Improved Performance of Perylene-Based Photovoltaic Cells Using Polyisocyanopeptide Arrays

Abstract

Photovoltaic devices incorporating perylene-substituted polyisocyanide materials have been demonstrated, using blend systems with polythiophene- and polyfluorene-based conjugated polymers. Prototypical structures with nominal (1:1) blend weight ratios of the polyisocyanide with poly(3-hexylthiophene) (P3HT) and poly(9,9-dioctylfluorene-co-bis(N,N-(4-butylphenyl))-bis(N,N-phenyl)-1,4-phenyldiamine) (PFB) readily showed an order of magnitude improvement in power conversion efficiency, as compared to analogous blend architectures using a perylene (PDI) monomer, whereas the performance of strongly phase-separated F8BT (poly(9,9- dioctylfluorene-co-benzothidiazole) blend devices showed no such improvement. We propose that the use of polyisocyanide chains as a molecular template offers a method by which the morphology and connectivity of photovoltaic blends can be modified and enhanced.Peer reviewe

Photophysical studies of poly-isocyanopeptide based photovoltaic blends

“…The TCPSC data and measured lifetimes are shown in Figure 3 and Table 1 respectively, along with PL efficiency and PL quenching efficiency, where known from previous studies. [15] These results have a clear interpretation in the case of the pery-PIC blends, where the PL lifetime of pristine pery-PIC is measured to be 6.5ns; the lower the PLE (and the higher the efficiency of luminescence quenching due to charge separation) the shorter the measured lifetime, commensurately. The lifetimes of emission from the PDI blends are generally rather longer than that of the pristine PDI; the most likely explanation for this is that the emissive state of PDI has been entirely quenched by geminate charge separation and the weak signal observed is from traps within PDI crystallites or aggregates.…”

“…The absorption spectra of the samples are shown in figure 2(a), with the optical densities of corresponding blends being comparable, as with the previous device studies. [15] In this case, to a first approximation, we do not expect optical density to affect our results, due to the relative nature of the photo-induced absorption measurements.…”

“…We note that, whilst both having a predominantly hydrophilic character, the surface energies of the spectrosil substrates used for optical measurements may be somewhat different from the PEDOT:PSSplanarised substrates used for device work. [15] Whilst the observed top-surface morphologies do not change qualitatively, it is possible that the blend composition and morphology at the substrate is modified, [19,20] with commensurate effects on the orientation of the pery-PIC/PDI chromophores in the interfacial layer.…”

“…The thin-film samples used in this paper were identical in preparation to those used for the morphological studies given in Foster et al [15] In brief, solutions were prepared by measuring a 5 1:1 weight ratio [16] (6mg/ml each) of pery-PIC/PDI with P3HT/PFB/F8BT in chloroform. A further 5 thin-film samples, one from each pristine material, were prepared for comparative TCSPC readings.…”

“…[12] Great advances have been made regarding issues such as low bandgap and spectral matching, morphology, stability [13] and facility of processing. [14] In our recent work, Foster et al, [15] it was found that introducing arrays of the organic n-type semiconductor perylene onto a backbone of helical polyisocyanopeptide (PIC) had a marked effect on the photovoltaic (PV) properties of blend systems with archetypal p-type polyfluorene and polythiophene conjugated polymers, including up to an order of magnitude increase in device power conversion efficiency, relative to the respective PDI-blend analogues. A summary of the 3 key results from this earlier work is given in tabular form in Appendix B.…”