Of course, that's energy we'd have to put in to water to raise the temperature--what we really want is to get energy out. What we need then is water that is spontaneously lowering its temperature, freezing, or best of all condensing. Changes of state in this direction emit heat.
This principle is already used in steam radiators. Steam is piped through to the radiator where it condenses, releasing massive amounts of heat to the room. Of course, the steam was previously created in the basement, so there's no theoretical net change of energy. A way must be found of spontaneously lowering the energy state of free steam or water to water or ice.
We now turn to another characteristic of water. Each molecule is an "electric dipole". You can think of this as being kind of like a magnet (which is a "magnetic dipole"). One of the properties of an electric dipole is that it turns to align itself in an electric field (exactly like a magnetic dipole aligns itself to a magnetic field). This principle is famously used in a microwave oven. The oscillating electric field causes the water molecules to flip back and forth, which molecular motion is heat, cooking the food.
But what if the field were not oscillating, or was oscillating in such a way as to destroy the natural movements of the water molecules? They would slow down, right? And once they got slow enough, hydrogen bonding would occur, linking the water molecules together more tightly--i.e., the water would freeze. This state change is accompanied by a release of energy, the same energy requirement calculated above to melt water. A similar arrangment could be applied to atmospheric water vapor, though steam's low density would make using enough of it in a smallish device a bit of a chore.
The only question that remains is to ask how much energy it would take to put this plan into action. If it's equal to or more than the amount of energy we get out, it's pointless. In fact, it takes next to no energy at all. The only requirement is the electric field. An electric field is merely a voltage--no current is required. From electronics, P = IV. With I = 0, the power drawn by this device would be essentially zero.
|< BarBeeQueue | BBC White season: 'Rivers of Blood' >|