JP2010249739A - Hyper lens and optical microscope system using the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To acquire information on the wavelength dependency of an optical characteristic of a sample by using a hyper lens in a size smaller than the wavelength of incident light. <P>SOLUTION: Each hyper lens 1007 designed and manufactured in accordance with a plurality of light source wavelengths and each light source wavelength is arranged on a lens holder 1001 of a laser microscope. A pigment 1006 of an observation object is arranged on a sample holder 1005, and incident light 1008 from a light source 1003 is made to enter, and light passing the hyper lens 1007 is received by a microscope 307 and measured by a CCD camera 309. A plurality of sample images are acquired at each light source wavelength, and the plurality of sample images are superimposed by performing position alignment or intensity alignment. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a hyper lens and an optical microscope system using the same.

Optical lenses are used in many fields such as optical microscopes. The optical microscope includes an objective lens and an eyepiece, but the object to be observed is magnified by the objective lens to become a real image, and the real image is magnified by the eyepiece. The resolution of an optical lens is limited by the diffraction limit of light. The diffraction limit of light is a limit determined by the size of the minimum optical spot obtained by the optical lens. If the diffraction limit is Δ,
Δ = λ / 4NA (1)
Can be written. Here, λ is the wavelength of light, and NA is the numerical aperture. For the diffraction limit, for example, the resolution limit of an image obtained in an optical microscope is determined, and a high resolution optical microscope can be obtained by reducing the diffraction limit.

The reason why the diffraction limit is defined by the formula (1) is as follows. Considering a three-dimensional space, it is assumed that light propagates in the z direction and the direction of polarization is the x direction. In this case, when the polarization direction of light is directed to an arbitrary direction on the xy plane, it is only necessary to newly consider the direction as the x axis, and thus generality is not lost. The electric field E of light is written as
E∝exp (−ik ₀ z) (2)
Where exp is an exponential function, i is an imaginary number, k ₀ is a wave number, and k ₀ and wavelength λ are
k ₀ = 2π / λ (3)
It is tied in the relationship.

wave number k _z in the z-direction, and the wave number of the x direction and _{k x, k 0, k z} , the relationship of k _x is,
k ₀ ² = (ω / c) ² = k _x ² + k _z ² (4)
It is. Here, c is the speed of light and ω is the frequency. If k _x exceeds ω / c, k _z becomes an imaginary number, and as can be seen from the above equation (2), the amplitude of the light traveling in the z direction is rapidly attenuated. For example, in a microscope, between the size X and k _{x in} the x direction of the object to be observed,
X∝1 / k _x (5)
There is a relationship. For this reason, when k _x > ω / c, the light traveling in the z direction is rapidly attenuated. This means that the information of k _x > ω / c, that is, the information of X <λ is abrupt as the light travels in the z direction. Means lost. Hereinafter, a light component that rapidly attenuates satisfying k _x > k ₀ is referred to as a near-field component, and a component that is transmitted without attenuation of k _x <k ₀ is referred to as a propagation component.

  Since the diffraction limit is proportional to the wavelength, attempts to realize a high-resolution optical microscope by reducing the wavelength of the light used have been continued. Further shortening of the wavelength of the light source is very difficult mainly for the following two reasons. One is that it is difficult to realize a semiconductor laser that is a small light source, and the other is that there are many samples that absorb ultraviolet rays, so that the signal from the sample becomes small. Therefore, several methods for propagating light beyond the diffraction limit using another method have been proposed. Hereinafter, a solid immersion lens (SIL), near-field light, and a hyper lens will be described as typical examples.

The solid immersion lens (SIL) is a solid lens that is selected at a position very close to the object to be observed while selecting a material having a high refractive index and selecting a shape so as not to add aberration. A hemispherical shape is often used. Place the object on the flat side of the hemisphere. In a substance having a refractive index n, the substantial numerical aperture is increased n times, and the diffraction limit is reduced 1 / n times as can be seen from the equation (1). Since the lens and the observation object are very close, a near-field component that is normally lost can tunnel to the lens. The light entering the lens is reduced by n times the normal value. Since the normal lens is far from the observation object, the near-field component is lost before reaching the lens. As a material of SIL, high refractive index glass (n = 1.6 to 2.0), sapphire (n = 1.76), ZrO ₂ (n = 2.16), GaP (n = 3.1-3.4) etc. are used. For this reason, NA is 3.4 at the maximum. These are discussed in detail in Applied Physics Letter, Vol. 57, 2615-2616 (1990) (Non-Patent Document 1).

  Next, a method for exceeding the diffraction limit using near-field light will be described. The method using near-field light is roughly divided into two. One is an aperture type, and the other is a plasmon type. The aperture type includes, for example, a type using a fiber probe and a type using a cantilever. Examples of the plasmon type include Nanobeak, c-aperture, and metal probe. Here, plasmon type, especially Nanobeak, will be described.

  First, plasmons will be explained. Plasmons are generated by a resonance phenomenon. Free electrons in the metal are statistically subjected to a certain force by electron-lattice interaction or electron-electron interaction. In general, an object receiving a certain force has a natural frequency. When light is incident, electrons vibrate due to the electric field vibration of the light. When the frequency of light and the natural frequency of an electron match, the electron has a large vibration energy. This is called plasmon resonance. The behavior of this plasmon differs between the metal interior and the metal surface. This is because the surface electrons are subjected to a force different from that of the internal electrons. Plasmons excited on this surface are called surface plasmons. Theoretically, surface plasmons have been shown to be longitudinal waves, so surface plasmons do not radiate light, but propagate through the metal surface with near-field light, which is an electromagnetic field localized on the metal surface. The wave number, vibration frequency, propagation distance, near-field light intensity, etc. of this propagating wave are the conditions of the incident light, the dielectric constant of the metal / dielectric, the thickness of the metal film, and the size of the metal pattern where the plasmon is generated. It depends strongly on.

  One of the methods of exceeding the diffraction limit using plasmons is a method using a metal probe having a shape close to a triangle called Nanobeak. Details are discussed in Journal of Applied Physics, Vol. 95, 3901-3906 (2004) (Non-Patent Document 2). When polarized light having an electric field in a direction parallel to the perpendicular drawn from the apex of the triangle to the base is incident on this probe, the free electrons in the metal vibrate, but the free electron density near the apex increases. Strong near-field light that is very localized in the vicinity is generated. This near-field light has information beyond the diffraction limit.

  Next, the hyper lens will be described. A hyper lens is an optical lens that enables resolution exceeding the diffraction limit, but the near-field light having a size smaller than the wavelength of the incident light, which is formed on the sample by entering the sample into the sample, is reduced in size. It is a lens that propagates far away while zooming in. The hyper lens has a spherical shape, and is formed of a multilayer thin film of metal and dielectric laminated along the spherical surface.

  Normally, near-field light does not propagate far, but decays exponentially as it moves away from the sample surface, but here it propagates plasmons formed in the lens by a hyperlens consisting of a multilayer film of metal and dielectric. As a result, the information of the near-field light is propagated far away. At this time, since the near-field light propagates in the normal direction of the sphere from the center of the sphere due to the spherical shape of the lens, the near-field light having a high spatial frequency on the sample has a low spatial frequency on the lens sphere. Converted. For this reason, although the spatial frequency of the light propagated far becomes low, the intensity profile reflects the near-field intensity distribution on the sample. A near-field intensity distribution, that is, an image having a resolution exceeding the diffraction limit can be obtained.

