No abstract

Dye adsorption plays a crucial role in dye-sensitized solar cells. Herein, we demonstrate an in situ liquid-phase analytical technique to quantify in real time adsorption of dye and coadsorbates on flat and mesoporous TiO 2 films. For the first time, a molar ratio of co-adsorbed Y123 and chenodeoxycholic acid has been measured.Dye sensitized solar cells (DSCs) provide a viable alternative to traditional semiconductor solar cells in that they have the potential to harvest solar energy with high efficiency and at low environmental and industrial costs.1 They are based mainly on cheap and nontoxic materials, are roll-to-roll compatible and offer short energy payback times.2 To date the highest certified power conversion efficiency (PCE) for a DSC using iodide/triiodide (I 3 À /I À ) electrolytes and a ruthenium dye is 11.4% although Yella et al. have recently achieved 12.3% using a Co (II/III) tris(bipyridyl)-based redox electrolyte in conjunction with a donor-p-bridge-acceptor (D-p-A) zinc porphyrin dye as sensitizer and Y123 as cosensitizer. [3][4][5][6] At the heart of the DSC is a self-assembled monolayer (SAM) of dye molecules adsorbed on a high surface area mesoporous TiO 2 photoanode and infiltrated with an electrolyte containing the redox shuttle molecule. Dye loading has been found to be crucial for the DSC performance in three complementary ways. First, the amount of the dye adsorbed on the TiO 2 surface and its extinction coefficient determine the fraction of sunlight that can be harvested, which in turn affects the cell's photocurrent (J sc ). Second, the photo-excited dye has to efficiently inject electrons into the semiconductor and must be regenerated by the redox mediator in the electrolyte. For these processes to be efficient, it is mandatory to have only a monolayer of dye on the surface and avoid multilayer buildup through aggregation.5 Third, the dye monolayer must act as a blocking layer that prevents recombination between the injected charge in the semiconductor and the oxidized form of the redox couple in the electrolyte. In addition, the recombination rate has been shown to decrease significantly when using molecular coadsorbates like dineohexyl bis-(3,3-dimethyl-butyl)-phosphonic acid or chenodeoxycholic acid (cheno). 7,8 The blocking layer becomes particularly important when using Co 2+ /Co 3+ or spiro-OMeTAD which offer higher redox potential but suffer from faster recombination rates than the two-electron (I 3Co-sensitization using multiple dyes of complementary spectral absorption has been shown to be a promising approach to improve PCE but is not yet fully understood. 10In addition to the combined optical absorption of the two dyes there is a reported ''concerto effect'', which is likely to be related to rearrangement of the dye molecules on the TiO 2 surface, as well as to the formation of a more effective recombination barrier. These results highlight the need for a better understanding of the dye adsorption dynamics as well as an optimization of the SAM of dye. However, most studies...

Indirect nanoplasmonic sensing (INPS) is an experimental platform exploiting localized surface plasmon resonance (LSPR) detection of processes in nanomaterials, molecular assemblies, and films at the nanoscale. Here we have for the first time applied INPS to study dye molecule adsorption/ impregnation of two types of TiO 2 materials: thick (10 μm) mesoporous films of the kind used as photoanode in dye-sensitized solar cells (DSCs), with particle/pore size in the range of 20 nm, and thin (12−70 nm), dense, and flat films. For the thick-film experiments plasmonic Au nanoparticles were placed at the hidden, internal interface between the sensor surface and the mesoporous TiO 2 . This approach provides a unique opportunity to selectively follow dye adsorption locally in the hidden interface region inside the material and inspires a generic and new type of nanoplasmonic hidden interface spectroscopy. The specific DSC measurement revealed a time constant of thousands of seconds before the dye impregnation front (the diffusion front) reaches the hidden interface. In contrast, dye adsorption on the dense, thin TiO 2 films exhibited much faster, Langmuirlike monolayer formation kinetics with saturation on a time scale of order 100 s. This new type of INPS measurement provides a powerful tool to measure and optimize dye impregnation kinetics of DSCs and, from a more general point of view, offers a generic experimental platform to measure adsorption/desorption and diffusion phenomena in solid and mesoporous systems and at internal hidden interfaces. KEYWORDS: Dye-sensitized solar cell, mesoporous titanium dioxide, localized surface plasmon resonance, indirect nanoplasmonic sensing, hidden interface T he properties and organization of dye molecules on the surfaces of the mesoporous TiO 2 anode of dye sensitized solar cells (DSCs) play a crucial role for the function of the DSC. Relevant properties are the density of adsorbed dye molecules, their mode of anchoring to the surface, and their electronic interaction with the TiO 2 surface that determines interfacial electron-transfer rates. It is therefore very important to examine, understand, and control the surface self-assembly process of the dye molecules, including the kinetics at the TiO 2 interface, as recently demonstrated by a 20% increase in cell performance after consecutive dye adsorption cycles. 1 Details regarding to what extent optimized dye adsorption may boost performance are to date poorly understood and are the main motivation for developing the experimental methods presented here.The mesoporous anode structure of a DSC is impregnated with dye molecules by exposing the anode to a dye solution. Mechanistically, the process is a combination of pore diffusion and adsorption/desorption events. The impregnation begins in the top layer of the porous structure, and then as time elapses, it reaches deeper and deeper, until eventually the whole sample is saturated with dye molecules. The interplay of the diffusion/ adsorption/desorption processes determines ...

Nanocomposite layers of Ag nanoparticles and a-Si:H film constitute attractive candidates for the realization of ultrathin "two-dimensional" plasmonic solar cells, with an ideal 18% efficiency predicted for an average layer thickness of only 20 nm. By combining optical spectroscopy with photoconductivity measurements, we here characterize different contributions to the light absorption and charge carrier generation in such nanocomposites. We focus in particular on the important role of the absorber layer thickness for these processes, by studying a range of a-Si:H thicknesses from 9 to 67 nm. Through detailed comparison with numerical calculations by the finite element method, observed experimental features are connected to specific resonance modes and charge carrier generation mechanisms. The influence of dipolar and quadrupolar near-field distributions are evaluated with respect to different figures of merit for plasmonic solar cells. We briefly discuss how the present findings may be implemented in practical solar cell configurations.

A dye molecule monolayer formed on a TiO 2 surface is a key component in dye-sensitized solar cells. It is usually formed by adsorbing dye molecules from a solution. The dye layer should absorb as much solar light as possible and convert the light to photoelectrons, which are injected into the TiO 2 conduction band. For that purpose the dye molecules should adsorb on TiO 2 with appropriate molecular orientation and close packing. We measured adsorption and desorption kinetics of dye Z907 on thin compact TiO 2 films in real time using indirect nanoplasmonic sensing. From kinetic curves, we derived adsorption and desorption rate constants in a direct way, which has not been done for such systems previously. We then derived the equilibrium adsorption constant from both kinetics (by the ratio of the adsorption and desorption rate constants) and from a measured Langmuir isotherm obtained experimentally using the same method, the same sample, and the same experiment. The two values are in reasonably good agreement considering possible error sources; our approach thus constitutes an effective method of determining more reliable equilibrium constants for dye−TiO 2 systems. Furthermore, by measuring a series of intermittent adsorption−desorption steps, we found successively less desorption at a given coverage after each rinsing step and conclude that there are different binding states and that reorganization of the dye molecules on the TiO 2 surface occurs over rather long time scales. The rearrangement process seems to accelerate by intermittent rinsing and associated desorption of loosely bound molecules. The results suggest that the detailed conditions for the dye impregnation kinetics can be used to optimize the dye layer.