256
T. C. Jagadale and S.-W. Chu
Second, different from incoherent fluorescent molecules, plasmonic structures
exhibit coherent coupling when two or more metallic nanoparticles are adjacent to
each other. As we can see in Fig. 10.7b, the arrowheads mark particles that aggregate
together, which change the resonant wavelength. Therefore, although these aggregates still exhibit strong scattering intensity in the 1f m image, their 2f m and 3f m
signals are relatively weaker compared to uncoupled nanoparticles. That is, SAX
may be used to detect plasmonic coupling beyond diffraction limit.
Third, most of the current super-resolution techniques are not able to enhance
resolution in tissues. For example, wide-field-based localisation techniques do not
provide optical sectioning capability; thus lacking axial contrast. STED and SIM
require spatial beam engineering, and thus are susceptible to beam distortion due to
scattering/aberration. On the contrary, SAX relies on temporal modulation that is less
affected when penetrating into tissues, and therefore should be the best candidate for
deep-tissue resolution enhancement. This has been realized recently with plasmonic
SAX, which provides three-fold resolution enhancement by 3f m signals throughout
the whole working distance of an objective, that is 200 µm, providing inspirational
possibility towards deep-tissue super-resolution imaging [42].
10.4.3 Super-Resolution Based on Optical Suppression
of Scattering Imaging (SUSI)
In Sect. 10.2.2, we have explained how STED improves spatial resolution via alloptical switch of fluorescence emission and a donut STED beam. In Sect. 10.3.3,
we presented optically switchable scattering from plasmonic nanoparticles. Here we
combine the two concepts to achieve super-resolution imaging based on suppression
of scattering imaging (SUSI), as demonstrated in Fig. 10.8. Similar to the condition
in Fig. 10.5, two lasers at 543 and 592 nm are aligned together.
In Fig. 10.8a, the top row shows a laser scanning image based on a single solid
543 nm beam (see inset), and the bottom row gives the PSF intensity profile for the
nanoparticles in the centre of imaging area, with FWHM of 180 nm. Apparently, the
resolution is inadequate to resolve whether there are two particles.
In Fig. 10.8b, the inset shows that a 592 nm beam is converted into a donut shape
by a vortex phase plate, and the central hole is overlapped with the solid 543 nm
excitation, similar to STED implementation. As mentioned in Fig. 10.5, when the
592 nm laser intensity increases to 10
6 W/cm
2 (about 100 µW in power), scattering
of 543 nm is significantly suppressed, leading to remarkable enhancement of spatial
resolution in Fig. 10.8b. Quantitatively, the averaged FWHM of a single particle is
120 ± 4 nm at this intensity, agreeing well with theoretical prediction, and it is now
possible to distinguish two particles in the centre of the image.
To further enhance spatial resolution, deconvolution is adopted in Fig. 10.8c,
which achieves resolution of 60 nm (i.e. λ/9). It is important to notice that not only
the central region exhibits enhanced resolution but all nanoparticles in the imaging
T. C. Jagadale and S.-W. Chu
Second, different from incoherent fluorescent molecules, plasmonic structures
exhibit coherent coupling when two or more metallic nanoparticles are adjacent to
each other. As we can see in Fig. 10.7b, the arrowheads mark particles that aggregate
together, which change the resonant wavelength. Therefore, although these aggregates still exhibit strong scattering intensity in the 1f m image, their 2f m and 3f m
signals are relatively weaker compared to uncoupled nanoparticles. That is, SAX
may be used to detect plasmonic coupling beyond diffraction limit.
Third, most of the current super-resolution techniques are not able to enhance
resolution in tissues. For example, wide-field-based localisation techniques do not
provide optical sectioning capability; thus lacking axial contrast. STED and SIM
require spatial beam engineering, and thus are susceptible to beam distortion due to
scattering/aberration. On the contrary, SAX relies on temporal modulation that is less
affected when penetrating into tissues, and therefore should be the best candidate for
deep-tissue resolution enhancement. This has been realized recently with plasmonic
SAX, which provides three-fold resolution enhancement by 3f m signals throughout
the whole working distance of an objective, that is 200 µm, providing inspirational
possibility towards deep-tissue super-resolution imaging [42].
10.4.3 Super-Resolution Based on Optical Suppression
of Scattering Imaging (SUSI)
In Sect. 10.2.2, we have explained how STED improves spatial resolution via alloptical switch of fluorescence emission and a donut STED beam. In Sect. 10.3.3,
we presented optically switchable scattering from plasmonic nanoparticles. Here we
combine the two concepts to achieve super-resolution imaging based on suppression
of scattering imaging (SUSI), as demonstrated in Fig. 10.8. Similar to the condition
in Fig. 10.5, two lasers at 543 and 592 nm are aligned together.
In Fig. 10.8a, the top row shows a laser scanning image based on a single solid
543 nm beam (see inset), and the bottom row gives the PSF intensity profile for the
nanoparticles in the centre of imaging area, with FWHM of 180 nm. Apparently, the
resolution is inadequate to resolve whether there are two particles.
In Fig. 10.8b, the inset shows that a 592 nm beam is converted into a donut shape
by a vortex phase plate, and the central hole is overlapped with the solid 543 nm
excitation, similar to STED implementation. As mentioned in Fig. 10.5, when the
592 nm laser intensity increases to 10
6 W/cm
2 (about 100 µW in power), scattering
of 543 nm is significantly suppressed, leading to remarkable enhancement of spatial
resolution in Fig. 10.8b. Quantitatively, the averaged FWHM of a single particle is
120 ± 4 nm at this intensity, agreeing well with theoretical prediction, and it is now
possible to distinguish two particles in the centre of the image.
To further enhance spatial resolution, deconvolution is adopted in Fig. 10.8c,
which achieves resolution of 60 nm (i.e. λ/9). It is important to notice that not only
the central region exhibits enhanced resolution but all nanoparticles in the imaging