher, but also for other > fundamental limits. Such as for example the speed of ligth. In vacuum > ligth goes like 300.000 km/s, which means that if you run your > processor at 1Ghz, parts that are up to 30cm apart from eachothers > could theoretically communicate. Multiply that by 60, and you see that > ligth in vacuum would only manage 2cm.
> Sincerely, > Eivind Kjrstad
Well I did over look light speed limit, my bad. still using diamond and maybe carbon nanotubes. We should get pretty high ghz speeds. We should be able to put many more layers on a chip using diamond and carbon nanotubes then we can using say silicon and copper, but only time will tell.
>From my limited study of Nano technology, we need much faster computers in
order to work with it. I agree with you that heat is always the problem with any computer. I just feel that the next step in computers will be carbon, diamond chips with nanotubes, as they can take much more heat then other materials. Also most heat from a computer today come from leakage. If I understand what I have read correctly, both diamond and nanotubes should have much less leakage then other materials. I feel in order for Nano-technology to be possible we will need the fastest computers possible. Computer speeds will be more of a problem then Smallies' sticky fingers will ever be.
> On 12 Feb 2004 00:30:37 GMT, "Robert V Hill" <t.blackm...@comcast.net> > wrote:
> Who is growing the diamond chips? Have they managed to dope them to > get transistor action? Diamond is nice... low dielectric constant and > something like 20x the thermal conductivity of silicon.
[ Moderator's note: The text from "peer through the glass," to "about $5" is material quoted from the Wired article. -JimL ]
peer through the glass. Four diamonds are growing beneath a shimmering green cloud. "It took me a long time to get to this point," says one of the men standing beside the machine. This is Robert Linares, Bryant's father. In the 1980s, he was a well-known researcher in advanced semiconductor materials. His company, Spectrum Technology, pioneered the commercialization of gallium arsenide wafers, the microchip substrate that succeeded silicon and allowed cell phones to become smaller and handle more bandwidth. Linares sold the company to PacifiCorp, a diversified utility, in 1985 and disappeared from the semiconducting world. It turns out he took the money and built a secret diamond research lab. "I knew diamonds were going to be the ultimate semiconductor at some point, but everybody thought it was impossible at the time," Linares says. "I had the freedom to do what I wanted after I sold my company, so I spent almost 15 years researching on my own."
To grow single-crystal diamond using chemical vapor deposition, you must first divine the exact combination of temperature, gas composition, and pressure - a "sweet spot" that results in the formation of a single crystal. Otherwise, innumerable small diamond crystals will rain down. Hitting on the single-crystal sweet spot is like locating a single grain of sand on the beach. There's only one combination among millions. In 1996, Linares found it. This June, he finally received a US patent for the process, which already is producing flawless stones.
By January, Apollo plans to start selling them on the jewelry market. But that's just the first step. Robert and Bryant Linares expect to use revenue from the gem trade to fund their company's semiconductor ambitions. Not surprisingly, the diamond industry is hostile to the idea, as the younger Linares discovered four years ago when he attended an industry conference in Prague. He was hoping to find out whether any other researchers - possibly De Beers scientists themselves - had discovered the sweet spot. During a break in the conference, a man approached Linares and told him to be careful. "He said that my father's research was a good way to get a bullet in the head," Linares recalls.
The diamond industry is in fact even more concerned about gems made using chemical vapor deposition than it is about Gemesis stones, though Gemesis poses a more immediate threat. The promise of CVD is that it produces extremely pure crystal. Gemesis diamonds grow in a metal solvent, and tiny particles of those metals get caught in the diamond lattice as it grows. CVD diamond precipitates as nearly 100 percent pure diamond and therefore may not be discernible from naturals, no matter how advanced the detection equipment.
But the greatest potential for CVD diamond lies in computing. If diamond is ever to be a practical material for semiconducting, it will need to be affordably grown in large wafers. (The silicon wafers Intel uses, for example, are 1 foot in diameter.) CVD growth is limited only by the size of the seed placed in the Apollo machine. Starting with a square, waferlike fragment, the Linares process will grow the diamond into a prismatic shape, with the top slightly wider than the base. For the past seven years - since Robert Linares first discovered the sweet spot - Apollo has been growing increasingly larger seeds by chopping off the top layer of growth and using that as the starting point for the next batch. At the moment, the company is producing 10-millimeter wafers but predicts it will reach an inch square by year's end and 4 inches in five years. The price per carat: about $5.
it seem I was wrong about the size of the wafers, but they are growing them.
> Yes, multiple CPU cores are a logical progression, especially when
> immense amounts of cache are available. That would be good for > simulation (of nanotechnology!) and could eliminate multithreaded > operating systems, the majority of which are awful. No context > switching!
> Nanotech and/or organics seem to me to have more interesting (and more > probable) potential for dense, nonvolatile data storage than for > superfast logic.
Eivind Kjorstad <e...@vestdata.no> writes: > Robert V Hill wrote:
>> With in 10 years we will be using diamond in computers anyways, not >> silicon. There are already at less one company that growing perfect >> diamond wafers right now.
> But that won't push out the limits very far. Tolerating higher > temperature means we can increase the clockspeed, assuming the > power-drain is acceptable, but assuming diamond behaves similar to > silicon, where the heat produced typically scales with something like > the cube of the frequency, then tolerating 5 times the temperature > don't amount to *that* much more clockability.
IMO, as heat dissipation becomes a more and more serious problem, it will eventually force a phase-transition to different approaches to hardware architecture. "Hot clocking" has the potential to greatly reduce heat dissipation via combining the functions of the clock bus and power bus, and only switching a transitor when the voltage across it is "low." And it has been shown that the only computing operations that _HAVE_ to dissipate energy are operations where bits are irreversibly created and destroyed, so that in principle, it is possible to build a "reversible" universal computer with _ZERO_ energy dissipation, except for I/O operations. (Albeit, _complete_ reversibility can come at a high price: Since it is not possible to create or erase bits without generating heat, one has to save any auxiliary data required to uniquely reconstruct every step or decision in the program, extra bits for any gate whose fan-out differs from its fan-in (including simple junctions!), etc.; as a result, many common algorithms will consume EXPONENTIAL amounts of memory unless _some_ heat dissipation is tolerated, e.g., each time one has to clear the "bit dump"... :-( For more information, Google on "reversible computing."
Finally, note that Quantum Computers are NECESSARILY and INTRINSICALLY also "reversible computers," and will therefore have zero power dissipation (modulo I/O and clearing the "bit dump").
> Mechanical nanotech-computers would also likely be limited by heat. > You can pack an enormous amount of transistor-equivalent mechanics > into a cube cm of volume, but if you operate them at a high enough > frequency, and each operation vastes just a little bit of energy in the > form of heat, you're going to have problems preventing that cubic cm from > melting. (in practice it'd offcourse stop functioning before melting)
I have never found Drexler's "rod logic" particularly compelling; I personally expect nanomachinery to use molecular or quantum electronics, and have effectors based on reversible molecular conformation changes...
> I have never found Drexler's "rod logic" particularly compelling; > I personally expect nanomachinery to use molecular or quantum electronics, > and have effectors based on reversible molecular conformation changes...
Same here they may have many applications, but I do not at this time think they will fully replace circuity. In the lab they are building circuits out of molecule and even atoms. Electron spin look promising for a way to make quantum computers.
