< Terug naar vorige pagina

Publicatie

Increasing the applicability of conjugated polymers for organic photovoltaics: continuous flow synthesis and elucidation of the importance of structural defects

Boek - Dissertatie

In the past decade, a tremendous amount of research has been performed in the field of organic photovoltaics, resulting in power conversion efficiencies (PCE’s) of over 10% for single junction polymer-based solar cells. Unfortunately, the photovoltaic performance can differ from batch to batch and it remains rather difficult to synthesize conjugated polymer materials on a large scale and with a high reproducibility in material properties, a necessity for organic photovoltaics to become an economically viable technology. In this PhD thesis, continuous flow chemistry was used as an easily scalable synthesis method with an intrinsically high output reproducibility. Also the influence of homocoupling, a synthesis based defect in the polymer backbone, on the optoelectronic properties and the photovoltaic performance was investigated, as this can also influence the reproducibility of the synthesis method. At first instance, we have focussed on the flow synthesis of the frequently used N,N’-dialkyl-6,6’-dibromoisoindigo building block. Both symmetrically and asymmetrically alkylated derivatives were synthesized. The synthetic pathway, which starts from the commercially available 6-bromoisatine, has three distinct reaction steps: a nitrogen alkylation, a reduction and a condensation step. Depending on the physical composition of the reaction mixture (homogeneous/heterogeneous) or scale, either a glass-chip, a tubular or packed-bed reactor was used. Through careful batch synthesis, reagent and solvent optimization, all reaction steps were successfully transferred to flow. In three cases the flow chemistry process outperformed the batch process, while in two other cases similar yields were achieved and in one case the batch process slightly outperformed the flow process. All flow protocols have the intrinsic property of being easily scalable, allowing for large scale production of N,N’-dialkyl-6,6’-dibromoisoindigo derivatives. A wide variety of push-pull conjugated polymers affording a reasonable high photovoltaic performance nowadays exist. Each donor or acceptor building block of the conjugated polymer has a different synthesis protocol and different reaction types have to be employed. One common aspect that these polymers share, is that the actual polymerization is frequently done by a Stille polycondensation reaction. Unfortunately, this reaction is difficult to scale in batch and frequently batch-to-batch variations arise, resulting in variations in the solar cell performance. Therefore, our next step was to translate the Stille polycondensation reaction to flow. The high-performance benzodithiophene–thienopyrroledione copolymer PBDTTPD was used for this purpose. The continuous flow production process showed a high reproducibility and delivered a constant output of high quality material with uniform characteristics. The flow process was also successfully upscaled, yielding 1.55 g of material. The photovoltaic performance of the PBDTTPD material was further increased from 7.2 to 9.1% by incorporation of an ionic polythiophene-based cathodic interlayer. The molecular weight and dispersity of conjugated polymers have a major effect on the final device performance through a combination of processing and morphological considerations. Proceeding from the previous work, we investigated the potential of continuous flow chemistry to tune the final molecular weight of the synthesized polymers. The low bandgap polymer PffBT4T-2OD or ‘PCE-11’, was used for this purpose as it provides highly efficient bulk heterojunction solar cells and its temperature dependent aggregation behaviour is dependent on the molecular weight of the polymer. The influence of various reaction parameters on the molecular weight of the polymer is investigated in terms of temperature, monomer concentration and injection volume of the polymerization mixture. The polymers were tested in organic solar cells in combination with PC71BM as the acceptor phase. It was observed that diffusion has a large influence on small scale injections and, in order for the process to be readily scalable to continuous operation, the injection volume has to be chosen large enough to screen for conditions. Variation of the monomer concentration allowed the highest control over the molecular weight. The same protocol was then also applied to a structurally similar polymer with longer alkyl side chains, PffBT4T-2DT, affording important advantages in terms of processing due to its higher solubility. Push-pull copolymers ideally show a perfect alternation of electron-rich (donor, D) and electron-poor (acceptor, A) building blocks. Unfortunately, a common side reaction of the Stille cross-coupling is the formation of homocoupled products, resulting in polymer main chain structures with a double donor or acceptor block and not a perfect alternation of the two. It is rather difficult to control the extent of homocoupling and this frequently leads to batch-to-batch variations and lower PCE’s than expected. The specific influence of homocoupling in the donor unit of the polymer has been previously investigated, but no records exist on homocoupling of the acceptor unit. Therefore, we used a dithienosilole-quinoxaline copolymer to investigate the actual influence of acceptor homocoupling on the polymer’s optoelectronic properties and photovoltaic performance. A homocoupled quinoxaline monomer was prepared and added in specific ratios to the polymerization mixture. The different polymers were analysed by UV-VIS, MALDI-TOF and cyclic voltammetry and then tested for their photovoltaic performance. Homocoupling induces a blue-shift in the absorption spectrum and, in higher quantities, causes a strong decrease in photovoltaic performance.
Aantal pagina's: 156
Jaar van publicatie:2017
Toegankelijkheid:Open