Thermodynamics prescribes entropy as the measure of irreversibility. Statistical mechanics then tells us that entropy is a measure of randomness. Mechanics is fundamentally deterministic and reversible, however. So it is not really clear where entropy comes from. Three possible sources are often suggested:
- quantum fluctuations;
- noise from the environment;
- dynamical chaos.
By exploiting the flexibility of modern cold atom techniques to make instructive systems that are both theoretically tractable and experimentally realizable, we can study all three of these fundamental phenomena. But each of them alone usually presents enough difficulty for a single project. So our work follows three complementary themes.
Quantum fluctuations as heat: Free quantum fields are trivial, unless they propagate in non-trivial classical backgrounds. Then quasi-particles may be created spontaneously. Can this represent entropy production, even without quantum nonlinearity or environmental noise?
Finding the bath within the system: Much has been learned from the model of 'quantum Brownian motion', in which a quantum system interacts with a large 'bath' of very simple systems that are not directly observed. We want to see how this artificial model can be a natural description for large systems in which the observed degree of freedom is not pre-defined, but stands out from its peers dynamically.
Microthermodynamics: Many texts explain entropy with hand-waving references to deterministic chaos. But if chaos is really sufficient for thermodynamics, then even very small systems may be thermodynamic. To see if this is possible, we construct model quantum systems with only a few degrees of freedom, but with nonlinear dynamics designed to emulate standard thermodynamic phenomena. Or is it more than emulation?