JP2001272343A - Double resonance absorption microscope - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a new double resonance absorption microscope, capable of suppressing deterioration in S/N caused by the contamination of fluorescence signals with exciting light, and has superior three-dimensional spatial resolution. SOLUTION: This double resonance absorption microscope is provided with a light source for a pumping beam which excites sample molecules from its ground state to a first electron excitated state, and a light source for erasing light exciting the sample molecule in the first electron excitated state to a second electron excitated state or to a higher excitated state, and detects the emission of light generated, when the sample molecule in each excitated state is made to be de-excited to the ground state. The pumping beam has a photon energy which is 1/2 or less that of the excitation energy for exciting the sample molecules from the ground state to first electronic excitation state.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a double resonance absorption microscope which is a microscope utilizing a double resonance absorption process. More specifically, the invention of this application is based on the fact that the background light (excitation light) is mixed into the fluorescence signal.
The present invention relates to a new double resonance absorption microscope capable of suppressing deterioration of / N and having excellent three-dimensional spatial resolution.

[0002]

2. Description of the Related Art In recent years, various types of high-performance and multifunctional microscopes have been developed with advances in peripheral technologies such as laser technology and electronic image technology. One of the inventors of the present invention is a microscope that enables a contrast control of an obtained image and a chemical analysis of a sample using a double resonance absorption process generated by illuminating a sample with light of a plurality of wavelengths. (Hereinafter referred to as double resonance absorption microscope) (Japanese Patent Application No. 6-32916).
5).

In this double resonance absorption microscope, a specific molecule is selected using a double resonance absorption process, and the absorption and fluorescence resulting from a specific optical transition can be observed. To explain the principle, first, the ground state (S ₀ state: FIG. 1)
The electrons in the valence orbitals of the sample molecules (that is, the molecules constituting the sample) are excited to the first electron excited state by the resonance wavelength λ ₁ light such as laser light (S ₁ state: FIG. 2). The light is excited to the second electron excited state or a higher excited state by the resonance wavelength λ ₂ light (S ₂ state: FIG. 3). The molecules return to the ground state by emitting fluorescence or phosphorescence from the excited state (FIG. 4). Then, an absorption image and a light emission image are observed using the absorption shown in FIG. 2 and the emission of fluorescence and phosphorescence shown in FIG.

[0004] In the excitation process to S ₁ state, the number of molecules S ₁ state of the unit volume increases as the intensity of light irradiation is increased. Since the linear absorption coefficient is given by the product of the absorption cross section per molecule and the number of molecules per unit volume, during the excitation process to the S ₂ state, the linear absorption coefficient for the subsequently irradiated resonance wavelength λ ₂ is first It depends on the intensity of the irradiated resonance wavelength lambda ₁ of light. That is, the linear absorption coefficient for the resonance wavelength λ ₂ (hereinafter abbreviated as wavelength λ ₂ ) can be controlled by the intensity of light having the resonance wavelength λ ₁ (hereinafter abbreviated as wavelength λ ₁ ). This sample was irradiated with light of two wavelengths of the wavelength lambda ₁ and wavelength lambda _2, if taking a transmission image due to the wavelength lambda _2, the contrast of the transmission image indicates that full control with light having a wavelength lambda ₁ ing. When a de-excitation process by fluorescence or phosphorescence from the S ₂ state is possible, the emission intensity is proportional to the number of molecules in the S ₁ state. Therefore, it is possible to control the image contrast even when used as a fluorescence microscope.

The double resonance absorption microscope enables not only control of contrast but also chemical analysis. FIG.
Since the outermost grain valence orbit shown in ( ₁₎ has an energy level unique to each molecule, the wavelength λ ₁ differs depending on the molecule. At the same time, the wavelength λ _{2 is} also unique to the molecule. Even with a conventional microscope that illuminates and observes at a single wavelength, it is possible to observe an absorption image or a fluorescence image of a specific molecule to some extent, but in general, the wavelength region of the absorption band of some molecules overlaps. Therefore, accurate identification of the chemical composition of the sample is not possible. On the other hand, in the double resonance absorption microscope, molecules that absorb or emit light are limited by two wavelengths of the wavelength λ ₁ and the wavelength λ ₂ , so that the chemical composition of the sample can be identified more accurately than in the prior art. In the case of exciting valence electrons, only light having a specific electric field vector with respect to the molecular axis is absorbed strongly determined polarization direction of the wavelength lambda ₁ and wavelength lambda _2, taken absorption image or fluorescence image Then, the orientation of the same molecule can be identified.

The inventor of the present invention has also proposed a double resonance absorption microscope having a high spatial resolution exceeding the diffraction limit (see Japanese Patent Application No. 8-322232). In the double resonance absorption process, there are certain molecules whose fluorescence from the S ₂ state becomes extremely weak as illustrated in FIG. The following very interesting phenomena occur for molecules having such optical properties. FIG. 6 is a conceptual diagram of the double resonance absorption process as in FIG. 5, but the X axis is provided on the horizontal line to express the spread of the spatial distance, and the wavelength λ ₁ light and the wavelength λ ₂ light are irradiated. Spatial area A ₁ (= fluorescence suppression area)
And a spatial region A ₀ (= fluorescent region) where only the wavelength λ ₁ light is irradiated and the wavelength λ ₂ light is not irradiated. S by the excitation by the spatial region A ₀ at a wavelength lambda ₁ light ₁
Many molecules in the state are generated. At this time, fluorescence emitted at the wavelength λ ₃ is seen from the spatial region A ₀ . However, in spatial regions A _1, since the wavelength lambda ₂ light is irradiated, is excited and the molecules of S ₁ state to immediately higher level S ₂ state, molecules of S ₁ state is no longer present. Therefore, in the spatial region A _1,
Since no fluorescence of the wavelength λ ₃ is generated and the fluorescence from the S ₂ state is not originally present, the fluorescence itself is completely suppressed. That is only the spatial region A ₀ of fluorescence is generated. The occurrence of such a phenomenon has been confirmed in several types of molecules.

