Anne Myers Kelley, PhD , University of California (Merced)
This project was inspired by recent reports that the power conversion efficiencies of solar photovoltaic devices can be improved by incorporating plasmonically active metal nanostructures into the device. For bulk heterojunction devices based on polymer/fullerene blends, power conversion efficiency enhancements of 15-70% have been reported. While observed enhancements could have a number of causes, including indirect effects such as changes in the morphology of the material, it appears likely that at least some of the reported effects have their origin in the electromagnetic properties specific to plasmonically active metals.
Local enhancement of electromagnetic fields by plasmonically active nanostructures manifests itself in a number of other ways including the phenomenon of surface enhanced Raman scattering (SERS). Standard electromagnetic SERS theory holds that the Raman scattering enhancement is simply the product of the enhancements of the local electromagnetic field intensities at the incident and Raman scattered wavelengths. Therefore, the magnitude of the SERS enhancement should provide a reasonable measure of the electromagnetic field enhancement, and the SERS experiment can be carried out in an actual device structure or closely related geometries. In addition, any structural changes in the molecules nearest to the metal surface are manifested as changes in the positions and/or relative intensities of the peaks in the SERS spectrum compared to the unenhanced Raman spectrum. This structural sensitivity allows for examination of chemically specific effects of metal nanoparticles on the organic materials which would also be expected to affect the performance of photovoltaic cells.
During the previous funding period we performed a SERS study of poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS), a polymer blend often used as a hole collecting contact in organic solar cells. Many of the device geometries reported to display power conversion efficiency enhancement have the metal nanoparticles in direct contact with PEDOT:PSS. In the oxidized form used, PEDOT:PSS is essentially transparent in the visible and increased light absorption should not be an available plasmonic enhancement mechanism. We observed SERS enhancement factors of about 2-30 depending on metal, excitation wavelength, and nanoparticle geometry. We also observed significant changes in the spectrum of the PEDOT component in the presence of metals, particularly Ag. These were interpreted to show that at low light intensities, silver nanoparticles promote the re-oxidation of chemically reduced PEDOT and may therefore protect against the formation of nonconducting dedoped regions of PEDOT:PSS coated ITO. At higher light intensities there was evidence for silver-induced photooxidation of PEDOT by addition of oxygen to the thiophene rings, disrupting the pi-conjugation.
During the past year we extended the SERS technique to films of poly(3-hexylthiophene)/[6,6]-phenyl C61-butyric acid methyl ester (P3HT/PCBM), one of the best studied polymer blends used in bulk heterojunction organic polymer solar cells. We focused on samples in which the polymer blend is deposited over triangular silver nanoprisms. These nanoprisms have many desirable properties including large electromagnetic field enhancements particularly at the corners, broad tunability of their plasmon resonances across the visible spectrum, and self-assembly without aggregation onto silanized glass surfaces. Furthermore, David Ginger’s group at the University of Washington has direct spectroscopic evidence for enhanced charge-carrier generation in this system. We used the SERS enhancements as a measure of the extent of electromagnetic field enhancement caused by the nanoparticles, and also to examine any changes in the Raman spectral characteristics that might indicate effects of the metal on the chemistry or morphology of the nearby organic material.
Resonance Raman spectra were obtained at excitation wavelengths ranging from 364 to 633 nm, both in the presence and absence of triangular Ag nanoprisms with varying plasmon resonance frequencies. For ~35 nm polymer films deposited over nanoprisms, the nanoprisms enhance the sample-averaged Raman scattering intensities by factors of 2 to 20, depending on wavelength and nanoprism density. The weak blend fluorescence is enhanced by approximately the same factor as the Raman scattering, implying negligible excited-state quenching by the metal. The Raman peak positions and relative intensities are unaffected by the presence of either the triangular nanoprisms or spherical Au or Ag nanoparticles, indicating negligible morphological or chemical changes to the P3HT in contrast to our previous observations on PEDOT:PSS. The magnitudes of the Raman enhancements are qualitatively consistent with the Ginger group’s observed enhancements in charge-carrier (positive polaron) yields for P3HT/PCBM deposited over silver nanoprisms.
PRF-ND funding has thus far resulted in one peer-reviewed publication (in J. Phys. Chem.) with a second manuscript currently under review by J. Phys. Chem and a third likely to be submitted during the fall. I have presented this work at several conferences, one of them leading to a collaboration with the group of David Ginger at the University of Washington. He and two of his postdocs are co-authors on our submitted paper. My graduate student Marina Stavytska-Barba has been supported in part as a research assistant on this grant. She is currently on maternity leave but expects to defend her Ph.D. dissertation during the fall.