Abstract
A high electron mobility polymer, poly{[N,N’‐bis(2‐octyldodecyl)‐naphthalene‐1,4,5,8‐bis(dicarboximide)‐2,6‐diyl]‐alt‐5,5’‐(2,2’‐bithiophene) (P(NDI2OD‐T2)) is investigated for use as an electron acceptor in all‐polymer blends. Despite the high bulk electron mobility, near‐infrared absorption band and compatible energy levels, bulk heterojunction devices fabricated with poly(3‐hexylthiophene) (P3HT) as the electron donor exhibit power conversion efficiencies of only 0.2%. In order to understand this disappointing photovoltaic performance, systematic investigations of the photophysics, device physics and morphology of this system are performed. Ultra‐fast transient absorption spectroscopy reveals a two‐stage decay process with an initial rapid loss of photoinduced polarons, followed by a second slower decay. This second slower decay is similar to what is observed for efficient P3HT:PCBM ([6,6]‐phenyl C~61~‐butyric acid methyl ester) blends, however the initial fast decay that is absent in P3HT:PCBM blends suggests rapid, geminate recombination of charge pairs shortly after charge transfer. X‐ray microscopy reveals coarse phase separation of P3HT:P(NDI2OD‐T2) blends with domains of size 0.2 to 1 micrometer. P3HT photoluminescence, however, is still found to be efficiently quenched indicating intermixing within these mesoscale domains. This hierarchy of phase separation is consistent with the transient absorption, whereby localized confinement of charges on isolated chains in the matrix of the other polymer hinders the separation of interfacial electron‐hole pairs. These results indicate that local, interfacial processes are the key factor determining the overall efficiency of this system and highlight the need for improved morphological control in order for the potential benefit of high‐mobility electron accepting polymers to be realized.