408
A. V. Maslov and V. N. Astratov
16.1 Introduction
16.1.1 Experimental Results
Since its invention at the end of the sixteenth century, the optical microscope has
been continuously improved. At the end of the nineteenth century, several prominent scientists, including Abbe, Helmholtz, and Rayleigh, addressed the question of
defining the microscope resolution and finding its ultimate limit. The standard formula for the resolution value δ used nowadays states δ = λ/(2n sin α), where λ is the
operating wavelength in vacuum, n is the refractive index of the immersion medium,
and α is the semi-aperture of the objective. In Abbe’s approach, the resolution δ is
the minimal resolvable period of a grating under oblique coherent illumination. Similar expression was obtained by Helmholtz for two self-luminous incoherent points
which are more relevant to true imaging [1]. This ultimate resolution corresponds to a
finite width of the image of a point source. The finite width appears from the current
emission property: the current components with spatial frequency higher than the
wavenumber in the surrounding medium cannot emit.
To push the resolution performance beyond λ/2, various new imaging techniques
have been proposed and successfully implemented. The last decade of the twentieth
century brought super-resolved fluorescence (FL) imaging represented in particular
by stimulated emission depletion (STED) [2, 3] and localization microscopies [4, 5],
which overperformed the conventional microscopy techniques. Due to extraordinary resolution beyond 20 nm and ability to highlight substructures of interest, these
methods pushed the limits of microscopy and resulted in the Nobel Prize in Chemistry awarded in 2014 to the pioneers of these methods. Despite all successes of
super-resolved FL microscopy, it has certain drawbacks such as damaging of living
biomedical samples by staining them with dyes and photobleaching.
Label-free imaging is, therefore, a preferred method for many applications. However, the arsenal of methods which can be used for increasing the resolution is narrower in this case. In principle, it can be divided into methods which are based on
the same far-field linear optics principles, which underlie the classical diffraction
limit, and methods involving different physical principles. The examples of the first
approach (far-field principles) are based on using solid immersion lenses (SIL) developed [6] in the early 1990s and structured illumination microscopy (SIM) developed
[7–11] in the early 2000s. Examples of the second approach include the methods
which are based on using the near fields as in near-field scanning optical microscopy
(NSOM) [12], nonlinear optics [13–15], and nanostructures with engineered dispersion properties such as nanoplasmonic arrays, metamaterials and metasurfaces.
Nanostructure-based optical elements include surface plasmon-polariton lens [16],
periodic metallo–dielectric structures [17], far-field superlens [18, 19], hyperlens
[20–23], Eaton lens [24], and adiabatic structures [25]. It should be noted, however,
that although the nanoplasmonic and metamaterials-based lenses represent exciting
new concepts, their deployment faces such barriers as the narrow spectral range of
operation, ohmic losses, and challenging fabrication.
A. V. Maslov and V. N. Astratov
16.1 Introduction
16.1.1 Experimental Results
Since its invention at the end of the sixteenth century, the optical microscope has
been continuously improved. At the end of the nineteenth century, several prominent scientists, including Abbe, Helmholtz, and Rayleigh, addressed the question of
defining the microscope resolution and finding its ultimate limit. The standard formula for the resolution value δ used nowadays states δ = λ/(2n sin α), where λ is the
operating wavelength in vacuum, n is the refractive index of the immersion medium,
and α is the semi-aperture of the objective. In Abbe’s approach, the resolution δ is
the minimal resolvable period of a grating under oblique coherent illumination. Similar expression was obtained by Helmholtz for two self-luminous incoherent points
which are more relevant to true imaging [1]. This ultimate resolution corresponds to a
finite width of the image of a point source. The finite width appears from the current
emission property: the current components with spatial frequency higher than the
wavenumber in the surrounding medium cannot emit.
To push the resolution performance beyond λ/2, various new imaging techniques
have been proposed and successfully implemented. The last decade of the twentieth
century brought super-resolved fluorescence (FL) imaging represented in particular
by stimulated emission depletion (STED) [2, 3] and localization microscopies [4, 5],
which overperformed the conventional microscopy techniques. Due to extraordinary resolution beyond 20 nm and ability to highlight substructures of interest, these
methods pushed the limits of microscopy and resulted in the Nobel Prize in Chemistry awarded in 2014 to the pioneers of these methods. Despite all successes of
super-resolved FL microscopy, it has certain drawbacks such as damaging of living
biomedical samples by staining them with dyes and photobleaching.
Label-free imaging is, therefore, a preferred method for many applications. However, the arsenal of methods which can be used for increasing the resolution is narrower in this case. In principle, it can be divided into methods which are based on
the same far-field linear optics principles, which underlie the classical diffraction
limit, and methods involving different physical principles. The examples of the first
approach (far-field principles) are based on using solid immersion lenses (SIL) developed [6] in the early 1990s and structured illumination microscopy (SIM) developed
[7–11] in the early 2000s. Examples of the second approach include the methods
which are based on using the near fields as in near-field scanning optical microscopy
(NSOM) [12], nonlinear optics [13–15], and nanostructures with engineered dispersion properties such as nanoplasmonic arrays, metamaterials and metasurfaces.
Nanostructure-based optical elements include surface plasmon-polariton lens [16],
periodic metallo–dielectric structures [17], far-field superlens [18, 19], hyperlens
[20–23], Eaton lens [24], and adiabatic structures [25]. It should be noted, however,
that although the nanoplasmonic and metamaterials-based lenses represent exciting
new concepts, their deployment faces such barriers as the narrow spectral range of
operation, ohmic losses, and challenging fabrication.