The field of organic photovoltaics research has witnessed a renaissance in recent years.
On the materials side, two recent developments have invigorated, if not revolutionized photoconversion
in organic materials. First, a new generation of highly light-absorbing
organic semiconductors (OS), often referred to as non-fullerene acceptors (NFAs), have propelled
device efficiencies of organic heterojunction solar cells to new levels, >18%[1], up from 11% for
fullerene based solar cells. Second, charge generation has recently been demonstrated for a number
of single component organic systems, or homojunctions.[2] This is significant because homojunctions
have a number of advantages compared to heterojunctions, e.g., simpler materials design and higher
morphological stability.

New regimes of photophysics are reached in these new materials that are currently not well understood
suggesting that more powerful experimental measurements and theoretical and computational models
are urgently needed to rationalize, explain and further build on these advances. This concerns in
particular the development of improved spectroscopic methods, with unprecedented, for device measurements,
time resolution as well as novel scale-bridging computational tools from the nanoscale, on which we
focus in the present contribution, to the device level and their integration with device measurements.

Here we present a powerful (non-adiabatic) molecular simulation method that we have developed
to propagate electronic excitations (excitons)[3,4] as well as charge carriers (electrons, holes)[5,6] through
nanoscale materials by solving the time-dependent Schrödinger equation coupled to intramolecular
and lattice vibrations. Our simulations give a ``first-principles" view on these processes
that is free of the many assumptions that strongly limit the applicability of traditional theoretical models
(band theory, polaronic hopping, Onsager-Brown,...). In particular, our methodology takes into
account the electronic quantum dynamics of delocalization/localization of excitons and charge carriers and allows
us to understand how the quantum dynamics is affected by the dynamic and static disorder that is so
characteristic for these materials.

Particular focus will be placed on discussing our recent non-adiabatic molecular dynamics simulation of exciton diffusion
in molecular organic crystals[3]. We find that in materials featuring some of the highest diffusion
lengths to date, e.g. the non-fullerene acceptor Y6, the exciton propagates via a transient quantum
delocalization mechanism, reminiscent to what was previously proposed for exciton transport in P3HT-based
nanofibres [7]. Yet, the extent of quantum delocalization of excitons is rather modest compared to charge carriers,
even in Y6, and found to be limited by the relatively large exciton reorganization energy.
Based on these simulations, we will chart out a path for rationally improving exciton transport in organic
optoelectronic materials that could remove some of the intrinsic limitations of present-day organic
optoelectronic devices.