posted on 2017-02-06, 06:14authored byDaeneke, Torben Jost
Dye-sensitized solar cells (DSCs) are regarded to be a promising technology with the ability to convert sun light directly into electricity at competitive costs. Unlike conventional solar cells, these devices are not made from high purity semiconductor junctions but from an assembly of functional materials, which under illumination undergo a series of charge transfer reactions leading to charge separation and the generation of electricity. In the typical assembly, a mesoporous wide band gap semiconductor is sensitized by a dye. This sensitized working electrode is in direct contact with an electrolyte containing a redox couple. The electrolyte layer is positioned between the working electrode and the counter electrode which is usually a conductive glass sheet coated with a thin layer of a catalyst. The benefit of the DSC is that each function in the solar cell is carried out by a different material, leading to lower requirements of the materials and allowing convenient fine tuning of each individual component.
While most components of the DSC have been thoroughly optimized, the electrolyte still is the limiting factor when it comes to up scaling and commercialization of this technology. In operation the redox couple in the electrolyte has a dual role. It is required to reduce the photooxidized dye and facilitate charge transport between the electrodes to allow for the continuous generation of electricity. To date, most DSC designs make use of the iodide-triiodide redox couple. While this leads to relatively high efficiencies, the complicated two electron redox chemistry of this couple causes intrinsic efficiency losses, limiting the achievable power conversion efficiency. Furthermore, the corrosive nature of I-/I3- complicates the design of durable, large scale solar cells, since cost reducing metal substrates and charge collection grids are dissolved by these electrolytes. As a result, DSCs were frequently efficient in the laboratory, but devices with the commercially sufficient dimensions were often comparatively inefficient and expensive.
At the beginning of this work, relatively little was known about the requirements for the ‘perfect’ redox couple to replace I-/I3-. A few alternatives had been published, but all the devices showed comparatively low efficiency. In this thesis, we have investigated iron complexes as DSC redox mediators. The ferrocene / ferrocenium redox couple had been tried previously but the resulting DSCs showed very poor energy conversion efficiency.
Ferrocene (biscyclopentadienyl iron) is the widely accepted reference redox couple for the work in non-aqueous solutions and one of the best characterized compounds in the field of electrochemistry. The ferrocene unit has been shown to be highly tunable in its electrochemical properties by derivatization. Furthermore, a wide range of ferrocene derivatives is commercially available. Due to the combination of being well characterized and the availability of a vast compound library, this group of complexes was considered ideal when attempting to define the properties of the ideal DSC redox couple.
In this work, high efficiency DSCs are reported that use a ferrocene / ferrocenium redox couple. It was identified that the decomposition of the ferrocenium ion in air leads to rapid device degradation and low efficiencies. After developing a technique that allowed the DSCs to be filled and sealed under the exclusion of oxygen, and optimizing the DSC design for the use with this new electrolyte, conversion efficiencies of up to 7.5 % were achieved. The reference I-/I3- solar cell reached 6.2 %.
A series of ferrocene derivatives was then studied to investigate the influence of the redox potential of the redox couple on the charge transfer kinetics in the solar cell. It was identified that a threshold exists until which an increased redox potential directly results into an increased voltage and efficiency. After this threshold, the charge transfer reaction between the photooxidized dye and the redox couple becomes sluggish, leading to increased losses and lower efficiencies. An in depth study, utilizing a range of sensitizers and a library of ferrocene derivatives, allowed us to investigate this crucial charge transfer process. Measurements of the reaction kinetics by nanosecond pulsed laser transient absorption spectroscopy allowed us to estimate that a driving force of 0.25 eV is necessary to achieve quantitative dye regeneration (yield > 99.9 %) for typical electrolyte concentrations. With the knowledge gained from this work, any particular sensitizer can be matched with an appropriate redox couple, reducing energy losses in the solar cell due to the poor alignment of redox potentials.
A further focus was the work with aqueous electrolytes containing the ferricyanide / ferrocyanide redox system. The use of water instead of organic solvents has clear benefits due to the abundance of this solvent and environmental considerations. Ferrocyanide is a standard redox couple for aqueous solutions, having a well defined and documented redox chemistry under a wide range of conditions. In this work, we present the first efficient DSCs utilizing this system. Water as an electrolyte solvent for DSCs has been rarely investigated. The influence of parameters, such as pH and buffers, on device performance will have to be studied and the ferricyanide / ferrocyandide redox system forms a solid base for such future work.
In summary, a range of iron complexes has been investigated for the use as redox couples for dye-sensitized solar cells resulting in high efficiencies. Certain requirements of the ideal redox couple have been established, allowing to match sensitizers with a redox couple that optimizes the system and minimizes internal energy losses.
Awards: Winner of the Mollie Holman Doctoral Medal for Excellence, Faculty of Science, 2012.