How water becomes ice
In development
No molecule carries the plan for a crystal. We are asking whether the lattice, and the way it bends light, can emerge from physics alone.
Water is the best place to begin, because everyone has held the result.
A single water molecule knows nothing about ice. The six-sided lattice, the hardness, the way ice floats: none of it is a property of one molecule. It appears when many molecules settle together, each one pulled on by its neighbours through hydrogen bonds.
So ice is a direct test of the central idea. If an engine is given only oxygen, hydrogen and the conditions, does ice come out?
What is known
A water molecule is changed by its surroundings. Quantum calculations in the published literature find that a molecule inside liquid water or ice is roughly 10 to 20 percent more polarisable, meaning more responsive to light and to electric fields, than the same molecule alone in vapour. Hydrogen bonding does that.
Curiously, the standard formula used to read molecular properties out of a refractive index reports almost the same value for vapour, liquid and ice. The formula assumes each molecule sits in a smooth, featureless surrounding. In a hydrogen-bonded network the surroundings are anything but featureless, and the formula averages the difference away.
It is a small, clean example of the larger point: treat a molecule’s properties as fixed numbers and you can miss what the network is doing.
What we are working on
- Building small clusters of water from the engine’s own physics, and watching how the molecules change as the cluster grows.
- A paper on the polarisability puzzle above, written with a collaborator.
What would tell us more
Current results are not yet written up here.