Therefore, in a conventional scanning laser microscope or the like, the size of a microbeam formed on an observation sample by condensing a laser beam is determined by the numerical aperture of the condenser lens and the diffraction limit by the wavelength. Nino contrast, the wavelength lambda ₁ light and the wavelength lambda ₂ light according to the phenomena shown by making spatially partially overlapping in Figure 6, the fluorescence region upon irradiation with wavelength lambda ₂ light spatial resolution is unable theoretically expected because There is inhibited, for example, when focusing on the irradiation area of the wavelength lambda ₁ light, fluorescent region is narrower than the size of the beam determined by the numerical aperture and the wavelength of the condenser lens, it is substantially improved in spatial resolution It is planned. The double resonance absorption microscope (see Japanese Patent Application No. 8-302232) by the present inventor has realized a super-resolution microscope exceeding the diffraction limit by using this principle.

The inventor of the present invention proposes that the double resonance
To further enhance the super-resolution of absorption microscopes,
Adjustment of sample and wavelength λ to make full use₁Light / wavelength λ_Twolight
We have already proposed the timing of irradiation of samples
(See Japanese Patent Application No. 9-255444). More specifically,
The material is stained with the staining molecules. As the dye molecule,
At least three quantum states including the bottom state (S₀State, S₁
State, S_TwoState) and S₁High except state
In the transition during deexcitation from the quantum state to the ground state
Thermal relaxation process is more dominant than fluorescence relaxation process
The use of various molecules (hereinafter referred to as fluorescent labeler molecules)
It is. Such fluorescent labeler molecules and biochemical staining
A sample that is chemically bonded to a biomolecule that has undergone technology
Wavelength λ ₁Irradiate the fluorescent labeler molecules with S₁Excited to state
And then immediately the wavelength λ_TwoFluorescent labeler by irradiating light
By exciting molecules to higher quantum levels, S₁Condition
Fluorescence from the state can be effectively suppressed. On this occasion
In addition, artificial suppression of the spatial fluorescent region as described above
By doing so, the spatial resolution can be further improved.
You.

The optical properties of the above fluorescent labeler molecule are as follows:
It can be explained from the quantum chemical point of view as follows. Generally, molecules are connected by σ or π bonds of each of the constituent atoms. In other words, the molecular orbit of the molecule is σ
It has a molecular orbital or a π molecular orbital, and electrons existing in these molecular orbitals play an important role of bonding each atom. Among them, the electron of the σ molecular orbit strongly binds each atom and determines the distance between atoms in the molecule, which is the skeleton of the molecule. On the other hand, electrons in the π molecular orbital make little contribution to the bonding of each atom, but rather are bound by an extremely weak force to the whole molecule.

In many cases, when electrons in the σ molecular orbit are excited by light, the atomic spacing of the molecule changes greatly, and a large structural change including dissociation of the molecule occurs. As a result,
Most of the energy that light gives to molecules to change the kinetic energy and structure of atoms changes into thermal energy. Therefore, the excitation energy is not consumed in the form of light called fluorescence. In addition, since the structural change of a molecule occurs at a very high speed (for example, shorter than a picosecond), even if fluorescence occurs in the process, the fluorescence lifetime is extremely short. However, even when electrons in the π molecular orbital are excited, the structure of the molecule itself hardly changes, stays at a high-order quantum discrete level for a long time, emits fluorescence in the order of nanoseconds, and escapes. Exciting.

According to quantum chemistry, the fact that a molecule has a π molecular orbit is equivalent to having a double bond, and it is necessary to select a molecule having abundant double bonds as a fluorescent labeler molecule to be used. It is a necessary condition. Also, it has been confirmed that even in a molecule having a double bond, fluorescence from the S2 state is extremely weak in a six-membered ring molecule such as benzene or virazine (for example, M .. Fujii et.al., Chem. Phys.). . Lett.
171 (1990) 341). Therefore, the fluorescence lifetime of the molecule containing a 6-membered ring molecule, such as benzene or pyrazine from the selection them if the state S1 as fluorescence probe molecule is prolonged, yet from a molecule by exciting the S ₁ state to S ₂ state by light irradiation Can be easily suppressed, and the super-resolution of the above-described double resonance absorption microscope can be effectively used.

That is, if a sample is stained with these fluorescent labeler molecules and observed, not only can a fluorescent image with high spatial resolution be obtained, but also the chemical groups of the side chains of the fluorescent labeler molecules can be adjusted. Thereby, only a specific chemical tissue of a biological sample can be selectively stained, and even a detailed chemical composition of the sample can be analyzed.

In general, the double resonance absorption process occurs only when the wavelengths and polarization states of two lights satisfy specific conditions, so that it is possible to know a very detailed molecular structure by using this. Become. That is, there is a strong correlation between the polarization direction of the light and the orientation direction of the molecules, and when the polarization direction of each of the two-wavelength light and the orientation direction of the molecules make a specific angle, the double resonance absorption process occurs strongly. Therefore, by irradiating the sample with two-wavelength light and rotating the polarization direction of each light, the degree of disappearance of the fluorescence changes, and from that state, information on the spatial orientation of the tissue to be observed can be obtained. This is also possible by adjusting the light of the two wavelengths.

On the other hand, by adjusting the irradiation timing of the wavelength λ ₁ light and the wavelength λ ₂ light to an appropriate timing (Japanese Patent Application No. 9-2980).
55544), it is possible to improve the S / N of the fluorescent image and more effectively suppress the fluorescence.

