Nowadays, modern science proposed and optimizes new materials and technologies, whose characteristics and performances are governed by many factors. However, the scientific community rarely adopts multivariate strategies for the comprehension of what is proposed. As a striking example, a standard Dye-sensitized Solar Cell (DSC) is a typical complex system assembled with different and heterogeneous layers (FTO/nanocrystalline semiconductor/sensitizer/electrolyte/Pt-FTO), each one affected by intrinsic variability; moreover the layers influence each other1 and this increases the number of variables involved at the same time in the photoconversion process. Despite the continuous advances of the last years,2 DSCs are not yet commercialized on large-scale, because they are still subjected to undesirable phenomena, i.e. photodegradation of the dye anchored on semiconductor (especially with metal-free sensitizers3), leakage of the electrolyte, diffusion of pollutants from the outside and corrosion of some components. The idea of the present work started from the need to identify all the components by which the photoconversion processes may be influenced. In order to obtain a significant improvement of photovoltaic performances, particularly in reproducibility, long-term stability and efficiency, each stage of the production process of the cell should be investigated. However, this work cannot be conducted with the standard univariate approach, in fact, ranging one variable at a time, it results very difficult to understand and predict the possible synergistic or antagonistic effects (very common in complex systems) due to the interactions between the variables themselves. Therefore, this research was performed with a multivariate approach, exploiting a quantitative chemometric approach (Design of Experiments), because it allows maximizing the information content with the least number of experiments. The encouraging results obtained before in the formulation of novel electrolytes (gel, polymer)4 and on ZnO-based DSC5 allowed us to exploit the potentiality of this approach also to the dye loading process6 and finally to the whole standard TiO2-based DSC. We are firmly convinced that this approach will make possible to find the optimal experimental conditions to achieve, within a good reproducibility, optimized performances, both in term of efficiency and long term stability.
Design of Experiments (DoE): a multivariate approach to DSC optimization
BAROLO, CLAUDIA;GALLIANO, SIMONE;VISCARDI, Guido
2014-01-01
Abstract
Nowadays, modern science proposed and optimizes new materials and technologies, whose characteristics and performances are governed by many factors. However, the scientific community rarely adopts multivariate strategies for the comprehension of what is proposed. As a striking example, a standard Dye-sensitized Solar Cell (DSC) is a typical complex system assembled with different and heterogeneous layers (FTO/nanocrystalline semiconductor/sensitizer/electrolyte/Pt-FTO), each one affected by intrinsic variability; moreover the layers influence each other1 and this increases the number of variables involved at the same time in the photoconversion process. Despite the continuous advances of the last years,2 DSCs are not yet commercialized on large-scale, because they are still subjected to undesirable phenomena, i.e. photodegradation of the dye anchored on semiconductor (especially with metal-free sensitizers3), leakage of the electrolyte, diffusion of pollutants from the outside and corrosion of some components. The idea of the present work started from the need to identify all the components by which the photoconversion processes may be influenced. In order to obtain a significant improvement of photovoltaic performances, particularly in reproducibility, long-term stability and efficiency, each stage of the production process of the cell should be investigated. However, this work cannot be conducted with the standard univariate approach, in fact, ranging one variable at a time, it results very difficult to understand and predict the possible synergistic or antagonistic effects (very common in complex systems) due to the interactions between the variables themselves. Therefore, this research was performed with a multivariate approach, exploiting a quantitative chemometric approach (Design of Experiments), because it allows maximizing the information content with the least number of experiments. The encouraging results obtained before in the formulation of novel electrolytes (gel, polymer)4 and on ZnO-based DSC5 allowed us to exploit the potentiality of this approach also to the dye loading process6 and finally to the whole standard TiO2-based DSC. We are firmly convinced that this approach will make possible to find the optimal experimental conditions to achieve, within a good reproducibility, optimized performances, both in term of efficiency and long term stability.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