Theoretically, a hyper lens is a lens in which the relationship between k _z and k _x is hyperbolic, and therefore the resolution is infinite in principle. Hereinafter, this principle will be briefly described. When the dielectric constant in the x direction is ε _x and the dielectric constant in the z direction is ε _z , the relationship between k _z and k _x is
^{(Ω / c) 2 = k} x 2 / ε z + k z 2 / ε x ... (6)
It is. Here, if ε _x , ε _z > 0, Expression (6) becomes an ellipse, and k _x has an upper limit. This means that it has a diffraction limit as described above. _If the signs of ε _x and ε _z are opposite, Equation (6) becomes a hyperbola, and k _x has a lower limit but no upper limit. For this reason, the diffraction limit becomes infinite in principle.

As a method for reversing the signs of ε _x and ε _z , a method of forming a multilayer film of metal and dielectric has been proposed. For example, Optics Express, Vol. 14, 8247-8256 (Non-patent Document 3) proposes a method of forming a multilayer film structure made of a metal and a dielectric in a semi-cylindrical groove provided in a substrate. The structure discussed in Non-Patent Document 3 is a multilayer film in which silver and aluminum oxide are alternately stacked in a semi-cylindrical groove formed in a substrate. This structure has not only the advantage that the near-field component does not attenuate, but also the advantage that the image is magnified by making the groove shape a semi-cylindrical shape. With this structure, a resolution three times smaller than the normally assumed resolution has been experimentally confirmed. In the hyperlens discussed in Non-Patent Document 3, the innermost film on the object side of the metal / dielectric multilayer film formed on the substrate has an arc shape. For this reason, only objects that fall within the innermost arc can be handled as objects to be observed. In order to solve this problem, Optics Express, Vol.16, 21142-21148 (2008) (Non-Patent Document 4) proposes a planar hyperlens that has a flat surface on the object side of the hyperlens and is simulated. Shows that the resolution is doubled.

  Here, when observing the light propagated from the hyper lens with a normal optical microscope, it is necessary to use, for example, an oil immersion lens used in a normal optical microscope. This is because the hyperlens transmits complex wave number information on the sample, but this propagating light is totally reflected when entering the air from the hyperlens substrate because of its large incident angle. is there. This total reflection can be prevented if it is incident on a liquid such as oil having a small refractive index difference from the substrate.

S. M. Mansfield and G. S. Kino, “Solid immersion microscope”, Applied Physics Letter, Vol.57, pp.2615-2616 (1990). T. Matsumoto, T. Shimano, H. Saga and H. Sukeda, “Highly efficient probe with a wedge-shaped metallic plate for high density near-field optical recording”, Journal of Applied Physics, Vol.95, pp.3901- 3906 (1995). Hyesog Lee, Zhaowei Liu, Yi Xiong, Cheng Sun and Xiang Zhang, “Development of optical hyperlens for imaging below the diffraction limit”, Optics Express, Vol.15, pp.15886-15891 (2007). Weo Wang, Hui Xing, Liang Fang, Yao Liu, Junxian Ma, Lan Lin, Changtao Wang and Xiangang Luo, “Far-field imaging device: planar hyperlens with magnification using multi-layer metamaterial”, Optics Express, Vol.16, pp .21142-21148 (2008).

  As described above, various lenses that exceed the diffraction limit of light have been proposed, but each has a problem.

  In the case of using SIL, a high refractive index substance is required to increase the resolution, but the maximum refractive index currently known in nature is about 3, and it is difficult to increase the resolution to more than three times. In the method using Nanobeak, when the near-field light generated between the Nanobeak tip and the sample is converted into propagating light, the intensity rapidly decreases, and the use as a lens is greatly limited.

  The hyper lens theoretically has a very high resolution and can convert the near-field component into a propagating component, thereby increasing versatility. However, the multilayer structure of metal and dielectric allows the film thickness and the optical constant of the material to be changed. The effect is manifested only in a narrow wavelength region to be determined. Therefore, it is impossible to know the optical dispersion of the object to be observed with a single hyper lens. In addition, the intensity of propagating near-field light is weak, and the efficiency with which the near-field light is converted into propagating light is poor. Moreover, in order to obtain the light propagated through the hyper lens, it is necessary to use an oil immersion lens or the like. The refractive index of a liquid such as oil used here is about 1.3 to 1.7, for example. When a near-field with a high spatial frequency is converted into propagating light, the propagating angle is close to the sample surface, so in order to obtain this propagating light with an optical microscope, a liquid with a very large refractive index is used as a hyper lens. Between the substrate and the lens of the optical microscope. However, the problem is that no such liquid exists.

  The present invention solves the above two problems related to the hyper lens by improving the structure of the hyper lens, using a plurality of hyper lenses, and adopting a system suitable therefor.

  An optical microscope system according to the present invention includes a sample holder that holds a sample, a plurality of light sources that generate sample irradiation light of different wavelengths, and a plurality of near-field components that convert the near-field components into propagation components in accordance with the wavelengths of the light sources. A hyper lens, a switching mechanism that switches between one of a plurality of light sources and one of a plurality of hyper lenses, a microscope that receives light transmitted through the hyper lens, and a magnified image by the microscope An image sensor and a calculation control unit that performs calculation and control of each unit of the apparatus are provided.

  The hyper lens has a flat light incident surface, and when viewed in one cross section passing through the optical axis, the hyper lens has a semicircular or substantially semicircular metal layer or dielectric layer centered on the light incident surface side, and the semi-lens. It has a laminated film structure comprising metal layers and alternating layers of dielectric layers which are concentrically laminated sequentially on the outside of a circular or substantially semicircular metal layer or dielectric layer. The three-dimensional shape of the hyper lens is hemispherical or semicylindrical. The hyper lens may include a light emitter layer outside the laminated film structure. Moreover, you may have a metal pattern for a near field component reinforcement | strengthening in the light-incidence surface.

  The arithmetic control unit controls the switching mechanism to sequentially select each of a plurality of light sources and a hyper lens that is paired with the light source, acquires a plurality of sample images with different wavelengths, and selects one of the acquired plurality of sample images. By combining the sample images of the sheets, information on the wavelength dependence of the optical characteristics of the sample is obtained with a size equal to or smaller than the wavelength of the incident light.

  The plurality of light sources are typically a light source that generates blue light, a light source that generates green light, and a light source that generates red light.

  According to the present invention, the wavelength characteristic of a sample can be obtained with a resolution exceeding the diffraction limit. Further, by combining the hyper lens and the light emitting layer, the optical system can be simplified, and the efficiency with which near-field light is converted into propagating light can be increased.