The inventors of the present invention have proposed that the wavelength λ ₁
According to a further contrivance irradiation timing of the light and the wavelength lambda ₂ light, also it proposed are (see Japanese Patent Application No. 10-97924) to realize a further improvement in S / N and fluorescence suppression Incidentally, the wavelength lambda ₁ light above The superposition of the irradiation area of the wavelength λ ₂ light on a part of the irradiation area is performed by converting the wavelength λ ₂ light into a hollow beam, that is, the central part (area near the axis) has zero intensity, and has an axis-symmetric intensity distribution. This can be realized by forming a hollow beam light, spatially overlapping a part of the wavelength λ ₁ light, and condensing it on a sample. FIG. 7 is a conceptual diagram illustrating the polymerization and the suppression of fluorescence by the polymerization. The superposition of the hollow beam of the wavelength lambda ₂ light, other than the region near the optical axis of the intensity of the wavelength lambda ₂ light is zero fluorescence is suppressed, fluorescence present in an area smaller than the extent of the wavelength lambda ₁ light labeler Only the fluorescence of the molecule is observed. As a result, super-resolution is developed. In the case of a hollow beam of λ ₂ light,
For example, as illustrated in FIG. 8, a phase plate that gives the passing light a phase difference π that is point-symmetric with respect to the optical axis can be used.

[0016]

As described above, the double resonance absorption microscope that has been developed by the inventor of the present invention as described above has outstanding utility and technical advantages in its super-resolution and analytical ability. In spite of its superiority, the fact remains that the following points need to be improved. Hereinafter, light that excites sample molecules (= molecules constituting the sample) from the S ₀ state to the S ₁ state is referred to as pump light, and light that excites molecules in the S ₁ state to the S ₂ state is referred to as erase light. And When a sample is stained using a fluorescent labeler molecule, the sample molecule is the fluorescent labeler molecule.

First, depending on the molecule to be excited,
The wavelength band of the fluorescence from the molecule and the wavelengths of the erase light and the pump light that excite the molecule may be close to or overlap with each other, so that when the fluorescence signal is detected, the excitation light becomes the background light, In some cases, it was difficult to extract a fluorescence signal to be measured originally. In particular, erasing light causes molecules to be S
_Since it is necessary to excite from the ₁ state to the S ₂ state, the intensity is relatively strong, and its influence needs to be particularly considered. For example, in the prior example (Japanese Patent Application No. 10-97924), when a sample is stained with a fluorescent labeler molecule, assuming that rhodamine 6G (Rhordamin-6G) is used as the fluorescent labeler molecule,
As illustrated in FIG. 9, the fluorescence band of the rhodamine 6G extends in a wavelength region from about 530 nm to 650 nm, and the wavelength of the pump light for the rhodamine 6G is 532n.
m, since the wavelength of the erase light is 599 nm, the fluorescence band and the excitation wavelength overlap. For this reason, there is a problem that the S / N of the fluorescent image is not good.

Secondly, realizing depth resolution in the direction of the optical axis, that is, three-dimensional spatial resolution, has become a major issue in recent years in improving the performance of microscopes. Similarly, the double resonance absorption microscope does not have a depth resolution in the optical axis direction. As described above, the improvement of the spatial resolution when the irradiation areas of the pump light and the erase light are partially overlapped is only for the planar resolution. Fluorescence is suppressed in the beam peripheral region of the pump light where the erase light overlaps, but the fluorescence is not suppressed at all on the optical axis of the hollow portion of the erase light, and molecules on the optical axis emit light. That is, in principle, there is no depth resolution in the optical axis direction. In order to have depth resolution, for example, it is possible to install a pinhole on the entire surface of the detector and at the confocal position. However, in practice, the alignment of the imaging optical system including the pinhole becomes complicated, There is a problem that the number of fluorescence photons reaching the detector is reduced. Therefore, it is possible to realize an extremely high-performance microscope that has never had a plane resolution and a depth resolution without complicating the configuration of the imaging optical system and the like.

The invention of this application has been made in view of the circumstances described above, and is an improvement of the double resonance absorption microscope, which has been developed by combining the background light (excitation light) with the fluorescence signal.
It is an object of the present invention to provide a new dual resonance absorption microscope that can suppress deterioration of / N and has excellent three-dimensional spatial resolution.

[0020]

Means for Solving the Problems The present invention solves the above-mentioned problems by providing a pump light source for exciting a sample molecule from a ground state to a first electronically excited state, An erase light source for exciting the sample molecules to a second electronic excited state or a higher excited state,
In a double resonance absorption microscope that detects light emission when sample molecules in each excited state are de-excited to the ground state, pump light
That the sample molecule in the ground state has a photon energy of １ ／ or less of the excitation energy to excite the sample molecule in the first electronic excited state; Having a photon energy that is less than or equal to half of the excitation energy to excite the state,
A double resonance absorption microscope (Claims 1 to 3) is provided, wherein each of the pump light and the erase light has the photon energy.

Further, the invention of this application is based on the double resonance absorption microscope, wherein the photon energy band of light emission from the sample molecule in the first electronically excited state and the sample molecule in the first electronically excited state are converted into the second electronically excited state. Alternatively, it is provided as an aspect that the excitation energy band for exciting to a higher excited state overlaps (claim 4).

Further, in the above double resonance absorption microscope, a superimposing means for partially superimposing the irradiation areas of the pump light and the erase light is provided so as to have super-resolution in a plane direction, and the superimposing means is provided through the superimposing means. By irradiating the sample with the pump light and the erase light, the light emission region when the sample molecules in the first electronically excited state are de-excited to the ground state is partially suppressed (Claim 5). In order to make the realization more effective, the sample is stained with a fluorescent labeler molecule having at least three electronic states including a ground state, and the sample molecule is the fluorescent labeler molecule. ) Is provided as an embodiment thereof.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION The invention of this application is as described above,
In the double resonance absorption microscope, the pump light has photon energy that is １ ／ or less of the excitation energy for exciting the sample molecule (or fluorescent labeler molecule) in the S ₀ state to the S ₁ state, light, are characterized by having a photon energy is less than 1/2 of the excitation energy to excite the sample molecules in S ₁ state (or fluorescence probe molecule) to S ₂ state. That is, the double resonance absorption microscope of the invention of this application utilizes a non-resonant multiphoton excitation process which is one of the non-resonant optical effects.

