Evidence of the Discrepancy between the Near-Field and the Far-Field using SERS

Nathalie Lidgi-Guigui Santé Médecine et Biologie Humaine, University Paris 13 -Sorbone Paris Cité, Bobigny, France Maximilien Cottat Santé Médecine et Biologie Humaine, University Paris 13 -Sorbone Paris Cité, Bobigny, France Florent Colas Santé Médecine et Biologie Humaine, University Paris 13 -Sorbone Paris Cité, Bobigny, France Nicolas Guillot Santé Médecine et Biologie Humaine, University Paris 13 -Sorbone Paris Cité, Bobigny, France Amina Belhadj Santé Médecine et Biologie Humaine, University Paris 13 -Sorbone Paris Cité, Bobigny, France Raymond Gilibert Santé Médecine et Biologie Humaine, University Paris 13 -Sorbone Paris Cité, Bobigny, France Andrea Toma -, Istituto Italiano di Tecnologia (IIT), Genova, Italy Enzo Di Fabrizio -, Istituto Italiano di Tecnologia (IIT), Genova, Italy Pietro Gucciardi Istituto per i Processi Chimico-Fisici, CNR-IPCF, Messina, Italy Marc Lamy de la Chapelle Santé Médecine et Biologie Humaine, University Paris 13 -Sorbone Paris Cité, Bobigny, France

Surface Enhanced Raman Scattering (SERS) is based on two main effects: the chemical and the electromagnetic. The enhancement factor from chemical effect is close to 102 whereas it can reach up to 108 for the electromagnetic one. As the electromagnetic effect is due to the strong enhanced electromagnetic field created by the Localized Surface Plasmon Resonance (LSPR) close to metallic nanostructures, it is usually defined as the near-field enhancement, whereas the plasmon resonance determined using the extinction spectrum can be defined as a far-field measurement.

Several studies have claimed that the best enhancement factor is reached for a LSPR position at the average between the excitation and the Raman wavelength [1, 2]. But it has also been demonstrated that such optimization rules depends strongly on the nanoparticle shape and on the excitation wavelength. In this work, we are studying the possibility of predicting the best enhancement factor for a given excitation wavelength. For three excitation wavelengths (633, 660 and 785 nm), we measured the SERS signal depending on the LSPR position for gold nanocylinders made by electron beam lithography. We demonstrate that the rule based on the average between the excitation and the Raman wavelength cannot be applied to all excitation wavelengths in the visible range. We interpret that this failure of the rule is due to a discrepancy between the far field and the near field. Based on experimental LSPR measurements, we applied the model of harmonic oscillators damping proposed by Zuloaga [3] to explain these results. However, it cannot completely explain the shift that is observed between the LSPR and SERS and it need to provide more accuracy on the near-field behavior to explain this near-field/far-field discrepancy. Nevertheless, by fitting our data, we are able to determine a law which can predict the best LSPR position for a given excitation wavelength, for highest wavelength in the visible range.

Acknowledgments

This work was supported by the Nanoantenna European project (FP7-HEALTH-F5-2009 241818) and the PIRANEX ANR project (ANR-12-NANO-0016).

[1] N. Felidj, J. Aubard, G. Levi, JR. Krenn, A. Hohenau, G. Schider, A. Leitner, FR. Aussenegg, Applied Physics Letters, 2003, 82(18),3095-3097

[2] C.L. Haynes, R.P. Van Duyne, J. Phys. Chem. B, 2003, 107, 7426-7433

[3] J. Zuloaga, P. Nordlander, Nano Letter, 2011, 11(3), 1280-1283

nathalie.lidgi-guigui@univ-paris13.fr









Powered by Eventact EMS