Schematic of the hyper lens used in the present invention. Explanatory drawing of the positional relationship of the sample and detector at the time of using the hyper lens of this invention. The relationship diagram of a hyper lens and a microscope. The figure which shows the relationship of the light transmittance which permeate | transmits a lens with the wave number of the x direction of the hyper lens by a Fresnel formula. Explanatory drawing of the polarization dependence of the spatial frequency and signal intensity of a hyper lens. Explanatory drawing at the time of providing a metal pattern in the sample surface of a hyper lens. Explanatory drawing of the manufacturing process of a hyper lens. The figure which shows the wavelength dependence of the relationship between the wave number of the lens designed according to each wavelength, and the transmittance | permeability at the time of using the light source from which a wavelength differs. Explanatory drawing of the sample which has a site | part from which the wavelength dependence of an optical characteristic differs. Explanatory drawing of the microscope system which replaces | exchanges the designed and produced lens automatically. Microscopic image of a sample obtained using multiple light sources and a hyperlens suitable for each light source. Explanatory drawing of position alignment at the time of synthesize | combining a microscope image and intensity | strength alignment. The figure which shows a synthetic | combination microscope image. Explanatory drawing of position alignment at the time of synthesize | combining a microscope image and intensity | strength alignment. Explanatory drawing of the system which acquires an image using different polarized light. Explanatory drawing of the optical signal obtained using different polarization | polarized-light, and its calculation. The figure which shows a synthetic | combination microscope image. Explanatory drawing of an observation sample with a random width and its scanning system. Explanatory drawing of an optical signal. Explanatory drawing of the system which controls the distance of a hyper lens and a sample with an electrostatic capacitance. Explanatory drawing of the hyper lens which provided the light-emitting body.

Embodiments of the present invention will be described below with reference to the drawings.
First, a method for observing the optical dispersion of the sample will be described. For this purpose, in the present invention, a plurality of hyper lenses having different structures are used. The structure of the plurality of hyperlenses is a structure in which the wavelength region for propagating near-field light is different. For example, three hyper lenses are created, each of which propagates near-field light of red (R), green (G), and blue (B) wavelengths, and the images obtained by the respective lenses are superimposed. Thus, the wavelength dependence of the optical properties of the sample can be observed.

  Next, a design method for a hyper lens used in the present invention will be described. The hyper lens of the present invention has a laminated film of two or more layers of a metal and a dielectric film formed in a structure such as a groove or pit formed on a substrate. FIG. 1A is a schematic cross-sectional view of an example of the hyper lens of the present invention. FIG. 1A is a cross section passing through the optical axis, and the light incident surface 104 is a flat surface. The hyper lens of the present invention includes a multilayer film of two or more layers of a metal 102 and a dielectric 103 formed on a substrate 101. When viewed in this cross section, the multilayer film is composed of a semicircular or substantially semicircular metal layer or dielectric layer having a center on the light incident surface 104 side, and a metal layer and a dielectric that are sequentially and concentrically stacked on the outer side. It has a laminated structure composed of alternating layers. The order of film formation may be the metal 102 first or the dielectric 103 first. When the cross section of the shape is circular, the near-field component can be efficiently propagated. FIGS. 1B and 1C are top views of two embodiments of the hyper lens. The hyper lens having the top view shown in FIG. 1B has a hemispherical overall shape. The hyper lens having the top view shown in FIG. 1C has a semicylindrical overall shape.

  Next, FIG. 2 shows the arrangement relationship between the hyper lens of the present invention and the object to be observed (sample). The observed object 201 is arranged on a flat surface 202 of the two surfaces of the hyper lens of the present invention. Incident light 203 is incident from the flat surface 202 side of the lens. The light 204 transmitted through the hyper lens is detected by a detection device 206 on the opposite surface 205 of the flat surface of the hyper lens. The distance between the lens-side uppermost surface of the object to be observed 201 and the flat surface 202 of the hyper lens is preferably shorter than the wavelength of the incident light 203. This is because when the distance between the lens-side surface of the object to be observed 201 and the flat surface 202 of the hyper lens is equal to or greater than the wavelength, the near-field component attenuates before entering the hyper lens.

  Next, an example in which the hyper lens of the present invention is incorporated in a microscope is shown in FIG. In FIG. 3A, an object to be observed 301 is arranged on a sample holder 302 arranged on the incident light 303 side. A hyper lens 304 of the present invention is disposed immediately above the object to be observed 301. The hyper lens 304 is fixed by the lens holder 305. The light 306 transmitted through the hyper lens 304 is received by the microscope 307, and the light 308 that has passed through the microscope 307 is observed by the detector 309. FIG. 3B shows an example in which the oil 310 is filled between the hyper lens 304 and the microscope 307. Filling the oil 310 increases the NA and improves the resolution. FIG. 3C shows an example in which the hyper lens 304 and the microscope 307 are brought close to each other. By bringing the hyper lens 304 and the microscope 307 close to each other, many components of the light 306 transmitted through the hyper lens 304 can be received by the microscope 307.

When designing the hyper lens, the k _x dependency of the electric field transmittance after passing through the hyper lens can be calculated using the Fresnel equation. The wavelength incident constant transmittance k _x in the region of calculation by k <k _x is not a zero, meaning that the near-field components exceeding the diffraction limit is propagated. Depending on the wavelength of the light incident on the hyper lens, the calculation is performed by changing the metal and dielectric materials, the film thickness, and the total number of the multilayer films, and the largest near-field component k> k _x is propagated. What is necessary is just to determine the structure.

FIG. 4 shows an example of transmittance calculated using the Fresnel equation. The wavelength of the incident wave used is 400 nm, and the configuration of the hyper lens is Ag (20 nm) / Al ₂ O ₃ (20 nm), which is a total of 10 layers. In FIG. 4, the horizontal axis represents a value k _x / k ₀ obtained by normalizing k _x with a wave number k ₀ corresponding to 400 nm, and the vertical axis represents the electric field transmittance after passing through the lens. As can be seen from FIG. 4, light (component of k _x / k ₀ > 1) exceeding the diffraction limit of 400 nm, that is, a near-field component propagates. Figure transmittance has become 0% at 4, k _x / k ₀ = 4 (arrow position), in the configuration described above it can be seen that is four times the diffraction limit resolution is normally expected.

  Next, a method for selecting a metal will be described. The metal for producing the hyper lens depends on the wavelength of the light source used for observing the sample. When this wavelength is visible light, it is desirable to use at least one of Au, Ag, Cu, Al, and Pt. In selecting the metal to be used, it is necessary to consider the plasma frequency and conductivity of the metal. If the conductivity is small, light is attenuated immediately in the metal, and the near-field component cannot be observed. Moreover, when light exceeding the plasma frequency is incident, electrons cannot follow the vibration of the light. As a result, the dielectric constant becomes positive above the plasma frequency and does not behave as metal and cannot be used for the film. For example, since the plasma frequency of Ag exists in the ultraviolet region, it can be used for all red, blue, and green lasers. Further, for example, the plasma frequency of Au exists near the middle of green and blue, and cannot be used for blue light.

After determining the metal for producing the hyper lens, the type of dielectric used for producing the lens, the film thickness of the metal and dielectric, and the total number of multilayer films are determined, and the lens performance is estimated using the Fresnel equation. At that time, the kind of dielectric suitable for use as a lens material and the film thickness ratio between the metal and the dielectric can be easily estimated by the following procedure. The dielectric constant of the metal epsilon _{m, p} the dielectric constant of the dielectric epsilon, the thickness of the metal d _m, and the thickness of the dielectric and d _p, these relationships are as follows.
d _m / d _p = −Re (ε _m ) / ε _p (7)

Here, Re (ε _m ) is the real part of the dielectric constant of the metal. For example, when the thickness of the dielectric used for the hyper lens is substantially equal to that of the metal, it is desirable to use a dielectric having a dielectric constant substantially equal to the absolute value of the real part of the dielectric constant of the metal used. This means an impedance matching condition.