Here, the non-resonant multiphoton excitation process in the case of two photons (referred to as a non-resonant two-photon excitation process) will be described. The non-resonant two-photon excitation process refers to a material having two energy levels (= electronically excited states: S ₀ and S ₁ ) as illustrated in FIG. when was the E _01, when the light having a high photon flux having a photon energy E _01/2 of half the E ₀₁ to the substance, is that the S ₀ effect of allowing the transition to S ₁ .

According to the perturbation calculation up to the second order in quantum mechanics,
Transition probability a from S ₀ to S ₁ by non-resonant two-photon excitation process
(S ₀ → S ₁ ) is given by the following equation.

[0026]

(Equation 1)

In equation (1), Fsin (ωt) = amplitude of excitation light−er = electric dipole moment of sample molecule ω = angular frequency of excitation light h = Planck constant t = time t ′, t ″ = t Integral variables ω _j0 , ω _1j = resonance frequency of the sample molecule _{仮 定 j1} , _χ0j = matrix element of electric dipole moment between S ₁ and S ₂ assuming a virtual quantum state of a certain j.

Χ _j1 and _0j are given by the following equations.

[0029]

(Equation 2)

[0030] In this number 2 is the wave function of u _k, u _m = final state k and the initial state m of molecules involved in the transition.

On the other hand, in the case of a normal one-photon excitation process, a transition process is performed in which excitation is performed without passing through a virtual intermediate state.
The transition probability is given by the following equation using first-order perturbation theory.

[0032]

(Equation 3)

As is clear from the comparison between Equations 1 and 3,
The transition probability from S ₀ to S ₁ is proportional to the fourth power of the amplitude of the excitation light in the case of a non-resonant two-photon excitation process (in other words,
It is proportional to the square of the intensity of the excitation light, and in the case of the one-photon excitation process, it is proportional to the square of the amplitude intensity of the excitation light (in other words, linearly proportional to the intensity of the excitation light). That is, in the case of the non-resonant two-photon excitation process, the fluorescence intensity is substantially proportional to the square of the intensity of the excitation light, and in the case of the one-photon excitation process, the fluorescence intensity is also linearly proportional.

Therefore, such a non-resonant two-photon excitation process is combined with a double resonance absorption process, ie, S
Non-resonant two-photon excitation for excitation from the ₀ state to the S ₁ state (hereinafter abbreviated as S ₀ → S ₁ excitation) and excitation from the S ₁ state to the S ₂ state (hereinafter abbreviated as S ₁ → S ₂ excitation) If the process is used, an extremely excellent S / N fluorescence signal can be obtained.

This principle will be described with reference to FIGS. FIG. 11 conceptually illustrates a combination of a double resonance absorption process and a non-resonant two-photon excitation process together with an energy diagram of a fluorescent labeler molecule. FIG. 12 illustrates the relationship between the photon energy of the excitation light and the absorption / fluorescence intensity spectrum of the sample molecule. First, as exemplified in FIG. 11, the photon energy E _01/2 , which is half the energy required for ordinary S ₀ → S ₁ excitation (in terms of wavelength, is twice that of ordinary S ₀ → S ₁ excitation). length of: 2 [lambda] ₁₎ a pump light with a virtually level from S ₀ S _V0
Excitation to S ₁ via Continue with normal S ₁ → S
₂ Photon energy E ₁₂ of half the energy required for excitation
/ 2 (in terms of wavelength, twice the length of ordinary S ₁ → S ₂ pumping: 2λ ₂ ), and is an erase light from S ₁ to S ₂ via virtual level S _V1 . Excitation is performed. In addition, E ₁₂ /
2 of the erase light after the irradiation of the pump light E _01/2, S ₁
The sample is illuminated before any of the molecules fluoresce.

For example, as shown in FIG. 12, in the case of only the double resonance absorption process, the wavelength λ ₁
The photon energy E ₀₁ of the pump light having the wavelength belongs to the S ₀ → S ₁ absorption band, and the photon energy E ₁₂ of the erase light having the wavelength λ ₂ belongs to both the S ₁ → S ₂ absorption band and the fluorescence band. are, but contrast, the photon energy E _12/2 of the erase light having photon energy E _01/2 and wavelength 2 [lambda] ₂ of the pump light of a wavelength 2 [lambda] ₁ is, S ₀ → S ₁ absorption band, S ₁
→ Does not belong to either the S ₂ absorption band or the fluorescence band.

[0037] Thus, in the pump light of wavelength 2 [lambda] ₁ · photon energy E _01/2 to generate a sample molecules of S ₁ state (or fluorescence probe molecule), wavelength 2 [lambda] ₂ before the molecules of the S ₁ state fluoresces and photon energy E ₁₂ / by the second erase light by exciting the S ₂ state, the excitation wavelength is present in much lower energy region than fluorescence band,
Since there is no overlap with the fluorescent band, it is possible to obtain an extremely good S / N fluorescent signal in which the excitation light is not mixed as the background light.

A non-resonant two-photon excitation process is used for either S ₀ → S ₁ excitation or S ₁ → S ₂ excitation, that is, one of the pump light and the erase light is reduced to の of the excitation energy. It may have photon energy. When the normal S ₀ → S ₁ excitation wavelength and the fluorescence band overlap,
This is effective when one of the cases where the normal S ₁ → S ₂ excitation wavelength and the fluorescence band overlap occurs.

