Thermal Stabilisation of Polymer–Fullerene Bulk Heterojunction Morphology for Effi cient Photovoltaic Solar Cells

International audiencelevel, and operational stability. [ 3 ] Most of BHJ photoactive blends are composed of a mixture of an electron-donor polymer and an electron-accepor fullerene derivative, where the latter material is typically a soluble C 60-fullerene (PC 61 BM) or C 70-fullerene (PC 71 BM) derivative (Figure 1). The BHJ layer is sandwiched between charge carrier selective interlayers and the electrodes. The bottom electrode is typically indium tin oxide (ITO) or other transparent conductors. Interlayers choice governs the polarity of the photovoltaic cells. Metal oxides such as TiO x or ZnO are commonly used as electron selective layer whereas MoO x or conducting polymers (PEDOT:PSS) are used as hole transporting layers. An optimised BHJ layer requires specifi c phase segregation of the BHJ donor-acceptor components to allow optimum charge carrier photogeneration in the blend and charge perco-lation pathways for effi cient electron and hole collection to the respective electrodes. An important morphological parameter of the BHJ blend to achieve large PCEs is that nano-sized fullerene crystallites are necessary within the polymer matrix to prevent electron-hole recombination mechanisms. [ 4–7 ] Thus, the domain size must be in the order of the excitons diffusion length, which typically ranks from 3 to 30 nm. [ 8 ] Such optimal polymer-fullerene blend morphology is achieved with a different efficiency depending on the material combinations. [ 9 ] Optimized phase segregation can be promoted using appropriate solvent(s) and/or specifi c solvent additives during blend deposition as well as post-deposition fi lm processing such as thermal or solvent annealing. [ 10 ] Semicrystalline polymers such as poly(3-hexylth-iophene) (P3HT, Figure 1) tend to expel fullerenes during their crystallization into nano-objects upon drying of the solvent or during post-fi lm deposition thermal annealing. This property enabled to fi nely tune P3HT:PCBM blend morphology and led to a tremendous amount of data concerning OPV cells based on this specifi c polymer. [ 11 ] However, P3HT cells are severely limited in terms of the maximum achievable PCEs. Therefore, low band gap polymers, which can harvest a larger portion of the solar spectrum, were developed to reach greater performances. Unfortunately, several of these high-potential polymers are less crystalline and do not have such a strong tendency for molecular organization. As a consequence, manipulating BHJ morphology of less crystalline/amorphous polymers is not trivial. Solvent additives, such as 1,8-diiodooctane or 1,8-octanedithiol for example, enable to preferentially solvate fullerene derivatives rather than the polymer, were chosen to tune BHJ morphology and achieve effi ciencies >9%. Thus, the major problem that the The use of a bulk heterojunction (BHJ) blend of an electron-donor and an electron-acceptor organic semiconductors to fabricate photovoltaic solar cells and to understand fundamental light-to-charge phenomena in organic solids has attracted the interest of the international scientifi c community for the last 20 years. These efforts recently led to the demonstration of lab-scale organic photovoltaic (OPV) cells with power conversion effi ciencies (PCE) of 9.2% and 10.6% for single cells [ 1 ] and tandem cells [ 2 ] confi gurations, respectively. OPV cells are becoming a credible revolutionary thin-fi lm photovoltaic technology with advantages such as lightweight, mechanical fl exi-bility, roll-to-roll large area and low-cost solar module production. The three major challenges to OPV module realization are the cost of the active/encapsulation layers, effi ciency at modul