  Next, a system for obtaining the wavelength dependence of the optical characteristics of the object to be observed with a size equal to or smaller than the wavelength of the incident light will be described. Observe the object to be observed using multiple hyper lenses corresponding to the wavelength of each incident light, and superimpose multiple images to obtain information on the wavelength dependence of the optical properties of the object to be observed. It can be obtained with a size less than the wavelength. When a plurality of images obtained at each wavelength are overlaid, alignment marks that can be observed at all wavelengths used for observing the object to be observed are required. Furthermore, it is necessary to match the intensities of a plurality of images obtained at each wavelength. The user can select what is used as the alignment mark. For example, the alignment mark is added on the sample, and alignment is performed using a system that can perform alignment using the mark. The alignment mark may be a scratch on the sample. An alignment mark may be added on the sample holder. For intensity adjustment, a method of calibrating with a calibration sample at each wavelength, an appropriate weight is applied to the intensity of the image obtained at each wavelength, and it is manually changed to obtain the clearest image. The method of obtaining is conceivable. It is desirable to use a material whose dispersion characteristics are well known for the calibration sample.

  Next, a description will be given of a method for emphasizing information having a size equal to or smaller than the wavelength of incident light in the wavelength dependence of the optical characteristics of the object to be observed. In a hyper lens made of a laminate of a metal and a dielectric, the optical characteristics obtained differ depending on the polarization direction of incident light. That is, when TM light is incident, the near-field component propagates far, but when TE light is incident, the near-field component is immediately attenuated. Here, for example, in the case of a semi-cylindrical hyperlens, TM light is light in which the magnetic field is directed in a direction perpendicular to the direction along the axis of the semi-cylinder, and TE light is relative to the axis of the semi-cylinder. The light is directed in a parallel direction. The light propagation direction can be described by the TE light and TM light. When TM light is incident on the object to be observed and the hyper lens of the present invention is used, information exceeding the diffraction limit can be obtained. However, since the background light is large, information exceeding the diffraction limit is usually used. It is buried in the component that propagates. Therefore, the effect of only the near-field component is measured by measuring the characteristics of the hyper lens with respect to the two polarized lights and subtracting them.

The above process will be described below with reference to FIG. First, an observation object is placed on a flat surface opposite to the curved surface of the hyper lens, a TM wave having a specific wavelength is incident, and the intensity is measured by an imaging device such as a CCD camera. This strength is _ITM . Next, the TE wave is incident and the intensity is measured. This strength is assumed to be _ITE . In order to align the incident positions of the TM wave and the TE wave, an alignment mark is made on the sample, and alignment is performed using the alignment mark. After alignment, components other than the near-field component are subtracted, but the signal strength obtained is different between I _TM and I _TE . Therefore, the Fourier transform of I _TM and I _TE, the value after the Fourier transform and F (I _TM) and F (I _TE).

FIG. 5A shows F (I _TM ) and F (I _TE ). When the horizontal axis in FIG. 5A is the spatial frequency and the vertical axis is the component obtained by Fourier transforming the signal intensity of the transmitted electric field, information exceeding the diffraction limit is propagated when TM polarized light is incident, and therefore F (I _TM ) is a space. The signal strength does not become zero even when the frequency is 2π / λ or more. On the other hand, since information beyond the diffraction limit does not propagate when TE polarized light is incident, F (I _TE ) has a spatial frequency of 2π / λ, the signal intensity becomes zero, and the above F (I _TE ) becomes background light. . In order to increase the signal intensity of the propagating near-field component, it is assumed that F (I _TM ) and F (I _TE ) are linear, and the TE wave component is weighted and F (I _TM ) −αF (I _TE ) In consideration of the above quantity, a system for determining α so that the sum of 2π / λ <(spatial frequency) portions of F (I _TM ) −αF (I _TE ) is minimized is constructed. FIG. 5B shows an example in which α is actually determined and F (I _TM ) −αF (I _TE ) is obtained. The signal intensity of a component having a spatial frequency of 2π / λ or more, that is, a component having a magnitude smaller than λ is increased, and the near-field component can be measured well. The polarization direction used depends on both the shape of the lens and the shape of the object to be observed. If the lens has a semi-cylindrical shape, TE polarized light and TM polarized light may be used. If the shape of the lens is hemispherical, it is necessary to change the direction of incident polarized light depending on the shape of the sample.

  Next, a method for increasing the signal of the near-field component with respect to the entire signal will be described. This can be realized by forming a metal pattern having a size equal to or smaller than the wavelength of incident light on the surface of the hyper lens on the observation object side. If a metal is used as a component when creating a hyper lens, light dissipation due to the metal occurs. For this reason, the near-field component of the light emitted from the lens becomes very weak, and information exceeding the diffraction limit is lost. When light is incident on a metal pattern or metal structure with a wavelength shorter than the wavelength of the incident light described above, free electrons in the metal vibrate, but the free electron density near the tip increases. Strong near-field light localized in the region is generated, that is, plasmons are concentrated. By the metal pattern, plasmons are concentrated, and the near-field light is enhanced to compensate for the loss by the metal, and a high S / N ratio can be obtained.

  The size and shape of the metal pattern that enhances plasmons depend on the wavelength of incident light. For this reason, several metal pattern structures that enhance plasmons are conceivable. For example, a triangular shape is conceivable. Moreover, although the number of the metal patterns to arrange | position is considered, several types are considered, for example. The distance between the apex or surface of the metal pattern on the hyper lens side and the center of the innermost metal hemisphere of the hyper lens is preferably equal to or shorter than the wavelength incident on the hyper lens. When this distance is greater than or equal to the wavelength of the incident light, the near-field component is attenuated and does not function as a near-field enhancing element. Plasmon concentration can also be realized by making the size of the innermost metal out of the metals constituting the hyper lens smaller than the incident wavelength. The conceptual diagram of the metal pattern produced on the board | substrate is shown in FIG. FIG. 6 shows a case where the shape of the metal pattern is a triangle. A triangular metal pattern is formed on the substantially flattened surface side of the hyper lens 601. 6A is a cross-sectional view, FIG. 6B is a top view when the hyper lens is hemispherical, and FIG. 6C is a top view when the hyper lens is groove-shaped. In FIG. 6, metal layers 603 and dielectric layers 604 are alternately laminated in order from the flat surface of the hyper lens.

  In order to form a metal pattern having a size equal to or smaller than the wavelength of incident light on the hyper lens, it is necessary that the surface on the object to be observed side of the two surfaces of the hyper lens to be manufactured is substantially flat. Hereinafter, an outline of a method for manufacturing a hyper lens will be described. First, pattern drawing is performed on the substrate using lithography or nanoimprint. As the substrate, a glass substrate, a Si substrate, a semiconductor substrate, a resin substrate, and the like are conceivable. Next, a hemispherical or semi-cylindrical depression according to the pattern is formed on the substrate by reactive ion etching. Next, a metal and a dielectric are formed alternately in the formed depression. As the film forming method, a sputtering method, a vapor deposition method, and a chemical vapor deposition method (CVD method) can be considered. Finally, the object side is flattened. As the planarization method, milling, reactive ion etching, and chemical mechanical polishing (CMP) can be considered. In CMP, polishing on plastic is also possible.

  Next, a method of scanning an object to be observed using the hyper lens of the present invention provided with a metal pattern for enhancing the near-field component will be described. A relatively large object to be observed can be observed by scanning. The scanning method using the hyper lens provided with the metal pattern includes a method of scanning while physically contacting the object to be observed and the metal pattern, and scanning without physically contacting the object to be observed and the metal pattern. There is a way. When scanning while physically observing the object to be observed and the metal pattern, scanning is performed while mechanically controlling the sample holder holding the object to be observed. In the method of scanning without physically touching the object to be observed and the metal pattern, if the object to be observed has conductivity, the object to be observed and the metal pattern are not in physical contact with each other. There is a method of controlling the interval by measuring the capacitance of the.

