© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2010, 22, 4987–4992 4987 www.advmat.de www.MaterialsViews.com COMMUNICATION wileyonlinelibrary.com By Andrea Maurano, Rick Hamilton, Chris G. Shuttle, Amy M. Ballantyne, Jenny Nelson, Brian O’Regan, Weimin Zhang, Iain McCulloch, Hamed Azimi, Mauro Morana, Christoph J. Brabec, and James R. Durrant* Recombination Dynamics as a Key Determinant of Open Circuit Voltage in Organic Bulk Heterojunction Solar Cells: A Comparison of Four Different Donor Polymers [∗] Mr. A. Maurano, Dr. R. Hamilton, Dr. C. G. Shuttle, Dr. B. O’Regan, Dr. W. Zhang, Prof. I. McCulloch, Prof. J. R. Durrant Departments of Chemistry Imperial College London South Kensington SW7 2AZ (United Kingdom) E-mail: j.durrant@imperial.ac.uk Dr. A. M. Ballantyne, Prof. J. Nelson Departments of Physics Imperial College London South Kensington SW7 2AZ (United Kingdom) Dr. H. Azimi, Dr. M. Morana, Prof. C. J. Brabec Konarka Austria Altenbergerstrasse 69, A-4040 Linz (Austria) Dr. H. Azimi Christian Doppler Laboratory for Surface Optics Johannes Kepler University Linz (Austria) DOI: 10.1002/adma.201002360 Solution-processed organic solar cells based on blends of semi- conducting polymers and soluble fullerene derivatives are showing impressive advances in photovoltaic power conversion efficiency, with recent reports of efficiencies in excess of 6%. [1] One of the key remaining factors limiting the performance of such blend or ‘bulk heterojunction’ solar cells is that they generally exhibit relatively modest voltage outputs, with the energy corresponding to the open circuit voltage, V OC , typically being less than half the optical gap. This V OC has been shown to be correlated to the energy levels of the donor and acceptor materials of the bulk heterojunction (BHJ). [2] In this paper, we compare the V OC of BHJ fabricated from four different donor polymers, and show that this voltage depends not only upon the material energetics but also upon the lifetimes of charge carriers within the blend. Previous studies of the role of material energetics in deter- mining V OC have led to the empirical relation: V oc = (1/e )( IP - donor EA acceptor ) - 0.3V (1) where IP donor and EA acceptor are the ionization potential and electron affinity of the donor and acceptor respectively and the constant 0.3 V was determined empirically. [2] Other studies have considered alternative factors that can limit V OC , including mor- phology, [3] shunt resistance, [4] electric field dependent geminate recombination, [5] reverse saturation current, [6] energetic dis- order [7] and the presence of interfacial charge transfer states. [8] We have recently undertaken a study of the role of bimolecular recombination dynamics in limiting the V OC of BHJ devices based upon poly(3-hexylthiophene) (P3HT) : [6,6]-phenyl C 61 butyric acid methyl ester (PC 61 BM) blend films. In particular, we determined the recombination flux as a function of charge den- sity in the blend film and demonstrated that device open circuit corresponds to the condition when the flux of charge photogen- eration ( J photo ) and bimolecular recombination ( J rec ) are equal and opposite, i.e.: J photo = – J rec . It follows from such analyses that device V OC should be dependent upon the dynamics of recombi- nation, and specifically upon the magnitude of the bimolecular recombination rate coefficient ( k rec ). [9] Whilst many studies have considered the role of such recombination dynamics in lim- iting device V OC , [5–8] such studies have not previously demon- strated that an empirical measurement of k rec can be employed to explain quantitatively differences in V OC between devices.. In this paper we demonstrate that we can employ transient pho- tovoltage and charge extraction measurements to calculate cor- rectly differences in open circuit voltage observed between BHJ devices employing different donor polymers. In the present study we analyze V OC for BHJ solar cells employing four different photoactive layers: P3HT blended with PC 61 BM (1:1 weight composition) (annealed at 140 °C), poly(3-hexylselenophene) (P3HS) blended with PC 61 BM (in 1:1 weight composition) (annealed at 150 °C), poly[2,6-(4,4-bis- (2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b ′]dithiophene)-alt-4,7- (2,1,3-benzothiadiazole) (PCPDTBT) blended with [6,6]-phenyl C 71 butyric acid methyl ester (PC 71 BM) (1:1 weight composition) and poly[(4,40-bis(2-ethylhexyl)dithieno[3,2-b:20,30-d]silole)-2,6- diyl-alt-(4,7-bis(2-thienyl)-2,1,3-benzothiadiazole)-5,50-diyl], here called Si-PCPDTBT, bended with PC 71 BM (1:1 weight composi- tion). By employing transient photovoltage, transient photocur- rent (TPV/TPC) [9] and charge extraction (CE) [10] techniques, we demonstrate that V OC is dependent not only upon the energy levels of the materials used, but also upon the charge carrier dynamics. In particular we demonstrate a simple relationship between V OC and k rec , which we believe to be a readily appli- cable and powerful tool to relate device V OC to film interface structure and nanomorphology. In Figure 1a we show typical current density–voltage ( J–V) curves under simulated AM 1.5 illumination for devices made