Oh, I know this one! Your enemy inhales it and it flays them from the inside.
Is it any wonder that pay-to-publish house keeps returning your novellas (and the money)? This isn’t nanotech warfare. The ‘sand’ in question will be bigger than that – the size of actual sand, eventually. Each grain is a tiny robot element, able to communicate with, and magnetically adhere to, its fellow grainbots. For why? To create objects. One minute it’s a pile of sand – the next, a chair.
I don’t need a sand-chair. If I have a pile of sand, I’ll sit on the pile of sand.
Again, you’re missing the beauty of this idea – developed, incidentally, by the Distributed Robotics peeps at everyone’s favourite ideas academy, MIT. Sand-sized robots are some way off, but the MIT guys have proven their concept with larger 1cm cubed models and computer simulation – basically, you drop a tiny model chair into a big pile of ‘smart sand’ and yank a full-size chair out.
Alright, that’s pretty cool. But how do those gritty critters know what to build?
By talking to each other. Most robots will be surrounded by others, while those on the edge of the pile will report air on one side. But the luckiest grainbots will border something that’s neither robot nor air – the model chair. They tell the free robots what shape to form, scaling it up as programmed. Excess sand will fall away, revealing your brand new sand chair. Beautiful.