Science of the Small

What is nanotechnology, and why do people study it?

The definition fed to us from dictionaries and the like is the rather bland and unhelpful: any science where at least one of the dimensions is less than 100 nanometres (nm), where 1 nm is equal to 10 to the minus 9 metres long (0.000000001 metres). For reference, the length of the bond between the two atoms in a hydrogen molecule is about one-tenth of a nanometre. The cells that make up your body are on the order of 1000 nm. So when I talk about the science of the small, hopefully you appreciate exactly how small I mean.

But that description is hardly sexy, and it tells us little. When asked by laymen, what is nanotechnology, I respond simply by saying it involves molecular machines, the design and construction of devices and systems that are built up from just a few molecules apiece. Easier to understand, though I can’t pretend this isn’t a very crude description. Still, it gets the point across.

What do I really think nanotechnology is? Nanotechnology is physics. Nanotechnology is chemistry, it is biology, it is materials science. It is a fusion of scientific disciplines with the intent of constructing and examining features so minute that no optical microscope could ever make them out.

When seen at this scale, matter isn’t intuitive or simple. Surfaces that appear to be smooth show significant roughness, things which seem to be static and unchanging are in a chaotic state of flux. Light as a tool to see what is going on becomes useless. Water acts like glue, ions act like cannonballs and scratches beneath the surface of a shiny piece of metal act with all the grace and subtlety of tectonic plates colliding. And this isn’t even factoring in quantum mechanical effects - currents which should be insulated leak like rusty faucets, matter acts like energy and energy behaves like matter.

It is difficult to come up with an everyday analogy for what the assembly of nanoscale devices is like, but I will try. Sometimes it is like trying to build a chair out of Lego pieces. Now, all the pieces need to be the same colour, but every hundredth, every fiftieth, maybe every tenth piece is not only the wrong colour but the wrong shape. You are blindfolded and have earmuffs on. You are wearing several layers of thick gloves that make you clumsy and unable to feel the pieces. The pieces are incredibly sticky - they are only slightly more likely to stick to other pieces than to the walls or your hands. Despite this none of the pieces sit still, even after locking them into place. You can only check your progress when you are finished, but this runs the real risk of blowing one of the legs clean off. Even if your chair survives examination, you are only given a rough sketch of the chair’s silhouette. Oh, and each Lego piece costs more than your first car and are one-use only, so the many screw ups you make will be very expensive.

It isn’t always this bad, but much of the time it is worse.

Working blind is no exaggeration. Electromagnetic waves, like light and radio signals, are defined by their wavelength. One property these waves have is that if they encounter an obstacle smaller than their wavelength they will pass through it. Submarines exploit this by using radio waves with wavelengths that are kilometres long - their signals are able to pass right through small islands as if they didn’t exist. Similarly, visible light has a range of wavelengths from around 700 nm for red to about 400 nm for blue. This means that anything smaller than 400 nm will not properly reflect light. Considering that the definition of nanotechnology involves scales of less than 100 nm, visible light will tell us nothing about our specimens.

Of course, smaller wavelengths of light could be used. Light with wavelengths equal to or smaller than the features that need to be resolved could, in theory, be used to examine the specimens. But reducing the wavelength of any wave increases its energy - soon we would be beyond visible light, beyond ultraviolet, possibly even beyond high energy X-rays. These are highly energetic, and highly destructive - any sample studied using these waves would be altered by them, if not completely destroyed.

There are alternatives to using light, thankfully. Simplest of these to understand is electron microscopy, where electrons are used in place of light. They are able to form waves with small wavelengths but not too high energy, and so can resolve features that are less than 100 nm across. Different samples interact differently with electrons, but there are different types of electron microscopy - some pass electrons through the sample, some reflect electrons off the surface, some stimulate the atoms of the sample and cause them to release electrons. These techniques can give partial information about a sample at tiny scales.

Electrons aren’t perfect. They carry a charge, and because of this they can be destructive towards the sample. Neutrons are another alternative - though physically larger than electrons, they have no charge. But this can create the opposite problem. Rather than interacting destructively with the sample, they might not interact at all, and no interaction means no information. Only specific materials, like deuterium, have a significant neutron absorbance cross-section.

