Your news round-up from the big world of the incredibly small. April 2001 edition
Smalltalk Logo (C)2000 vik@olliver.family.gen.nz
By now you may have got the impression from SmallTalk that the nanoworld is nuts about diamonds. Well, this is pretty much true although you'd need a billion or so nanomachines to make enough diamond to fit even a very small engagement ring for someone you weren't too sure about. Timothy Fisher is taking a somewhat unorthodox approach to using diamond though, and has devised a cunning way of using minute quantities of diamond to get solar power.
 

ITimothy Fisher's diamond solar cellt works off heat rather than light, in the same kind of way that your TV screen does. At the back of your average TV are some electron guns, which fire electrons into the bit you can see, which glows obligingly in the shape of your least favourite TV presenter or occasionally something worthy like a footy game. These tasteless electrons are spat out from a very hot wire, which you can sometimes see glowing if you peek through the vents in the back of the TV set.
 

So with the aid of a US$348,000 grant from the US National Reconnaissance Office, Timothy and his colleagues developed a way of turning natural gas into a diamond film with lots of little points like pyramids, as you can see on his picture to the right where they pack in at 10 million to the square centimetre. When it's warmed up to 1,000oC - a temperature which needs reflecting mirrors to concentrate sunlight but which diamond can comfortably take - electrons flow off the pyramid's points. Strapping a heat-absorbing metal plate on the back and a conducting screen on the front gives the electrons a way out into the world, providing us with electricity. The weird physics of the incredibly small makes these things capable of working at 50% efficiency, compared with the 15% for a standard solar cell.
 

They're not quite finished with the design yet, but once these things are being made for a few dollars per square centimetre with each centimetre giving 10W of power, they should be a durable and practical way of using concentrated solar energy on and off this planet.

R. Freitas' Microbivore by VikBaby Bugcruncher

An aquaintance of mine called Robert Freitas, well known in nanotechnology circles as the author of Nanomedicine, has been writing about the ultimate antibiotic. Called a 'microbivore' his device follows the usual concept of his creations - do what the human body does, but do it better. In this case, it tracks down bacteria and other nasties that are invading its human host, and minces them to make an easily digested soup for the body to absorb at its leisure.
 

The outside of the microbivore is covered with patches that can be programmed to behave in the same way as our antibodies, so that they stick to any unwanted bacteria, viruses, parasites like malaria, or even fungi. Once one of these is detected, several of 277 arms each less than 1/30,000mm long telescope out from the body, grab the bug and wave it towards the "mouth". The artist's impression on the left shows all arms extended which they normally would not be, making it look a lot like a puffer fish. Apologies from the artist.
Once a bug is in the mouth, an iris closes and it is moved into a "morcellation chamber" where it is cut up with nanoscale knives, squirted into a digestion chamber and turned into aforementioned nutritious broth by artificial enzymes so you don't get a nasty reaction to any bits or toxins in the bug. This is pretty much what the human white blood cell does, except it's not equipped with little arms and knives. There are little creatures that do sweep food along their bodies with tiny hairs in a similar way, like "sea gooseberies" and other small strange sea creatures that make a living out of eating plankton.
But this little microbivore - when we can build it - can be programmed to destroy just about any nasty bug, has no side-effects, and does the job more efficiently than an antibiotic or our own body's defences.

Behold The Mighty Microbat

So it doesn't look a lot like an insect. But it does fly by flapping its wings, and it is 20cm long which isn't too far off the size of some dragonflies. Not too many battery-powered dragonflies about though.

But microbat, shown here on the left in an image from its parent company AeroVironment, is the best attempt so far that we've got at a robotic insect. AeroVironment have produced small planes before, mostly for battlefield and specialist surveillance. Microbat is also a step on the way to a new surveillance device, but one that can fly around quietly and unobserved inside. Obviously, it's going to need some work before they get that far - little details like telling it how to avoid flying into a ceiling fan are required, as this is not a good look for a lightweight model that looks as if its major structural component is gladwrap.

The advantage of ornithopters - flappy flying things - is that they can fly slowly, and change direction fairly easily once you've got independent control of each wing. Insects even manage to hover, and so AeroVironment looked to insects for their solutions. Unfortunately, insects wave their wings in an awkward figure-of-eight patter. Well, it's awkward for engineers anyway, and then there's the tricky design problem of that flexible wing. It seems to work, but it's limited by battery life. There is no camera or transmitter on this model, partly due to weight, partly due to battery life, but when they get it right it'll bring a whole new meaning to the phrase "I'm being bugged." Who wants to develop a bug repellent?

Wires Crossed But Working

In the last SmallTalk we saw how researchers at Penn State University were creating arrays of nanowires by using ANZAC cookie-like moulds. Well now Hewlett-Packard Laboratories have patented a design for memory that uses such "crossbar" nanowire arrays. Their design works like this. You take a molecule that works as a memory cell and train it to stand up straight next to its neighbours. Then you make a lot of them and put them together. They form thin sheets one molecule thick, with the molecules pointing up. To access the memory molecules, you take a "crossbar" array of nanowires with the wires going from "North" to "South" as it were. You then take another array and line them up "East/West". If you sandwich a sheet of your smart memory molecules between these two arrays, you'll have memory cells between wires where the "North/South" lines cross the "East/West" ones. Measure across the right wire, and you read the right memory cell.

Now with the "crossbar" arrays being fairly easy to make and the memory building itself, you have a cheap way of making plentiful memory. HP's research group (led by Philip Kuekes, R. Stanley Williams and University of California, Los Angeles chemist James Heath) having got a patent on the process, they reckon they'll have a working device within a year. When the HP project started 3 years ago, they projected working circuits in 2005. Nanotechnology is moving faster than you think. Smalltalk Logo (C)2000 vik@olliver.family.gen.nz

http://olliver.family.gen.nz/launchpad 25th April 2001 vik@olliver.family.gen.nz

Back to the Launchpad Main Page