376
Z. Wang and B. Luk’yanchuk
In the following, fundamentals about optical super-resolution in microspheres and
metamaterial particle-lenses will be firstly presented. The key progresses will then
be reviewed, followed by the outlook of the field.
15.2 Super-Resolution Mechanisms: Photonic Nanojet,
Super-Resonance and Evanescent Jets
The super-resolution mechanisms for microsphere and metamaterial particle-lens
are fundamentally different; they are thus presented separately below. Briefly speaking, for microspheres, the super-resolution is a combined result of several superresolution effects, including photonic nanojet effect (weak super-resolution, nonresonant), super-resonance effect (strong super-resolution, resonant) and others
including substrate effect and partial and inclined illumination effect, etc. For the
second-generation metamaterial lens, strong super-resolution is a result of a new
nanophotonic effect: nanocomposite media made of high refractive index nanoparticles can effectively convert evanescent waves into propagating waves, and vice
versa.
15.2.1 Microsphere and Microcylinder
Optical resolution can be measured and calibrated by different means. In fluorescent imaging, ‘point’ source objects like fluorescent molecules are readily available.
The system resolution can be precisely determined by measuring the point source
function of the sources. In contrast, ‘point’ objects are rarely available in label-free
imaging. To circumvent this issue, researchers often use finite size nanostructures and
the resolution was measured based on observability of minimal discernible feature
sizes which can be represented by the widths of the stripes, diameters of nanopores,
edge-to-edge separations in dimers and clusters, etc. This has resulted in the resolution claims spanning the range from λ/6 to λ/17 [31]. However, such claims were
questioned and the resolution should be calibrated. A calibration procedure was suggested by Allen et al. by convolution imaging object with a two-dimension PSF
and compare with experimental results. The calibrated resolution for microsphere
imaging is about ~λ/6–λ/8 [20, 27, 32], instead of λ/17 as claimed by Lin [31]. It is
important to note that a λ/7 resolution optical system can resolve λ/25 features, e.g.
15 nm gap in a bowtie sample as in [20]. In the following, we proceed to explain
why a λ/6–λ/8 resolution is possible in microsphere-assisted imaging.
Despite of simplicity, the underlying physical mechanism for microsphere imaging is quite complex. A complete theoretical model should consist of three elements:
focusing of the incident light by microsphere, interaction of the incident light with
the sample, and imaging of the scattered light in virtual mode. These processes are
Z. Wang and B. Luk’yanchuk
In the following, fundamentals about optical super-resolution in microspheres and
metamaterial particle-lenses will be firstly presented. The key progresses will then
be reviewed, followed by the outlook of the field.
15.2 Super-Resolution Mechanisms: Photonic Nanojet,
Super-Resonance and Evanescent Jets
The super-resolution mechanisms for microsphere and metamaterial particle-lens
are fundamentally different; they are thus presented separately below. Briefly speaking, for microspheres, the super-resolution is a combined result of several superresolution effects, including photonic nanojet effect (weak super-resolution, nonresonant), super-resonance effect (strong super-resolution, resonant) and others
including substrate effect and partial and inclined illumination effect, etc. For the
second-generation metamaterial lens, strong super-resolution is a result of a new
nanophotonic effect: nanocomposite media made of high refractive index nanoparticles can effectively convert evanescent waves into propagating waves, and vice
versa.
15.2.1 Microsphere and Microcylinder
Optical resolution can be measured and calibrated by different means. In fluorescent imaging, ‘point’ source objects like fluorescent molecules are readily available.
The system resolution can be precisely determined by measuring the point source
function of the sources. In contrast, ‘point’ objects are rarely available in label-free
imaging. To circumvent this issue, researchers often use finite size nanostructures and
the resolution was measured based on observability of minimal discernible feature
sizes which can be represented by the widths of the stripes, diameters of nanopores,
edge-to-edge separations in dimers and clusters, etc. This has resulted in the resolution claims spanning the range from λ/6 to λ/17 [31]. However, such claims were
questioned and the resolution should be calibrated. A calibration procedure was suggested by Allen et al. by convolution imaging object with a two-dimension PSF
and compare with experimental results. The calibrated resolution for microsphere
imaging is about ~λ/6–λ/8 [20, 27, 32], instead of λ/17 as claimed by Lin [31]. It is
important to note that a λ/7 resolution optical system can resolve λ/25 features, e.g.
15 nm gap in a bowtie sample as in [20]. In the following, we proceed to explain
why a λ/6–λ/8 resolution is possible in microsphere-assisted imaging.
Despite of simplicity, the underlying physical mechanism for microsphere imaging is quite complex. A complete theoretical model should consist of three elements:
focusing of the incident light by microsphere, interaction of the incident light with
the sample, and imaging of the scattered light in virtual mode. These processes are