Further, the above description is based on the case of two photons.
About the nonresonant multiphoton excitation process of
Even using a non-resonant multiphoton excitation process of three or more photons,
In the double resonance absorption microscope of the present invention, the mixing of the excitation light is extremely small.
Can be effectively prevented. nth-order nonresonant multiphoton excitation
In the starting process, the pump light and the erase light ₀
→ S₁Excitation, S₁→ S_Two1 of excitation energy required for excitation
/ N photon energy. For example, the non-shared
In the three-photon excitation process, pump light and erase light
The electron energy may be 1/3 of the excitation energy.

In many fluorescent labeler molecules, the photon energy used in the non-resonant multiphoton excitation process is small, there is no so-called ordinary electronic excitation level other than the virtual level, and it is transparent to pump light and erase light. become. By utilizing this property, a super-resolution microscope with extremely excellent performance can be realized.

Further, by using this non-resonant multiphoton excitation process, it is possible to impart a depth resolution in the optical axis direction, that is, a three-dimensional spatial resolution, to the double resonance absorption microscope. Here, for simplification of the description, the description will be made based on the non-resonant two-photon excitation process.

The wavelength lambda ₁ light as excitation light, when assuming an optical system aplanatic as a condenser lens, an optical system opening (numerical aperture: NA) when collecting light of wavelengths lambda ₁ light is full, the image space
The three-dimensional complex amplitude distribution I (x, y, z) at (x, y, z) is given by the following equation.

[0043]

(Equation 4)

Where f is the focal length, and (ξ, ζ) are the coordinates of the pupil plane. The integration area is a full aperture.

According to the above equation (3), light of wavelength λ ₁ is irradiated,
When a sample molecule is excited to the S ₁ state in a normal one-photon excitation process, the excited molecular weight qone (x, y, z) is proportional to I (x, y, z) ² . On the other hand, according to Equation 4, when the sample molecules are excited by the non-resonant two-photon excitation process with the wavelength 2λ ₁ light, the excited molecular weight qtwo (x, y, z) is proportional to I (x, y, z) ⁴ . . When the qone (x, y, z) and qtwo (x, y, z) are normalized and Fourier-transformed into a frequency space, a physical quantity (three-dimensional OTF) indicating a resolution that can be actually observed is obtained. That is, if the normalized excited molecular weight in the one-photon excitation process and the excited molecular weight in the non-resonant two-photon excitation process are written as Qone (x, y, z) and Qtwo (x, y, z),
And the relationship as shown in Equation 6.

[0046]

(Equation 5)

[0047]

(Equation 6)

Here, k1 _x , k1 _y , and k1 _z are wave vector components for the excitation light of wavelength λ ₁ , and k2 _x , k1
2 _y and k 2 _z are wave number vector components for the excitation light of wavelength 2λ ₁ .

Assuming that the pupil plane is symmetric with respect to the optical axis, r = (x ²
+ X ² ) ^1/2 is introduced, and the cross sections of the spatial distributions of (k1 _r , k1 _z ) and (k2 _r , k2 _z ) are obtained by using Equations 5 and 6, and FIG. As shown in (b). k1 _r and k2 _r are wave number vector components of the wavelength λ ₁ light and the wavelength 2λ ₁ light in the radial direction from the optical axis, respectively.

According to FIGS. 13A and 13B, in both the non-resonant two-photon excitation process and the one-photon excitation process, the upper limit Kmax and the lower limit Kmin of the spatial frequency band in the radial direction use the numerical aperture NA of the optical system. hand,

[0051]

(Equation 7)

The spatial resolution is almost the same.
However, there is a remarkable difference in the resolution in the optical axis direction.
In the one-photon excitation process, there is no value in the k1 _z- axis direction including the origin. In other words, this indicates that there is no resolution in the optical axis direction. On the other hand, in the non-resonant two-photon excitation process, the k2 _z- axis direction,

[0053]

(Equation 8)

It can be seen that it has the value That is, it has a depth resolution in the optical axis direction. This is because the amount of excited molecules in the S ₁ state is equal to 2
This is proportional to the power and corresponds to the fact that the excited molecule is localized in a very narrow space region.

[0055] Thus, by using the non-resonant two-photon excitation process in generating the excited molecules of S ₁ state, i.e. 1/2 in which photon energy of the excitation energy to excite the sample molecules in the ground state to the S ₁ state By using the pump light having the above, depth resolution in the optical axis direction can be added to the double resonance absorption microscope.

As described above, the double resonance absorption microscope is provided with superimposing means (for example, an optical system including a phase plate for converting the erase light into a hollow beam) for partially overlapping the irradiation area of the pump light and the erase light. By irradiating the sample with the pump light and the erase light through the light source, the light emission region when the sample molecules in the S ₁ state are de-excited to the ground state is partially suppressed.
An extremely high-performance super-resolution microscope that can have depth resolution can be realized.

[0057] Of course, more than that is the same for the erase light for the S ₁ → S ₂ excitation, usually the S ₁ → S ₂
By setting the photon energy to 励 起 of the excitation energy required for excitation, a three-dimensional spatial resolution is realized. Even when a non-resonant multiphoton excitation process of three or more photons is used, three-dimensional spatial resolution can be similarly achieved by setting the photon energies of the pump light and the erase light to 1 / n.

Hereinafter, among the fluorescent labeler molecules that stain the sample, those in which the excitation light wavelength and the fluorescence wavelength in the one-photon excitation process are close to or overlap with each other will be described. When the sample is stained using these fluorescent labeler molecules, the above-described effect of the invention of the present application caused by the non-resonant two-photon excitation process, that is, the improvement of the S / N of the fluorescent signal free of background light is improved. Will be more prominent.