  Moreover, it is desirable that the cross-sectional shape of the hyper lens of the present invention is a semicircular shape or a semi-elliptical shape. By making it semicircular, the near-field component emitted from the lens is propagated. Moreover, a near-field component can be propagated even in a semi-elliptical shape.

  In order to obtain the propagation light formed by the hyper lens with an optical microscope, a light emitter layer may be added to the hyper lens without using the oil immersion lens. When a light emitter layer is provided outside the outer spherical surface of the hyper lens, the light emitter layer absorbs propagating light and generally emits light having a different wavelength. This light propagates in all solid angle directions regardless of the propagation direction of the propagation light converted by the hyper lens. As a result, since this light is incident on the hyperlens substrate at a shallow angle, it propagates into the air without being totally reflected.

  As this light emitter, for example, a phosphor such as a polymer or a semiconductor quantum dot can be used. It is necessary to install the luminous body layer at a location far from at least the light source wavelength from the center of the hyper lens. This is because the spatial frequency of the light profile is high in the portion below the light source wavelength from the center of the hyper lens, that is, in the near-field region. For this reason, when a light emitter layer is installed in the near-field region, the component having a high spatial frequency in the propagating light generated from the light emitter layer is attenuated, so that the resolution obtained by the optical microscope is reduced. is there.

Furthermore, by using the light emitting layer, background light having a low spatial frequency can be removed. The light source wavelength lambda _1, and the wavelength of fluorescence emitted from the phosphor layer and lambda _2, is generally a λ ₁ <λ _2. The light emitted from the hyperlens substrate is a mixture of these two wavelengths. Here, when the oil immersion lens is not used, the high spatial frequency component of the light of λ ₁ is not emitted from the substrate due to total reflection, so that only the low spatial frequency component is emitted from the substrate. On the other hand, the light of λ ₂ includes both a low spatial frequency and a high spatial frequency. Here, an optical filter is installed in the optical microscope to obtain an image of light having only the wavelength λ ₁ and an image of light having only the wavelength λ ₂ . If the intensities of light at one point λ ₁ and λ ₂ on the image are I (λ ₁ ) and I (λ ₂ ), respectively, an appropriate value α is adopted and I (λ ₂ ) −αI (λ ₁ ), It is possible to obtain an image of only the component light having a spatial frequency higher than the wavelength of the incident light on the sample. Usually, since the intensity of the near-field light is weak, a signal having a high spatial frequency component is buried in a background signal having a low spatial frequency component, and the S / N of the image is lowered. A resolution image can be obtained. A phosphor layer and a near-field component enhancing metal pattern may be used in combination.

Hereinafter, the present invention will be described in more detail with reference to examples.
[Example 1]
A method for obtaining color information of an object to be observed with a size equal to or smaller than the wavelength of incident light using the optical microscope system of the present invention will be described. The hemispherical hyperlens shown in FIGS. 1 (a) and 1 (b) was designed to maximize performance at the wavelengths of the red laser, blue laser, and green laser. The red laser light source used was He—Ne with a wavelength of 633 nm, the green light source used was a YAG laser double wave with a wavelength of 532 nm, and the blue laser light source used was a He—Cd laser with a wavelength of 405 nm. The spot diameter is 2 μm for the red laser, 1 μm for the green laser, and 1 μm for the blue laser.

The procedure from the production of the hemisphere to the substrate, the formation of the dielectric / metal film, and the substantially flattening of the surface of the object to be observed is as follows. First, a hemispherical depression was formed on a substrate having a thickness of 1 mm. SiO ₂ was used for the substrate. The hemispherical pattern was formed by a method using a stamper. FIG. 7 shows the procedure.

First, as shown in FIG. 7A, a Cr layer 702 and a resist layer 703 were sequentially applied on a SiO ₂ substrate 701, and a circular pattern having a diameter of 1 μm was drawn on the resist layer 703 by laser drawing. Next, development was performed as shown in FIG. 7B, and the drawn Cr layer 702 was removed by reactive ion etching (RIE) as shown in FIG. 7C. Then, wet etching was performed on the SiO ₂ substrate 701 to form a hemispherical depression 704 as shown in FIG. The diameter of the hemispherical depression 704 was 2.0 μm. The diameter of the hemispherical depression 704 is determined by the radius of the circular pattern drawn on the SiO ₂ substrate 701. Finally, as shown in FIG. 7E, the resist layer 703 and the Cr layer 702 were removed. Using the hemispherical depression 704 formed as described above as an original, Ni was vapor-deposited and Ni-plated on the original to transfer the original hemisphere to form a stamper 706 as shown in FIG. Using the stamper 706, the hemisphere of the stamper 706 was transferred to the UV resin 708 applied to the glass substrate 707 by injection molding to form a hemispherical depression 709 as shown in FIG.

Next, a metal or dielectric film was formed on the substrate by sputtering. Considering the plasma frequency, the metal used for the hyper lens was Ag for the blue laser hyper lens, Au for the green laser hyper lens, and Au for the red laser hyper lens. The film structure of the hyper lens for blue laser _{Ag (20nm) / Al 2 O} 3 (20nm), a total of 10 layers in radius 400 nm, the film structure of the green laser for hyper lens _{Au (20nm) / Al 2 O} 3 (20nm) A total of 10 layers had a radius of 400 nm, and the red laser hyperlens film structure had a total of 10 layers of Au (20 nm) / Al ₂ O ₃ (20 nm) with a radius of 400 nm. Finally, in order to flatten the surface of the hyperlens to be observed, mechanical chemical polishing (CMP) that can polish the resin was performed.

FIG. 8 shows the calculation result of the transmittance using the Fresnel equation when a laser is incident on the prepared three hyperlenses from the observed object side. The horizontal axis value k _x / k ₀ normalized wave number k ₀ of the light incident to k _x is the x component of the wave number, and the vertical axis represents the transmittance of the electric field strength. FIG. 8A shows the calculation result of the blue laser hyperlens, FIG. 8B shows the calculation result of the green laser hyperlens, and FIG. 8C shows the calculation result of the red laser hyperlens. The arrow in the figure represents the normalized wave number at which the electric field intensity transmittance is zero. According to the above calculation, the resolution of the blue laser hyperlens is 4 times the diffraction limit, the resolution of the green laser hyperlens is 4 times the diffraction limit, and the resolution of the red hyperlens is 5 times the diffraction limit. all right.

  Next, three types of dyes were observed using the three hyperlenses. The dyes used were fluorescein isothiocyanate having an excitation wavelength at 495 nm, rhodamine having an excitation wavelength at 555 nm, and Alexa Fluor 647 having an excitation wavelength at 650 nm. As shown in FIG. 9, three types of dyes were periodically arranged on the sample holder 901. 902 is fluorescein isothiocyanate, 903 is rhodamine, and 904 is Alexa Fluor 647. Reference numeral 905 denotes an alignment mark. The size of the dye was 1 μm in diameter, and the distance between the dyes was 200 nm.

