The simulation of nonadiabatic molecular processes that involve a strong correlation between electronic and nuclear motions has made significant strides in recent years [1-5]. What was once a collection of specialized methods - developed independently within various groups - has now evolved into a more unified and collaborative approach [6,7]. While early challenges consisted of fragmentation within the research community and a lack of standardized benchmarks for evaluating different methods, recent efforts have focused on bridging these gaps [8].

Community-driven efforts have enhanced the understanding of various approaches among researchers in the field: from rigorous quantum wavepacket methods, such as those related to the multiconfigurational time-dependent Hartree approach [9] or based on variational Gaussian wavepackets [10,11], semiclassical methods using classically moving Gaussians or coupled trajectories [12,13], to mixed quantum-classical techniques in the surface-hopping [14,15] or Ehrenfest-type [16,17] formalism.

Furthermore, the collaboration within the community has led to the standardization of benchmarks for various classes of photochemical and photophysical phenomena [8]. Although significant work remains to fully understand the limitations of different theoretical methods, the computational tools for simulating ultrafast photoinduced molecular processes in small and medium-sized molecules are becoming well established [18-23].

Looking ahead, a pivotal question is how to validate, integrate, and advance existing methodologies to tackle increasingly complex nonadiabatic problems [24-27]. These encompass phenomena extending from photostability, photoswitching, photoinduced electron and proton transfer in active environments [27-32] - e.g. solvents, biological or inorganic matrices - to excitonic systems with numerous electronic states [3,4,33-35], photoactive molecular crystals [36], and molecules in optical cavities generating hybrid light-matter states [37-41]. Further issues involve developing methods to simulate multi-step processes spanning femtoseconds to hundreds of picoseconds [42,43].

Modeling such complex processes from first principles involves challenges in two main directions [44]: (i) the description of the interactions between the different components of the photoactive systems at the purely static level, as well as the definition of protocols to construct realistic multi-state multi-mode Hamiltonian models; (ii) the simulation of the light-initiated molecular dynamics involving a large number of degrees of freedom and electronic states and spanning different time scales.

The workshop has two main objectives. First, it seeks to identify the critical bottlenecks that must be overcome to realistically model complex photoactive molecular systems, considering both static and dynamical aspects. Second, it aims to develop common strategies to improve and combine current methods for nonadiabatic dynamics, enabling them to handle increasingly complex problems, beyond gas phase photochemistry, bridging the ultrafast and the longer timescales.

Key phenomena that need a collective computational effort encompass the photochemistry or photophysics of medium-sized molecules in solution, the first principles description of exciton mobility and trapping in molecular materials, and the dynamical simulation of hybrid light matter states. We aim to bring the members of the nonadiabatic dynamics community together to discuss future computational strategies to tackle the challenges posed by these relevant complex problems.

Registrations will open in March 2024.