[Example of fluorescent labeler molecule] 2,2 "-Dimethyl-p-terphenyl: P-terphenyl (PTP): 3,3 ', 2", 3 "-Tetramethyl-P-quaterphenyl: 2,2'''-Dimethyl-P-quaterphenyl: 2-Methyl-5-t-butyl-p-quaterphenyl: 2- (4-Biphenylyl) -5- (4-t-butylphenyl) -1,3,4-oxiazol
(BPBD-365): 2- (4-Biphenylyl) -phenyl-1,3,4-oxadiazol: 2,5,2 '''' 5, ''''-Tetramethyl-p-quinquephenyl: 3,5, 3 '''' 5, ''''-Tetra-t-butyl-p-quinquephenyl: 2,5-Diphenyloxazol: 2,5-Diphenylfuran: PQP (p-Quanterphenyl: 2,5-Bis- (4-biphenylyl ) -1,3,4-oxadiazol: p-Quaterphenyl-4-4 '''-disulfonicacid Disodiumsalt: p-Quaterphenyl-4-4'''-disulfonicacid Dipotassiumsa
lt: 4,4 '''-Bis- (2-butyloctyloxy) -p-quanterphenyl: 3,5,3''''5,''''-Tetra-t-butyl-p-sexiphenyl: 2- ( 1-Naphthyl) -5-phenyloxazol: 2- (4-Biphenylyl) -6-phenylbenzoxazotetrasulfonicaci
d Potassium Salt: 2- (4-Biphenylyl) -6-phenylbenzoxazol-1,3: 4,4'-Diphenylstilbene: [1,1'-Biphenyl] -4-sulfonic acid, 4,4 "-1,2- ethene-d
iylbis-, dipotassium salt: 2,5-Bis- (4-biphenylyl) -oxazol: 2,2 '-([1,1'-Biphenyl] -4,4'-diyldi-2,1-ethenediyl)-
bis-benzenesulfonic acid Disodium Salt: 7-Amino-4-methylcarbostyryl: 1,4-Di [2- (5-phenyloxazoly)] benzene: 7-Hydroxy-4-methylcoumarin: p-Bis (o-methylstyryl) -benzene: Benzofuran, 2,2 '-[1,1'-biphenyl] -4,4'-diyl-bis-tetr
asulfonic-acid: 7-Dimethylamino-4-methylquinolom-2: 7-Amino-4-methylcoumarin: 2- (p-Dimethylaminostyryl) -pyridylmethyl Iodide: 7-Diethylaminocoumarun: 7-Diethylamino-4-methylcoumarin: 2,3,5 , 6-1H, 4H-Tetrahydro-8-methylqinolizino- [9,9a,
1-gh] -coumarin: 7-Diethylamino-4-trifluormethylcoumarin: 7-Dimethylamino-4-trifluormethylcoumarin: 7-Amino-4-trifluormethylcoumarin: 2,3,5,6-1H, 4H-Tetrahydroquinolizino- [9,9a, 1-gh] -co
umarin: 7-Ethylamino-6-methyl-4-trifluormethylcoumarin: 7-Ethylamino-4-trifluormethylcoumarin: 2,3,5,6-1H, 4H-Tetrahydro-9-carboethoxyquinolizino-
-[9,9a, 1-gh] coumarin: 2,3,5,6-1H, 4H-Tetrahydro-9- (3-pyridyl) -quinolizino
-[9,9a, 1-gh] coumarin: 3- (2'-N-Methylbenzimidazolyl) -7-n, n-diethylaminoco
umarin: 2,3,5,6-1H, 4H-Tetrahydro-9-acetylquinolizino- [9,9
a, 1-gh] -coumarin: N-Methyl-4-trifluormethylpiperidino- [3,2-g] -coumar
in: 2- (p-Dimethylaminostyryl) -benzothiazolylethyl Iodi
de: 3- (2'-Benzimidazolyl) -7-N, N-diethylaminocoumarin: Brillantsulfaflavin: 3- (2'-Benzothiazolyl) -7-diethylaminocoumarin: 2,3,5,6-1H, 4H-Tetrahydro-8- trifluormethylquinolizi
no- [9,9a, 1gh] coumarin: 3,3'-Diethyloxacarbocyanine Iodide: 3,3'-Dimethyl-9-ethylthiacarbocyanine Iodide: Disodium Fluorescein (Uranin): 9- (o-Carboxyphenyl) -2,7-dichloro -6-hydroxy-3H-xant
hen-3-on2,7-Dichlorofluorescien ・ Fluorescein 548: Fluorol 555 (Fluorol 7GA): o- (6-Amino-3-imino-3H-xanthen-9-yl) -benzonic acid
(Rhodamine 560): BenzoicAcid, 2- [6- (ethylamino) -3- (ethylimino) -2,7-d
imethyl-3H-xanthen9-yl], perchlorate (Rhodamine 57
5): Benzonic Acid, 2- [6- (ethylamino) -3- (ethylimino) -2,
7-dimethyl-3X-xanthen-9-yl] -ethylester, monohydroch
loride (Rhodamine 590): 1,3'-Diethyl-4,2'-quinolyloxacarbocyanine Iodide: 1,1'-Diethyl-2,2'-carbocyanine Iodid: 2- [6- (diethylamino) -3- (diethylimino) -3H-xanthen-9-
