This work provides some clarifications on the long-known Redfield theory of spin relaxation, which considers a small system coupled to a large ensemble of particles in thermodynamic equilibrium. The derivation is rewritten so as to emphasize the usual assumptions and their consequences. In particular, the respective roles of the detailed balance balance assumption and the non-commutativity of the spin–bath coupling are discussed.

We studied spin dynamics in nuclear spin systems at low magnetic fields, where strong coupling among nuclear spins of the same kind, here ¹³C, is expected. Such conditions are known to be favorable for coherent polarization transfer. However, to our surprise, interactions with other nuclei, i.e., protons, lead to a breakdown of the strong coupling conditions. By using a two-field nuclear magnetic resonance approach, we can manipulate low field-spin dynamics and reintroduce strong coupling.