There are other approaches to seeing without seeing, some quite inventive. Like Atomic Force Microscopy, where a superfine point is moved over the surface of the material and the mutual atomic repulsion is measured. But few of these techniques can be applied outside a vacuum, or on an unfinished product. Significant work has to go into showing that your sample isn’t being altered by the conditions of examination. And even then, the information is partial. Different techniques need to be applied, any irregularities addressed and a more complete picture built up before you can conclude anything about the nature of your sample. In nanotechnology you work blind, because the act of seeing is both impractical and destructive.

But, assuming that you successfully examine your sample, you will immediately observe other problems. Your pristine metal sample carries deep gouges in the surface, your intricate device is lacking a component and several of the longer chemical chains have entwined themselves in an embrace that would almost be romantic if it weren’t so frustrating. Because that is the reality of matter at this scale - short of a multibillion dollar manufacturing plant, with a high tolerance for wastage, matter will not do what you want it to. Even when it should, even when all the laws of chemistry, logic and awesomeness are on your side, matter will do whatever it wants.

Consider the following example: take strands of DNA and attach a sulfur atom to the end of each. Suspend these strands in solution over a polished gold surface. Sulfur bonds strongly to gold, so there is no reason (assuming you control the salinity, pH, temperature, etc) that the DNA wont fix itself to the surface, end first.

But what will happen it reality? DNA strands will collide and entangle with each other, some will simply denature, some will crash into the surface and lay there “side on”, the polished surface will be covered with gunk and structural irregularities. Oh, don’t get me wrong - many, maybe even most, of the strands will do exactly what they should. But enough will just do their own thing that, unless you find some way of removing the failed strands, any device you are trying to build will have incredibly low efficiency, if it works at all.

And that’s not even mentioning impurities. The gold surface will not be pure gold, no matter what you try. The sulfur ends will not be pure sulfur - there will be some strands floating around that tried to bind with iron, or butanol, or who-knows-what. No volume of material you use will be pure, and when your device relies on the precise placement of molecules, you will find this to be a problem. The silicon that goes into microchips needs to be completely pure, with an exact amount of specific additives uniformly mixed through it, but it is only something like 99.9999% pure, give or take a few decimal places. Sounds pretty damn pure, until you realise people have spent decades designing plants which cost hundreds of billions of dollars to build whose sole purpose is to make completely pure silicon. That is our upper limit, the purest silicon we can reliably make, even with virtually limitless funds.

Matter tends to be very active at this scale. In the example above, even once the sulfur attaches itself to the gold surface it will zip around the metal, occasionally detaching and reattaching itself. Mobility can be restricted - in this case the sulfur is bound to the surface, but if a chain were imbedded in the gold substrate with a sulfur atom sticking out, it would be far more static. It would still be thrashing about, jiggling, dancing in the thermal fluctuations like a flower in the breeze, but at least it wont be moving around the surface. Much. Problematic if you are trying to use these as the foundation for further modification.

So that is what nanotechnology is, an approach to construction where you are blind and the building materials have minds of their own. As for the why? Who cares why? Seriously, there are people out there who are building devices or enhancing materials by the deliberate shuffling of components so tiny that photons are impossibly huge. They are making rope out of carbon-carbon chains, miniature railway tracks one atom thick, corporate logos eight atoms tall. Why ask why? Even if these had zero real world application, wouldn’t it be enough that we are able?

Science isn’t about the “why”. You ask a scientist why they designed a protein that reversibly unfolds when exposed to light and they will tell you, “because I could”. The achievement itself is the why, any applications that arise from the work are a pleasant bonus. Of course, if you want them to tell you the why behind their work ask them around the time of grant applications… but don’t expect them to be completely honest. Hyped up explanations are given to reporters and those who control the flow of funding, not to mention when trying to pick up nerdy girls. Applications are for engineers, scientists are simply trying to push the boundaries of human knowledge.


I really enjoyed this article but I think it could be formatted better. That is, I think it needs more illustrations to break up different sections and the background colour clashes with the text colour. The article is hard work (in a good way) but I think more people would be apt to read it if it were more easy on the eye. I dunno where you would source relevant royalty free images but I really think the right images would lift the whole thing. As I say this is more an issue of formatting than with the content of the article, which I found fascinating.