yl] benzonic acid (Rhodamine 610): Ethanaminium, N-[(6-diethylamino) -9- (2,4-disulfophe
nyl) -3H-xanthen-3-ylidene] -N-ethylhydroxid, inner
salt, sodium salt: Malachit Green: 3,3'-Diethylthiacarbocyanine Iodide: 1,3'-Diethyl-4,2'-quinolylthiacarbocyanine Iodide: 8- (2-Carboxyphenyl) -2,3,5,6,11,12 , 14,15-octahydro-
1H, 4H, 10H, 13H-diquinolizino [9,9a, 1-bc: 9 ', 9a', 1-hi] xanthylium Perchlorate (Rhodamine640): 4-Dicyanmethylene-2-methyl-6- (p-dimethylaminostyry
l) -4H-pyran: 3,3'-Diethyloxadicarbocyanine Iodide: 8- (2,4-Disulfophenyl) -2,3,5,6,11,12,14,15-octahydr
o-1H, 4H, 10H, 13H-diquinolizino [9,9a, 1-bc: 9 ', 1-hi] x
anthene (Sulforhodamine 640): 5,9-Diaminobenzo [a] phenoxazonium Perchorate: 9-Diethylamino-5H-benzo (a) phenoxazin-5-one: 5-Amino-9-diethylimino (a) phenoxazonium Perchlorat
e: 3-Ethylamino-7-ethylimino-2,8-dimethylphenoxazin-5
-ium Perchorate: 8- (Trifluoromethyl) -2,3,5,6,11,12,14,15-octahydro-
1H, 4H, 10H, 13H-diquinolizino [9,9a, 1-bc: 9 ', 9a, 1-hi] x
anthylium Perchlorate: 1-Ethyl-2- (4- (p-Dimethylaminophenyl) -1,3-butadieny
l) -pyridinium Perchlorate: Carbazine 122: 9-Ethylamino-5-ethylimino-10-methyl-5H-benzo (a) phe
noxazoniumPerchlorate: 3-Diethylamino-7-diethyliminophenoxazonium Perchlo
rate: 3-Diethylthiadicarbocyanine Iodide: Oxazine 750: 1-Ethyl-4- (4- (p-Dimethylaminophenyl) -1,3-butadieny
l) -pyridininum Perchlorate: 1,1 ', 3,3,3', 3'-Hexamethylindodicarbocyanine Iodid
e: 1,1'-Diethyl-4,4'-carbocyanine Iodide: 2- (4- (p-Dimethylaminophenyl) -1,3-butadienyl) -1,3,3
-trimethyl-3H-indoliumPerchlorate: 2- (4- (p-Dimethylaminophenyl) -1,3-butadienyl) -3-eth
ylbenzothoazolium Perchlorate: 1,1'-Diethyl-2,2'-dicarbocyanine Iodide: 1-Ethyl-4- (4- (9- (2,3,6,7-tetrahydro-1H, 5H-benzo (i,
j) -chinolinozinium))-1,3-buta dienyl) -pyridinium P
erchlorate: 3,3'-DimethyloxatricarbocyanineIodide: 1-Ethyl-4- (p-Dimethylaminophenyl) -1,3-buta dienyl)
-quinolinium Perchlorate: 8-Cyano-2,3,5,6,11,12,14,15-octahydro-1H, 4H, 10H, 13
H-diquinolizino [9,9a, 1-bc: 9a ', 1-hi] xanthylium Perc
hlorate (Rhodamine800): 2- (6- (4-Dimethylaminophenyl) -2,4-neopentylene-1,3,
5) -3-methylbenzothiazoliumPerchlorate: 1,1 ', 3,3,3', 3'-Hexamethylindotricarbocyanine Iodid
e: IR125: 3,3'-Diethylthiatricarbocyanine Iodide: IR144: 2- (6- (9- (2,3,6,7, -Tetrahydro-1H, 5H-benzo (i, j) -chin
olizinium))-2,4-neopentylene-1,3,5-hexatrienyl) -3-
methyllbenzothiazolium Perchlorate: 3,3'-Diethyl-9,11-neopentylenethiatricarbocyanine
Iodide: 1,1 ', 3,3,3', 3'-Hexamethyl-4,4 ', 5,5'-dibenzo-2,2'-i
ndotricarbocyanine Iodide: 3,3'-Diethyl-4,4 ', 5,5'-dibenzothiatricarbocyanine
Iodide: 1,2'-Diethyl-4,4'-dicarbocyanine Iodide: IR140: 2- (8- (4-p-Dimethyhlaminophenyl) -2,4-neopentylene-
1,3,5,7-octatetraenyl) -3-methylbenzothiazolium Per
chlorate: IR132: 2- (8- (9- (2,3,6,7-Tetrahydro-1H, 5H-benzo (i, j) chinol
izinium))-2,4-neopentylene-1,3,5,7-octatetraenyl)-
3-Methylbenzothiazolium Perchlorate: IR26: IR5 Hereinafter, examples will be described with reference to the accompanying drawings, and embodiments of the present invention will be described in further detail.

