Self-assembling three-dimensional crystals of DNA.

Scientists working in the burgeoning field of nanotechnology at Brookhaven National Laboratory have announced another breakthrough in molecular technology: They figured out how to use DNA to guide the construction of crystals on the molecular scale. It goes a little something like this: Because the nucleotides that link together to create DNA (adenine, thiamine, guanine, and cytosine) have regions with different patterns of electrical charges, individual molecules are attracted to one another and can stick together, rather like complimentary pieces of velcro. Moreover, each nucleotide has more than one region which can create a bond - this is how other parts of those molecules stick to the sugar/phosphate scaffoldings that give DNA its shape. In other words, think of them like double-ended Lego bricks: One end sticks to the "business end" of another Lego brick, while the other end attaches itself to the framework. However, the properties of these links make it possible to tow whatever other molecules they stick to around in a solution and guide them into specific three-dimensional arrangements.

What the researchers at Brookhaven did was attach small bits of DNA that have exposed complimentary bonding groups to nanoparticles (particles of matter on the scale of billionths of a meter in diameter) of gold and then dropped them into a solution. The bits of DNA attached to each particle were attracted to bits of DNA on other nanoparticles that happened to match up and hooked together, which dragged the nanoparticles into a particular position. Repeat this process again, and again, and again, millions of times. The samples were then heated and allowed to cool, very much like the processing of annealing metals in a foundry, which gave the nanomaterial an opportunity to reshuffle its structure a bit and find more stable arrangements of nanoparticles. The net result: The establishment of ordered crystals of matter on a scale heretofore unvisited. Interestingly, the structure of the crystals isn't very dense at all - of the total volume of space that they take up, only about 5% of it is taken up by the nanoparticles, and another 5% or so is taken up by the bits of DNA used to manipulate them. The rest is empty space, which will no doubt be put to practical use in the near future once the process is refined and made practical (in a large-scale application sense).

In theory, you could start tucking nanoparticles of other materals (or possibly single atoms) into those empty spaces in the nanocrystals. For example, you could use this technique to construct lenses. Or circuitry (come on, you knew I was going to go there.) It would be possible to give such crystals exacting patterns of magnetic or electrical attractions to engineer other properties. It should even be possible to construct parts of mechanisms, like geartrains or simple gear pairs that could be incorporated into more complex nanotech. This is possible because the nanocrystals are sensitive to minute changes in heat which cause their structures to shift and become more manipulable in other ways, and we all know how different patterns of atoms in molecules are attracted to one another and attach as a result.