420
A. Pors et al.
the response of the fabricated arrays. Individual and double gold nanorods being
∼50 × 50 nm 2 in the cross section with the lengths of ∼90 nm and ∼125 nm and the
center-to-center distance of ∼130 nm were considered to be embedded in dielectric
with the refractive index of 1.45. Transmission spectra were then calculated using
their computed extinction cross sections and the corresponding size (400 × 750 nm 2 )
of unit cell (Fig. 11.9c). It is seen that the positions of local minima and maximum
in the transmission spectra correspond well to the experimental results (Fig. 11.9b),
albeit for a slightly shorter length of the long nanorod, but the contrast obtained in
simulations is much better than that achieved in the experiment. It should be borne in
mind that the use of divergent illumination and detection of focused radiation as well
as inhomogeneous broadening (due to fabrication-induced deviations of nanorod
dimensions in the fabricated arrays) could definitely contribute to smearing out the
spectral features in these preliminary experiments.
The fabricated DED array (Fig. 11.9a) covered with the PMMA layer was further characterized in the setup for leakage-radiation microscopy (LRM) [49] using a
continuous-wave-operated tunable Ti:sapphire laser, whose radiation polarized parallel to the nanorod axes was slightly focused and directed at normal incidence on
the sample (Fig. 11.10a) with the incident light powers being on the level of a few
milliwatts. The ± first diffraction orders observed in the back focal plane of an LRM
objective exhibited strong asymmetry (Fig. 11.10b) in their wavelength-dependent
behavior as expected [cf. Fig. 11.8c]. Wavelength dependencies of the corresponding
diffraction efficiencies measured with a calibrated photodetector resulted in a very
sharp peak of their ratio at the wavelength of ∼785 nm with the FWHM of ∼20 nm
(Fig. 11.10c), a value which is significantly smaller than the FWHM of plasmonic
resonances for individual nanorods (Fig. 11.9b). Such a drastic narrowing of the system response at resonance constitutes the main forte of the DED-based plasmonic
nanostructures. Since the sensitivity (nm/RIU) of plasmonic resonances is difficult to
appreciably increase, the FOM can radically be increased only by narrowing the resonance line width [37]. Note that the fabricated and characterized DED array was not
optimized with respect to the nanorod parameters and their separation, resulting in a
significantly smaller maximum ratio as compared to the model case (Fig. 11.8c). The
balance between the strengths, resonance positions, and spatial separation of individual nanorods (that should also be related to the array periodicity) is quite delicate:
one should achieve the same scattering strength of interacting DED scatterers simultaneously while fulfilling the optimum conditions [Eq. (11.12)] in order to ensure
the highest ratio between the diffraction efficiencies. Still, the effect of destructive
interference of DED pairs is rather robust as demonstrated with our proof-of-principle
experiments (Fig. 11.10b and c). It should be mentioned that we characterized also
similar arrays of DED pairs with a smaller center-to-center separation (∼100 nm)
and found a similarly sharp system response with the maximum ratio (between
the diffraction efficiencies) of ∼40 and the FWHM of ∼30 nm. Finally, assuming
that the sensitivity of the fabricated array is on the same level as that obtained in
our simulations, i.e., ∼600 nm/RIU, one obtains the FOM of ∼30, a value which
exceeds the record one obtained with multiscale patterned hole arrays in gold films
fabricated by soft interference lithography [42].
A. Pors et al.
the response of the fabricated arrays. Individual and double gold nanorods being
∼50 × 50 nm 2 in the cross section with the lengths of ∼90 nm and ∼125 nm and the
center-to-center distance of ∼130 nm were considered to be embedded in dielectric
with the refractive index of 1.45. Transmission spectra were then calculated using
their computed extinction cross sections and the corresponding size (400 × 750 nm 2 )
of unit cell (Fig. 11.9c). It is seen that the positions of local minima and maximum
in the transmission spectra correspond well to the experimental results (Fig. 11.9b),
albeit for a slightly shorter length of the long nanorod, but the contrast obtained in
simulations is much better than that achieved in the experiment. It should be borne in
mind that the use of divergent illumination and detection of focused radiation as well
as inhomogeneous broadening (due to fabrication-induced deviations of nanorod
dimensions in the fabricated arrays) could definitely contribute to smearing out the
spectral features in these preliminary experiments.
The fabricated DED array (Fig. 11.9a) covered with the PMMA layer was further characterized in the setup for leakage-radiation microscopy (LRM) [49] using a
continuous-wave-operated tunable Ti:sapphire laser, whose radiation polarized parallel to the nanorod axes was slightly focused and directed at normal incidence on
the sample (Fig. 11.10a) with the incident light powers being on the level of a few
milliwatts. The ± first diffraction orders observed in the back focal plane of an LRM
objective exhibited strong asymmetry (Fig. 11.10b) in their wavelength-dependent
behavior as expected [cf. Fig. 11.8c]. Wavelength dependencies of the corresponding
diffraction efficiencies measured with a calibrated photodetector resulted in a very
sharp peak of their ratio at the wavelength of ∼785 nm with the FWHM of ∼20 nm
(Fig. 11.10c), a value which is significantly smaller than the FWHM of plasmonic
resonances for individual nanorods (Fig. 11.9b). Such a drastic narrowing of the system response at resonance constitutes the main forte of the DED-based plasmonic
nanostructures. Since the sensitivity (nm/RIU) of plasmonic resonances is difficult to
appreciably increase, the FOM can radically be increased only by narrowing the resonance line width [37]. Note that the fabricated and characterized DED array was not
optimized with respect to the nanorod parameters and their separation, resulting in a
significantly smaller maximum ratio as compared to the model case (Fig. 11.8c). The
balance between the strengths, resonance positions, and spatial separation of individual nanorods (that should also be related to the array periodicity) is quite delicate:
one should achieve the same scattering strength of interacting DED scatterers simultaneously while fulfilling the optimum conditions [Eq. (11.12)] in order to ensure
the highest ratio between the diffraction efficiencies. Still, the effect of destructive
interference of DED pairs is rather robust as demonstrated with our proof-of-principle
experiments (Fig. 11.10b and c). It should be mentioned that we characterized also
similar arrays of DED pairs with a smaller center-to-center separation (∼100 nm)
and found a similarly sharp system response with the maximum ratio (between
the diffraction efficiencies) of ∼40 and the FWHM of ∼30 nm. Finally, assuming
that the sensitivity of the fabricated array is on the same level as that obtained in
our simulations, i.e., ∼600 nm/RIU, one obtains the FOM of ∼30, a value which
exceeds the record one obtained with multiscale patterned hole arrays in gold films
fabricated by soft interference lithography [42].