Minimal ingredient for CISS magnetoresistance in two-terminal junctions
In plain words
Symmetry theorems say that in the simplest setup, with one entrance, one exit, low voltage and no energy loss along the way, a molecule cannot produce the measured spin signal. Experiments still see a signal, so something must break the assumptions of these theorems.
Precise statement
For a two-terminal junction of a chiral molecule, time-reversal symmetry forbids spin polarization of the current in single-channel coherent transport between nonmagnetic leads (Bardarson theorem, J. H. Bardarson, J. Phys. A 41, 405203, 2008, https://doi.org/10.1088/1751-8113/41/40/405203), and Onsager reciprocity forbids any magnetoconductance asymmetry $G(M) - G(-M)$ in linear response when one electrode is a ferromagnet of magnetization $M$, for any number of channels. Identify which ingredient (dephasing or leakage channels, electron-electron interaction, electron-vibration coupling, nonlinear bias, interface spinterface effects) produces magnetoconductance asymmetry of the measured size, and prove which are necessary. An answer is a model calculation giving the measured asymmetry and its bias dependence.
What would settle it
Model and first-principles calculations of nonlinear magnetotransport compared with bias- and temperature-dependent measurements on the same molecule.
Related problems
- Special case of Why chiral molecules polarize electron spins so strongly