  Next, observation will be described. The hyper lens of the present invention was incorporated into a laser microscope. FIG. 10 shows a schematic diagram of a microscope system incorporating the hyper lens of the present invention. Three hyper lenses corresponding to each color were placed on one black plastic plate (lens holder) 1001 and controlled by a lateral control device 1002 so that they could be moved mechanically. The three colors of lasers are arranged in the light source unit 1003 so that the light exits on the same optical axis from one exit port. Of the three types of laser light, unused laser light is shielded by a light shielding plate. A light shielding plate exchanging device 1013 was used for light shielding. Actually, the black plastic plate 1001 and the light shielding plate are controlled on a personal computer (PC) as the arithmetic control unit 1004. If a wavelength to be used is determined and a button on the PC is clicked, the corresponding wavelength is selected. The lens is automatically switched to the hyper lens, and the light shielding plate is also switched so that the laser light of that wavelength is emitted from the light source unit 1003.

  As the object to be observed, the above three kinds of dyes 1006 were arranged in a sample holder 1005, and a hyper lens 1007 was installed immediately above. The hyper lens 1007 is fixed to the black plastic plate 1001. Incident light 1008 was incident, light 1009 that passed through the hyper lens 1007 was received by the microscope 307, and light 1011 that passed through the microscope 307 was observed by the CCD camera 309. The observed data was stored in the calculation control unit 1004. In this embodiment, the space between the hyper lens 1007 and the microscope 307 is filled with the immersion oil 310. By filling the space between the optical microscope 307 and the hyper lens 1007 with the immersion oil 310 whose refractive index is almost equal to that of glass, the total reflection component can be reduced.

  The measurement was performed while exchanging the laser light source and the hyper lens. First, measurement was performed using a blue laser and a blue laser lens. A blue laser and a lens were selected by clicking on the PC screen as the arithmetic control unit 1004, and measurement was performed. After that, measurements were performed using a green laser and a green laser lens, and finally measurements were performed using a red laser and a red laser lens.

  FIG. 11 shows images obtained at the respective wavelengths. FIG. 11A shows the measurement result of the blue laser, FIG. 11B shows the measurement result of the green laser, and FIG. 11C shows the measurement result of the red laser. As expected from the calculation results, structures below the resolution could be observed at any wavelength. In FIG. 11A, since the excitation wavelength of fluorescein isothiocyanate 1102 is 495 nm and is longer than the blue wavelength of 405 nm, a slightly unclear image was obtained when observed with a blue laser hyperlens. In FIG. 11 (b), rhodamine 1104 has an excitation wavelength of 555 nm, so that an image can be clearly seen. In FIG. 11 (c), only Alexa Fluor 647 was observed.

  After the three types of measurements, alignment and intensity adjustment are required to superimpose the three images obtained. Two sets of two straight lines each having a width of 2 μm and a length of 5 μm were prepared on the sample holder 901 as alignment marks. This alignment mark 905 was produced by laser drawing.

FIG. 12 shows the procedure for alignment and intensity alignment. Specify the number of wavelengths to be used at the beginning and the size of the wavelength. In this embodiment, red, green, and blue are used, so 3 is entered. Next, the sample was measured at each wavelength. Image data obtained at each wavelength was captured by a CCD camera and stored in a memory in the arithmetic control unit. The obtained three data were displayed on the display. Next, each of the two alignment marks on the display was designated by the cursor as a point indicating the same point on the sample. With these two points, the coordinates (x, y) on the two images can be shared. After the registration, coefficients α _B , α _G , and α _R for matching the intensities of the obtained images were designated. is _{α B I B + α G I} G + α R I R. This coefficient was set manually to perform calculation, and set so that the image of the object to be observed became clearest. The values were α _B = 0.2, α _G = 0.5, α _R = 0.4.

  The results are shown in FIG. As a result, an image having a resolution exceeding the optical resolution including the color information of the dye was obtained.

  Similar results were obtained even when a semiconductor, a Si substrate, or a resin was used as the substrate for manufacturing the stamper.

  In addition, as a measurement system, arrangement | positioning like Fig.3 (a) and FIG.3 (c) is also considered. In FIG. 3A, the space between the hyper lens 304 and the microscope 307 is air. For this reason, although the sharpness of the image obtained compared with the arrangement of FIG. 3B is low, an image including information on the near-field component was obtained. Further, in FIG. 3C, almost the same result as in FIG. 3B was obtained.

  In addition, when a hyper lens with an elliptical cross-section was used to measure the same experimental system, measurement method, and observed object, the resolution was higher than when a hyper lens with a circular cross-section was used. Although it was low, almost the same result was obtained.

[Example 2]
Another embodiment will be described in which the color information of the object to be observed is obtained with a size smaller than the wavelength of the incident light using the optical microscope system of the present invention. The experimental system, the measurement method, the hyper lens used, and the object to be observed were the same as in Example 1, except that a calibration sample was used when the intensity of the three images was adjusted. This time, an Al flat plate (longitudinal and transverse 1 μm, thickness 10 nm) with well-known dispersion characteristics was used as a calibration sample. It is desirable to use a calibration sample whose dispersion characteristics are well known.

FIG. 14 shows a flowchart for adjusting the intensity. The procedure is almost the same as in Example 1, but after specifying the number of wavelengths, the transmission intensity of Al of the calibration sample is measured at each wavelength of the blue laser, green laser, and red laser, and stored in the memory of the arithmetic control unit. did. Then, after the alignment, a coefficient α for matching the intensities of the three obtained images was specified in comparison with the transmission intensities of the respective colors obtained by the calibration sample. Intensity I _B of the observed object image corresponding to the blue laser, the intensity of the observed object image corresponding to the green laser I _G, the intensity of the observed object image corresponding to the red laser and I _R, a blue laser of Al the ratio of the transmitted intensity and the transmitted intensity and I _B of the red laser ratio of alpha _G, Al between transmitted intensity and I _G of the ratio between I _B green laser α _B, Al α _R and in Then, the whole image is α _B I _B + α _G I _G + α _R I _R. In this example, α _B = 0.17, α _G = 0.61, and α _R = 0.35, which were almost the same as those in Example 1. As a result, almost the same result as in Example 1 was obtained.

[Example 3]
Next, an embodiment using a semi-cylindrical hyperlens having a groove shape as shown in FIG. The material used to fabricate the hyper lens is the same as in Example 1. In order to make a groove-shaped lens, a rectangular groove having a width of 2 μm and a length of 1 mm was drawn on the substrate by laser drawing. Subsequent hyper lens fabrication procedures are the same as in Example 1. The cross-sectional shape of the hyper lens of this example is shown in FIG. 1A, and the top view is shown in FIG. An observation object was placed on the innermost metal on the surface. The substrate 101 was a glass substrate, and the metal 102 was silver for a blue laser, and gold for a green laser and a red laser. The dielectric 103 is Al ₂ O ₃ . When the hyper lens has a semi-cylindrical shape, the performance of the hyper lens changes depending on the polarization direction of the incident light. That is, when the electric field is parallel to the axis of the semi-cylinder (TM polarized light), it works as a hyper lens that transmits the near-field light component, but when the electric field is perpendicular to the axis of the semi-cylindrical (TE polarized light) It works as a hyper lens and cannot transmit near-field components. By utilizing this, if the intensity of transmitted light when observed with TE polarized light is subtracted from the intensity of transmitted light when the same observation object is measured with TM polarized light, only the near-field component can be observed.

