Say I promote a molecule to a triplet state via repeated electronic excitation events. Here, this is going to mean that an electron from the HOMO level, originally with a spin-UP and spin-DOWN electron, will be promoted to the LUMO level such that its spin is the same as the electron left-over in the HOMO level (we can then use the Pauli exclusion principle to partially understand that the triplet state will generally have a longer lifetime than the typically higher energy singlet state). The classic example of a molecule in a triplet state is oxygen (though here, this is also the ground state, which is somewhat unusual):
Prior to returning to the ground state via phosphorescence or a reverse intersystem crossing event, let's say that the triplet $T_1$ state molecule reacts with a reducing agent via an exothermic bond formation process, and that the reaction is reversible. Consider that there are numerous instances where triplet state reactivity is higher than ground state reactivity.
The following two things are unclear to me:
(1) When the reducing agent dissociates, is the molecule in the $T_1$ triplet state or the $G_0$ ground state? If the molecule is still in the triplet state, can it enter the ground state while remaining bound to the reducing agent?
(2) Call the energy gap between the relevant $T_1$ triplet state and the $G_0$ ground state $E_1$. Call the energy released into a surrounding thermal bath by the reducing agent reacting with the molecule in the $T_1$ triplet state $E_2$. Must $||E_2|| > ||E_1||$?
My guess is that the pair of electrons donated by the reducing agent will necessarily set $||E_2|| > ||E_1||$ and that the molecule will be left in the ground state upon dissociation of the reducing agent since the electrons will be reallocated to properly satisfy spin-pairing requirements.