Morphology and electro-optical properties of materials in solar cell
The interest for conjugated systems, which particular photo-physics characteristics emanate from the existence of a cloud of π electrons delocalized, is coming among others from the possibility to incorporate them as active components in a large range of optoelectronic devices, such as organic light emitting diodes , solar cells or photovoltaics , field effect transistor , and (bio)chemical sensors. This sector, also called plastic electronics, is growing today by leaps and bounds. Even if these discoveries have been applied to concrete commercial applications, a lot of questioning remains about the principle of functioning of those devices and about the photo-physics properties of conjugated materials. Consequently the full potential of those materials and devices has not been exploited yet.
In previous studies, we investigate with a combination of ab-initio, semi-empirical and micro-electrostatic calculations, the case of the Pentacene – Fullerene heterojunction [4,5].
This first work on the Pentacene – Fullerene heterojunction has been realized on a planar (and not realistic) interface. Indeed, in reality, this kind of interface is much more complex, the interface must be in three dimensions with inclusion of pentacene molecules in the bulk of fullerene and vice versa. To determine several possible structures, we have performed large-scale molecular dynamics at high temperature on heterojunction. Then with the geometries of the complex heterojunction generated, we will study all the electronic process involved in this kind of system: transport of excitation, charge separation/recombination, and transport of charge.
Thanks to the previous allocation we have been able to generate with MM/MD new interfaces between TQ1 (a low band gap polymer - donor) and PC70BM (acceptor). Geometries were then used for the following application:
(i) to perform QM calculation and study the relationship between the morphology and the charge transfer state separation.
(ii) to perform MC simulation with a home made code compiled on triolith.
The MC code is now efficiently parrallelized OpenMP/MPI and allow us to treat large system (1000 nm3). 
We have obtained very promising results for the mobility-field and mobility-temperature dependance in C60 with inclusion of polarization.  Moreover, we are now able to treat several charges simultaneously at the interface between donor and acceptor materials. 
To continue in this direction we would need an allocation of the same amount of hours than the previous one (200 000 cpu hour/month).
 Burroughes, J.H.; et al., Nature, 1990, 347, 539–541
 A.; Scully, S.; Hardin, B.; Rowell, M.; McGehee, M., Mater. Today, 2007, 10 (11), 28-33
 Reese, C.; Roberts, M.; Ling, M.; Bao, Z. Mater. Today, 2004, 7 (9), 20–27
 Linares, M.; Beljonne, D. ; Cornil, J.; at el.J. Phys. Chem. C 2010, 114, 3215.
 Verlaak, S.; Beljonne, D.; et al. Adv. Funct. Mat. 2009, 19, 3809.
 Volpi, R.; Stafström, S.; Linares, M. JCP, 2015, 142, 094503.
 Volpi, R.; Kottravel, S.; Nørby Steen, M.; Stafström, S.; Linares, M. J. Chem. Theory Comput., 2016, 12, 812−824
 Volpi, R.; Nassau, R.; Steen Nørby, M:; Linares, M. ACS Appl. Mater. Interfaces, 2016, 8, 24722-24736