  A configuration example of the microscope is shown in FIG. The configuration of the microscope is almost the same as the configuration of the first embodiment shown in FIG. 10, but a polarizer 1514 is installed immediately after the laser emission port of the light source unit 1003 to select the polarization direction, and the laser microscope 307. An analyzer 1515 is installed at the observation port. Further, an image obtained by the microscope 307 was picked up by the CCD camera 309 to obtain a signal. The observation object 1006 used is a line and space with a pitch of 100 nm, and the total length is 700 nm.

First, measurement was performed using a blue laser and a hyperlens for blue laser. The screen on the arithmetic control unit 1004 was clicked to select a blue laser and a hyper lens for the blue laser, and measurement was performed. The measurement result using a blue laser is shown in FIG. FIG. 16A shows the electric field strength 1601 in the case of TM waves and the electric field strength 1602 in the case of TE waves. The horizontal axis represents the distance Δ from the end of the line and space, and the vertical axis represents the detected electric field intensity. As described above, the overall intensity of the TE wave is smaller than that of the TM wave. Therefore, the electric field intensity was adjusted using a system for matching the intensity, and the background component was subtracted. In this example, the obtained signal intensity was Fourier transformed to adjust α so that the sum of 2π / λ <(spatial frequency) portion of F (I _TM ) −αF (I _TE ) is minimized. In this example, α = 0.7. FIG. 16B shows the result of subtracting the electric field strength at the TE wave from the electric field strength at the TM wave. As can be seen from FIG. 16B, the S / N clearly increased.

  Thereafter, measurement was performed by changing the laser to a green laser and a red laser. FIGS. 16C and 16D show the measurement results using the green laser and the green laser hyperlens, and FIGS. 16E and 16F show the red laser and the red laser hyperlens. The measurement results are shown. An increase in S / N was also observed in the case of the green laser. In the case of a red laser, the S / N ratio did not increase because the resolution did not exceed 100 nm even when the hyper lens of the present invention was used.

[Example 4]
In order to obtain a clearer image, a measurement example in which the near-field component is enhanced by using a hyper lens provided with a metal pattern on the sample surface will be described. The experimental system of this example is the same as that of Example 1, and the microscope system shown in FIG. 10 was used.

  The near-field enhancement pattern was an equilateral triangle, and the number was two. Using a hemispherical hyper lens, an equilateral triangular metal pattern 602 is formed on the substantially flattened surface of the hyper lens as shown in the cross-sectional view of FIG. 6A and the plan view of FIG. The triangular tip was formed to face the observation object. When a TM wave is incident on an object with a sharp tip, plasmons are enhanced, so that the near-field component is enhanced before entering the lens, and the amount weakened in the process of passing through the lens can be compensated. The metal material, width, and length used for the triangular pattern are determined by the wavelength used. This time, an equilateral triangular metal pattern having a thickness of 55 nm and a side length of 100 nm was formed on the flat surface of the blue laser hyper lens using Au. On the flat surface of the green laser hyperlens, a regular triangular metal pattern having a thickness of 75 nm and a side length of 150 nm was formed using Au. On the flat surface of the red laser hyperlens, an equilateral triangular metal pattern having a length of 85 nm and a side of 200 nm was formed using Au.

  The observed observation object is the same as in Example 1. The experimental procedure was also the same as in Example 1, and measurements were performed by sequentially irradiating a blue laser, a green laser, and a red laser, and the three images were superimposed. The results are shown in FIG. Compared with Example 1, a clearer image was obtained.

[Example 5]
Next, a measurement example in which a one-dimensional scan is performed using a hyper lens having a metal pattern for enhancing a near-field component on a flat surface while bringing the metal pattern into contact with an object to be observed will be described. The experimental system and the hyper lens used are almost the same as those in Example 1. However, the sample holder control was performed with a piezoelectric element. The same metal pattern as in Example 4 was used. As the object to be observed, a line and space formed on a polycarbonate was used.

  As shown in FIG. 18 (a), the width of the space was randomly changed from the left to 200 nm, 150 nm, 100 nm, and 150 nm. The total width of the line and space was 1 μm, which was sufficiently larger than the width of the hyper lens of 400 nm.

  FIG. 18B shows a difference from the first embodiment in the experimental system. As shown in FIG. 18B, the metal pattern 1802 fabricated on the hyper lens 1801 and the object to be observed 1803 are brought into physical contact, and the sample holder 1804 on which the object to be observed 1803 is placed is placed with respect to the optical axis. A one-dimensional scan was performed by driving in the vertical direction. The moving distance of the sample holder 1804 was controlled by adding a piezoelectric element 1805 below the sample holder and applying a voltage in the XY axis direction of the piezoelectric element 1805. The positional accuracy of the movement by the piezoelectric element 1805 was about 5 nm.

The laser and the hyper lens were observed in the same manner as in Example 1 while changing the order of blue, green, and red. First, the signal intensity obtained by using the blue laser is shown in FIG. A CCD was used as the detector. The scan range was set to 0 to 1.5 μm, with the left edge of the line and space being 0. In FIG. 19A, a peak 1901 corresponds to a space having a width of 200 nm, a peak 1902 corresponds to a space having a width of 150 nm, and a peak 1903 corresponds to a space having a width of 100 nm. The observation results using the green laser and the red laser are shown in FIGS. 19 (b) and 19 (c). The green laser has a resolution of about 120 nm and a low signal intensity. In addition, since the red laser has a resolution of about 200 nm, only a 200 nm space can be observed.
From the above, it was found that an object to be observed that is sufficiently larger than the lens can be observed.

[Example 6]
Next, a measurement example in which a hyper lens having a near-field component enhancing metal pattern on a flat surface is used and one-dimensional scanning is performed without contacting the object to be observed will be described. The configuration of the experimental system and the hyper lens used are almost the same as in Examples 1 and 5. As the observation object, conductive line and space 2003 having the shape shown in FIG. 18A was used. The line and space 2003 was shaken at random with the width of the space being 200 nm, 150 nm, 100 nm, and 150 nm from the left as in Example 5. The total width of the line and space was 1 μm, which was sufficiently larger than the width of the hyper lens of 400 nm.

  FIG. 20A shows a difference from Example 5 in the experimental system. The metal pattern 1802 formed on the hyper lens and the line and space 2003 were electrically connected, and the capacitance between them was measured to control the interval. The metal pattern 1802 and the object to be observed 2003 were connected using a Cu wire 2006. A capacitance bridge 2007 was used for the capacitance measurement. The movement of the sample holder 1804 was controlled by adding a piezoelectric element 1805 below the sample holder 1804 and applying a voltage in the XY axis direction of the piezoelectric element 1805. The positional accuracy of movement by the piezoelectric element 1805 is 5 nm.

The capacitance is inversely proportional to the distance between the metal pattern and the line and space. The capacitance of the line and space used in this example was calculated in advance, and the value at which the distance between the metal pattern and the top of the line and space was 10 nm was determined. In the actual scanning stage, scanning was performed while adjusting so that the capacitance did not exceed the value obtained by the above calculation. For example, in the case of blue, since one side of the near-field component enhancing metal pattern is 100 nm and the vacuum dielectric constant is 8.85 × 10 ⁻¹² (m ⁻¹ F), the near-field enhancing metal pattern and the space portion Has a capacitance of 3.83 (nF). FIG. 20B shows the obtained capacitance profile. The scan range was set to 0 to 1.5 μm, with the left edge of the line and space being 0. The horizontal axis is the distance from the left of the observed object, and the horizontal axis is the capacitance.

