Smalltalk article for The Roundabout

Your news round-up from the big world of the incredibly small. August 2000 edition
Smalltalk Logo (C)2000 vik@olliver.family.gen.nz

Another giant leap in the world of the small. Six months ago, not far into the fabulous year 2000, the best state-of-the art microscopic grippers available were about half a millimetre long from end to end. Just to show us how fast things are changing in the world of the incredibly small, Swedish researchers have built an entire robotic arm the same size. It has a shoulder, elbow and three-fingered hand capable of picking up bits less than a tenth of a millimetre across. I asked one of the researchers if they could make it any smaller, and he replied that they could, but if they did they would have a heck of a job trying to see what they were doing with it!
Sweedish Robotic Arm

The original research is aimed at producing very small tools for surgeons to use inside the body, to stitch small, delicate arteries together with and so forth.The current tool is certainly small enough to be inserted into the hole left by a mere pinprick, but there is still no suggestion of how the surgeon is going to see what he is doing. In case you're wondering, yes the arm will work in blood. Also salt water, tears, urine and just about any other bodily fluid. One of the dreams of nanotechnology is to have little machines that can run around inside the human body, patching it up so that it can continue to work indefinitely. This device is certainly a step in this direction, but there's a long way to go yet.

Tiny Tweezers

Smaller still is an experimental pair of tweezers made with salmon DNA. These things use 3 strands of DNA to make the tweezers themselves, and two different kinds of "fuel" strands to make them open and close. There's a long strand with two short strands twisted on from each end, leaving a bare bit in the middle - the hinge of the tweezers. It's a bit like a short bit of bent, 2-strand rope cut in half at the middle. They float freely and are so small that they can't be seen, so we don't actually know if they pick anything up at this point! But complicated experiments involving lasers and fluoresence and test tubes full of millions of the critters tell us that they do open and close when the appropriate "fuel" is added. In theory this device could either grip atoms, bring very small things together, or make for a handy little pushme-pullyou motor.

Building With DNA Lego

When people hear that DNA is used in making this kind of device, they go all a-quiver and start worrying about genetic engineering, self-replicating machines going wild and many other scares invented by the media. And believe me, you'll see scares a-plenty as big businesses start to realise that their markets will disappear, made obsolete by nanotechnology. You and I will be bombarded with lots of good reasons why we should buy their stuff and not make our own.

Anyway, the DNA is not used in the same way as us animals and our edible friends, the plants, use it. DNA is a thick, sticky liquid made of long, thin thread-like molecules that like to wrap up in pairs. This is 'cos the threads are made of 4 kinds of building blocks, which only join up in pairs. If it helps, think of one pair as being velcro fluff/hooks and the other pair as being north-south magnets; there is only one thing that velcro fluff will stick to, one thing that a north poles stick to, and so forth. When they stick, they stick well, but try hard enough and you can still separate them.

Us living things use the pairs as a coded string to store the recipe for life on; it is interpreted by complex molecules and weird chemistry in a way that makes us live and grow. Little molecular machines on the other hand, just useDNA as a handy molecule that can be stuck together in the right way automatically. They can't reproduce with it, because they lack the means to make more DNA - a process more complex than it sounds. To them the famous DNA double helix is mere scaffolding, and the order of the building blocks which is so important to us is nothing more than the right set of sticky bits to ensure that tab A goes in slot B.

Bearing Up

Another fun thing to build atomic machinery with is the nanotube. It's a kind of atomic chickenwire made from carbon atoms, except that when you roll it up it forms concentric tubes, a bit like a pushed together telescope. Scientists get all excited about these and are constantly trying to think of new ways to play with them. The latest game involves grabbing the inner cylinder of the bundle and turning it around. Because atoms are nice and smooth and quite unlike chickenwire really, the inner bundle can turn freely inside the outer bundles. It's a kind of atomic roller bearing. But wait, there's more: Start pulling the centre tube out, and the other atoms in the roll try to pull it back in again - it's an atomic spring free!

We can't do much with these little parts right now, but some groups have discovered how to weld these tubes together. Now as far as I'm concerned, welding tube is a darn tricky thing to do. The boffins have to learn a whole new range of welding tricks, and they can't even see what they're doing. However, when they get it right, their bearings and springs are already available.

And Finally...

We all expect the next version of the Playstation to have faster, smoother action, store heaps more levels, play a 54-piece orchestral composition when you hit the X button, and be smart enough to get a degree in astrophysics at Massey University. That means that the chip inside has to be "improved". This is done by making it smaller, so that the electrons inside it's tiny box take less time to get from one part to another, and when they get there have less space to fill up. This makes everything happen faster. So far it's been doubling every 18 months since 1970, which computerniks call "Moores Law". If it holds, some time in 2010 our computer's parts will be the size of single atoms, 1,000 times smaller than they are now. It is likely that this time will see nanotechnology take off, as we will then have the ability to build things one atom at a time. As someone in the early 70's had no idea exactly how we would make today's computers to small and fast, we have only an inkling of which one of a number of possible ways is the "right" one. It may be something completely unexpected, which would speed us up a bit.

So far our intelligence has managed to chug along, producing new discoveries to order, and we know roughly where we're headed. Recently we have discovered how to use atoms to be the parts of a computer - using atoms not to represent a "bit" in a computer, but to make a singe atom perform a real calculation. This is called "Quantum Computing", and may be a way of effectively making computers smaller than atoms. With scientists from IBM recently having made a molecule with 5 atoms lined up to form a working 5-bit quantum computer unit, you have to wonder where it is going to end.

25th August 2000 vik@olliver.family.gen.nz

Back to the Launchpad Main Page