Inconsistency between Thermodynamics and Probabilistic Quantum Processes

We have developed a Gedankenexperiment that reveals an inconsistency between quantum theory and thermodynamics. The experiment, which may also be practically performed, introduces an interferometer coupled to a collection of two-level systems that absorb and emit radiation statistically. The experiment therefore combines coherent superposition of wave functions with the probabilistic description of absorption and emission processes. We show that this combination of coherence and collapse forces an isolated system to reduce its total entropy, starting from standard, thermodynamic equilibrium. The Gedankenexperiment demonstrates that the basic constituents of quantum physics, namely coherence and probabilistic events, contradict the basic constituent of thermodynamics, the second law.

Inconsistency between Thermodynamics and Probabilistic Quantum Processes, D. Braak and J. Mannhart, arXiv: 1811.02983 (2018)

Closely related articles are available here:

Lossless Currents at High Temperatures, J. Mannhart and D. Braak, Journal of Superconductivity and Novel Magnetism (2018)

Non-reciprocal Interferometers for Matter Waves, J. Mannhart, Journal of Superconductivity and Novel Magnetism 31 (2018)

Phase Filters for a Novel Kind of Asymmetric Transport, J. Mannhart, P. Bredol, and D. Braak, Physica E: Low-dimensional Systems and Nanostructures 109 (2019)

 
 
 

FIG. 1: Sketch illustrating in simple terms photons being absorbed in a probabilistic process (“quantum jump”) and later reemitted by spontaneous emission. An atom anchored in a crystal lattice may absorb a photon impinging on it from any direction, as long as the incoming mode has finite overlap with the dipole mode of the TLS. The excited atom later relaxes by emitting a photon as an outgoing spherical wave with arbitrary polarization, independent of the direction or polarization of the previously absorbed photon ((a) and (b)).
FIG. 2: The deviation of the total entropy of a system according to Fig. 1 from its (maximal) initial value.
Go to Editor View