  The laser and the hyper lens were observed in the same manner as in Example 1 while exchanging in the order of blue, green and red. As a result, the fluctuation of the obtained signal intensity was small compared to the case of Example 5. This is because the distance in the z direction is controlled in detail by capacitance without contact. From the above, it has been found that an observation object larger than the size of the hyper lens can be observed while performing a one-dimensional scan without contact when the electrostatic capacitance is used.

[Example 7]
An example in which a phosphor layer is provided on a hyper lens will be described. The produced hyper lens is shown in FIG. The manufacturing method is the same as the above method, but the radius of the hemispherical pit formed on the substrate is 5 μm. In addition, rhodamine molecules were deposited by 2 μm as the luminous body layer 2101 before forming the dielectric. Thereafter, Al ₂ O ₃ was sputtered with 2.5 μm, and 10 sets of Ag (20 nm) / Al ₂ O ₃ (20 nm) were laminated. Thus, the produced hyper lens can propagate near-field light in the green wavelength band.

  With this lens, a diffraction grating having a pitch of 100 nm was observed. The lens shown in FIG. 21 was placed on the diffraction grating, and light emitted from the back surface of the lens was observed with an optical microscope. Here, air was used between the back surface of the lens and the objective lens of the optical microscope. In addition, a wavelength filter can be inserted between the objective lens and the eye lens in the optical microscope. A YAG laser was used as the light source.

  First, a wavelength filter for cutting light having a wavelength of 600 nm or less was inserted to obtain a microscopic image. This image was taken with a CCD camera and stored in a computer. This image is named Image 1. Next, an image was obtained in the same manner as described above using a wavelength filter that cuts light having a wavelength of 600 nm or more. This image is named Image 2.

  Image 1 is an optical image by fluorescence emitted from rhodamine of the luminescent layer 2101, and image 2 is an optical image of the wavelength of the YAG laser. The computer incorporated a program having a function of subtracting the intensity profiles of the two images.

On the computer, two points of image 1 and image 2 were designated by the cursor as points indicating the same point on the sample. With these two points, the coordinates (x, y) on the two images can be shared. Next, a coefficient α for subtracting the image intensity profile was specified. The coefficient α is I ₁ (x, y) −αI ₂ , where intensities at coordinates (x, y) on the images 1 and 2 are I ₁ (x, y) and I ₂ (x, y), respectively. This is a coefficient for calculating (x, y). Here, several values of α were manually set, and a value was obtained where the diffraction grating image was most clearly obtained. As a result, when α was 0.2, the image became clearest. This reason is due to the fact that only the high spatial frequency component in the image 1 is shown on the image by subtracting the image 2 which is an image by the background light.

  An experiment using CdSe as a material of the light emitting layer 2101 was also conducted. The hyper lens film, optical microscope, observation sample, wavelength filter, and YAG laser used are the same as described above. As a result, the image of the diffraction grating was most clearly obtained when α was 0.08.

101: substrate, 102: metal, 103: dielectric, 201: object to be observed, 203: incident light, 206: detector, 301: object to be observed, 302: sample holder, 303: incident light, 304: hyper lens, 305: Lens holder, 307: Optical microscope, 309: Detector, 310: Oil, 601: Hyper lens, 602: Metal pattern, 603: Metal layer, 604: Dielectric layer, 901: Sample holder, 905: Alignment mark , 1001: Lens holder, 1002: Lateral direction control device, 1003: Light source unit, 1004: Calculation control unit, 1005: Sample holder, 1006: Object to be observed, 1007: Hyper lens, 1013: Shading plate exchange device, 1514: Polarization 1515: Analyzer, 1801: Hyper lens, 1802: Metal pattern, 1803: Observation object, 1804: sample holder, 1805: a piezoelectric element, 2007: capacitance bridge, 2101: phosphor layer

Claims (14)

A sample holder for holding the sample;
A plurality of light sources for generating sample irradiation light of different wavelengths,
A plurality of hyper lenses that convert near-field components into propagating components according to the wavelength of each light source,
A switching mechanism for switching one of the plurality of light sources and one of the plurality of hyper lenses in pairs;
A microscope on which light transmitted through the hyper lens is incident;
An image sensor for capturing an enlarged image by the microscope;
An arithmetic control unit that controls the arithmetic and device parts,
The hyper lens has a flat light incident surface, and when viewed in one cross section passing through the optical axis, a semicircular or substantially semicircular metal layer or dielectric layer having a center on the light incident surface side, and Having a laminated film structure composed of alternating metal layers and dielectric layers, which are concentrically laminated on the outside of the semicircular or substantially semicircular metal layer or dielectric layer;
The arithmetic control unit controls the switching mechanism to sequentially select each of the plurality of light sources and a hyper lens paired with the light source, obtain a plurality of sample images with different wavelengths, and obtain the plurality of samples An optical microscope system characterized in that, by synthesizing one sample image from an image, information on the wavelength dependence of the optical properties of the sample is obtained in a size equal to or smaller than the wavelength of incident light.   2. The optical microscope system according to claim 1, wherein the plurality of light sources are a light source that generates blue light, a light source that generates green light, and a light source that generates red light.   The optical microscope system according to claim 1, wherein the hyper lens has a hemispherical shape or a semicylindrical shape.   4. The optical microscope system according to claim 1, wherein the hyper lens has a light emitting layer on an outer side of the laminated film structure. 5.   5. The optical microscope system according to claim 4, wherein the light emitter is a phosphor.   2. The optical microscope system according to claim 1, wherein the hyper lens has a near-field component enhancing metal pattern on the light incident surface. 3.   2. The optical microscope system according to claim 1, wherein coordinates common to all sample images are determined by designating at least two points on the sample image obtained at each wavelength.   2. The optical microscope system according to claim 1, further comprising a mechanism for adjusting the intensity of the plurality of sample images.   2. The optical microscope system according to claim 1, wherein each wavelength light is switched to two types of polarized light and irradiated to the sample, and two sample images obtained with each polarized light are calculated to obtain a sample image corresponding to the wavelength. An optical microscope system characterized by   The optical microscope system according to claim 9, further comprising a drive unit that drives the sample holder, and detecting a change in light intensity in synchronization with driving of the sample holder in a direction perpendicular to the optical axis by the drive unit. An optical microscope system characterized by   The optical microscope system according to claim 9, further comprising a measurement unit that measures a capacitance between the hyper lens and the sample, and using the measurement value obtained by the measurement unit during the driving, the hyper lens and the sample. An optical microscope system characterized by maintaining a constant distance between the two. A flat light incident surface;
A semicircular or substantially semicircular metal layer or dielectric layer having a center on the light incident surface side when viewed in one section passing through the optical axis, and the semicircular or substantially semicircular metal layer or dielectric layer A laminated film structure composed of alternating layers of metal layers and dielectric layers laminated concentrically sequentially outside
A phosphor layer formed outside the laminated film structure;
A hyperlens having a three-dimensional shape that is hemispherical or semicylindrical.   The hyperlens according to claim 12, wherein the light emitter is a phosphor. A flat light incident surface;
A semicircular or substantially semicircular metal layer or dielectric layer having a center on the light incident surface side when viewed in one section passing through the optical axis, and the semicircular or substantially semicircular metal layer or dielectric layer A laminated film structure composed of alternating layers of metal layers and dielectric layers laminated concentrically sequentially outside
A near-field component enhancing metal pattern formed on the light incident surface,
A hyperlens having a three-dimensional shape that is hemispherical or semicylindrical.

Images