372
Z. Wang and B. Luk’yanchuk
The resolution of optical microscope was first described by German physicist
Ernst Abbe in 1873 [1]: The minimum distance, d’, between two structural elements
to be imaged as two objects instead of one, is given by d’ = Kλ/NA = 0.5λ/NA
(K = 0.5), where λ is the wavelength of light and NA the numerical aperture of the
objective lens. Other ways of resolution definition, such as Sparrow [2], Houston
[3] or Rayleigh [4] criteria, with K = 0.473, 0.515, or 0.61 respectively, has since
been developed in the history. The underlying physics of resolution limit is optical
diffraction and loss of evanescent wave components when light travels into far-field
from objects; Subwavelength spatial information of an object was carried by the
high-frequency evanescent waves which decay exponentially with distance from the
object. For a white light microscope, the resolution limit is about 200–250 nm. The
Abbe resolution limit was considered the fundamental limit of optical microscope
resolution for a century.
The emerge of near-field optics breaks the resolution limit. In 1984, GermanSwiss physicist Dieter Pohl and colleagues invented Near-field Scanning Optical
Microscopy (NSOM or SNOM), the first optical instrument that provided optical
resolution far beyond Abbe’s limit, e.g. 20 nm at wavelength 515 nm [5]. Here,
a super-resolution image of a structure is constructed by scanning a tiny tip with
nano-sized aperture in the proximity (~tens nanometres) of an illuminated specimen.
From late 1990s, stimulated by the rapid advancements of plasmonics, nanophotonics and metamaterials, new super-resolution microscopy/nanoscopy techniques
have emerged, including metal-based metamaterial superlens [6], STED (stimulated
emission depletion microscopy) [7], optical superoscillatory lens [8], SIM (structured
illumination microscopy) [9] and more [10].
In 2000, British scientist John Pendry proposed the intriguing ‘metamaterial superlens’ (also known as Pendry superlens) concept, which uses a slab of NIM (negativeindex medium) as superlens to enhance and transfer the evanescent waves for perfect
imaging [6]. The fascinating ‘perfect lens’ proposal sparked a real surge in metamaterials and plasmonics research. Based on Pendry’s recipe, several variations of
metal-based metamaterial superlenses were developed and demonstrated by groups
across the world [11–14]. Among them the hyperlens [12, 13] received most attention. Hyperlens is engineered such that the evanescent waves of objects are converted
into propagating waves forming a magnified image of the sample on a distant screen.
It is plausible to think it is a far-field imaging device. However, projection to a distant
screen does not alert the fact that the hyperlens relies on the sample’s near-field to
achieve super-resolution. Therefore, it is still a near-field imaging device. Due to
intrinsic losses in metals and nanofabrication challenges, superlenses constructed
with metal elements show a limited resolution about 70 ~ 100 nm at visible frequencies and being seldom used in biomedical imaging.
Another major route to super-resolution is fluorescent microscopy, which seeks
super-resolution from the labelling of samples rather than the lenses. The superresolution fluorescence microscopy techniques can be divided into two categories:
spatially patterned excitation (STED, SSIM, RESOLFTs) [15] and single-molecule
localization (STORM, PALM, FPALM) [15] of fluorescence molecules. The techniques have been widely used in biological sciences, like microbiology, cell biology
Z. Wang and B. Luk’yanchuk
The resolution of optical microscope was first described by German physicist
Ernst Abbe in 1873 [1]: The minimum distance, d’, between two structural elements
to be imaged as two objects instead of one, is given by d’ = Kλ/NA = 0.5λ/NA
(K = 0.5), where λ is the wavelength of light and NA the numerical aperture of the
objective lens. Other ways of resolution definition, such as Sparrow [2], Houston
[3] or Rayleigh [4] criteria, with K = 0.473, 0.515, or 0.61 respectively, has since
been developed in the history. The underlying physics of resolution limit is optical
diffraction and loss of evanescent wave components when light travels into far-field
from objects; Subwavelength spatial information of an object was carried by the
high-frequency evanescent waves which decay exponentially with distance from the
object. For a white light microscope, the resolution limit is about 200–250 nm. The
Abbe resolution limit was considered the fundamental limit of optical microscope
resolution for a century.
The emerge of near-field optics breaks the resolution limit. In 1984, GermanSwiss physicist Dieter Pohl and colleagues invented Near-field Scanning Optical
Microscopy (NSOM or SNOM), the first optical instrument that provided optical
resolution far beyond Abbe’s limit, e.g. 20 nm at wavelength 515 nm [5]. Here,
a super-resolution image of a structure is constructed by scanning a tiny tip with
nano-sized aperture in the proximity (~tens nanometres) of an illuminated specimen.
From late 1990s, stimulated by the rapid advancements of plasmonics, nanophotonics and metamaterials, new super-resolution microscopy/nanoscopy techniques
have emerged, including metal-based metamaterial superlens [6], STED (stimulated
emission depletion microscopy) [7], optical superoscillatory lens [8], SIM (structured
illumination microscopy) [9] and more [10].
In 2000, British scientist John Pendry proposed the intriguing ‘metamaterial superlens’ (also known as Pendry superlens) concept, which uses a slab of NIM (negativeindex medium) as superlens to enhance and transfer the evanescent waves for perfect
imaging [6]. The fascinating ‘perfect lens’ proposal sparked a real surge in metamaterials and plasmonics research. Based on Pendry’s recipe, several variations of
metal-based metamaterial superlenses were developed and demonstrated by groups
across the world [11–14]. Among them the hyperlens [12, 13] received most attention. Hyperlens is engineered such that the evanescent waves of objects are converted
into propagating waves forming a magnified image of the sample on a distant screen.
It is plausible to think it is a far-field imaging device. However, projection to a distant
screen does not alert the fact that the hyperlens relies on the sample’s near-field to
achieve super-resolution. Therefore, it is still a near-field imaging device. Due to
intrinsic losses in metals and nanofabrication challenges, superlenses constructed
with metal elements show a limited resolution about 70 ~ 100 nm at visible frequencies and being seldom used in biomedical imaging.
Another major route to super-resolution is fluorescent microscopy, which seeks
super-resolution from the labelling of samples rather than the lenses. The superresolution fluorescence microscopy techniques can be divided into two categories:
spatially patterned excitation (STED, SSIM, RESOLFTs) [15] and single-molecule
localization (STORM, PALM, FPALM) [15] of fluorescence molecules. The techniques have been widely used in biological sciences, like microbiology, cell biology