[0060]

[Embodiment 1] FIG. 14 is a schematic diagram of a configuration of a main part showing an example of a double resonance absorption microscope according to the present invention. The double resonance absorption microscope illustrated in FIG. 14 constitutes a laser scanning fluorescence microscope, and uses not only the double resonance absorption process and the non-resonant multiphoton excitation process but also the irradiation region of the pump light and the erase light. This is an embodiment in which partial polymerization can be performed. With respect to such a double resonance absorption microscope of FIG. 14, [A] a light source unit, [B] a sample observation system,
[C] The detection system will be described separately below. Note that the non-resonant multiphoton excitation process uses a non-resonant two-photon excitation process, and the sample (10) has rhodamine 6 as a fluorescent labeler molecule.
Assume that the sample is a biological sample stained with G.

[A] Light Source Section An Nd: YAG laser (1) emitting a picosecond pulse laser beam using a supersaturated dye is used as a basic light source, and its fundamental wave (wavelength 1064 nm, photon energy about 1.16) is used.
5eV) to generate pump light and erase light. Here, the photon (electron) energy eV = 1239.
8 / wavelength nm is used as the conversion formula.

<1> Pump Light First, the pump light will be described. To excite Rhodamine 6G from S ₀ → S ₁ in the usual one-photon excitation process, about 2.
Photon energy E ₀₁ of 3eV is required. In Rhodamine 6G, the S ₀ → S ₁ absorption band starts at about 550 nm, that is, about 2.25 eV in terms of photon energy (see FIG. 9). Therefore, the wavelength 2λ ₁ = about 1064n
m · photon energy E _01/2 = Nd having about 1.165EV: a fundamental wave of a YAG laser (1) by the pump light, S ₀ → S ₁ pumped rhodamine 6G in a non-resonant two-photon excitation process Can be done.

[0063] required intensity of the pump light having such photon energy E _01/2 · Wavelength 2 [lambda] _1, that is rhodamine 6
The intensity required to excite G from S _{0 to} S ₁ in the non-resonant two-photon excitation process will be described. First, in Equation 1,
The fonton flux of the pump light having a wavelength of 2λ ₁ is represented by I (2
λ ₁ ), the number of fluorescent labeler molecules that excite S ₀ → S ₁ in a non-resonant two-photon excitation process within a unit time is:

[0064]

(Equation 9)

Is given by In Equation 9, n ₀ (t)
And n ₁ (t) are the number of fluorescent labeler molecules per unit volume in the S ₀ state and the S ₁ state at the elapsed time t from the irradiation of the laser beam. Here, the physical quantity σ
₀₁ (2λ ₁ ) is defined as Equation 10 below, and is defined as the absorption cross section in the non-resonant two-photon excitation process.

[0066]

(Equation 10)

The number of fluorescent labeler molecules per unit volume is expressed as N
_{Assuming 0} , the differential equation of Equation 9 is

[0068]

[Equation 11]

Can be solved.

Here, if the following expression 12 is satisfied,
Most of the fluorescent labeler molecules are excited to the S ₁ state, and the amount of light required for fluorescence observation can be obtained.

[0071]

(Equation 12)

For example, if the pulse width t of the laser beam from the Nd: YAG laser (1) is about 30 psec,
Since the absorption cross section σ ₀₁ (2λ ₁ ) of the rhodamine 6G during the non-resonant two-photon excitation process is about 10 ⁻⁵⁰ cm ⁴ sec / photon or less, the S of rhodamine 6G during the non-resonant two-photon excitation process is reduced.
₀ → The fonton flux I (2λ ₁ ) of the laser beam required for S ₁ excitation is about 10 ³⁰ photon / cm ² / sec or less. N
d: fundamental wave from YAG laser (1) (pulse width t =
(About 30 psec), the laser intensity per pulse when condensed to a submicron size is 10 nJ / pulse or less. This is energy that can be sufficiently covered by a commercially available Nd: YAG laser. Therefore, there is an advantage that an inexpensive and easily available laser can be used by the non-resonant two-photon excitation process.

<2> Erase Light Next, the erase light will be described. In order to excite Rhodamine 6G from S _{1 to} S ₂ in the ordinary one-photon excitation process, a photon energy E _{12 of} about 1.92 eV is required. In Rhodamine 6G, the S ₁ → S ₂ absorption band starts at about 640 nm, that is, 1.94 eV in terms of photon energy. Therefore, Nd: a fundamental wave of a YAG laser (1) and wavelength 2 [lambda] ₂ = about 1197Nm · photon energy E _12/2 = about 1.036eV to wavelength conversion, by using it as the erase light, Rhodamine 6G S ₁ → S ₂ can be excited in the non-resonant two-photon excitation process.

The generation of the erase light from the Nd: YAG laser (1) is described, for example, in Japanese Patent Application No. 10-97924.
Raman shifter made of Ba (NO ₃ ) ₂ crystal (3)
Can be performed. When a laser beam is passed through a Raman shifter (3) made of Ba (NO ₃ ) ₂ crystal, the wave number becomes 1
Stokes light shifted to the low energy side by 045 cm ^-1 is generated. This Stokes light is completely coherent and has a wavelength of 1 from the Nd: YAG laser (1).
When a fundamental wave of 064 nm is incident, the wavelength is 1197.4.
nm primary Stokes light is generated. Rhodamine 6G is generated by the non-resonant two-photon excitation process using this primary Stokes light.
Is excited by S ₁ → S ₂ , a transition between levels having a gap of about 599 nm (photon energy: 2.07 eV) is possible. Therefore, for Rhodamine 6G, this primary Stokes light of 1197 nm can be used as erase light.

Subsequently, the photon energy E ₁₂ /
2. The required intensity of the erase light having a wavelength of 2λ ₂ , that is, the intensity required to excite Rhodamine 6G from S _{1 to} S ₂ in the non-resonant two-photon excitation process will be described.

First, irradiation of pump light and erase light
The measurement of the fluorescence signal is performed, for example, using the timing shown in FIG.
It should be done in a group. The timing illustrated in FIG.
Time-resolved measurement disclosed in Japanese Patent Application No. 10-97924.
Illumination and measurement timing used in the method
is there. This time-resolved measurement method uses pump light and erase
Light pulse width t_p, T_eIs the fluorescence lifetime of the fluorescent labeler molecule
This is a particularly effective method when it is shorter than τ. First, the pulse
Width t_pFluorescent labeler molecules with S pump light₀→ S ₁Excited
Immediately after the end of the pump light irradiation, the pulse width t_eIre
Irradiates the fluorescent labeler molecule with S₁→ S_TwoExcited
You. At this time, in this embodiment, as described above, the pump light
And partially overlap the erased light irradiation area,
Suppresses fluorescence in areas that are spatially unnecessary (see Fig. 6).
See). Then, the fluorescence suppression area A₁Fluorescent area A other than₀From
Of the fluorescent labeler molecule from the end of the erase light irradiation
Time t to end of light_gFluorescence signal is detected
You.

[0077] When assuming such radiation and measurement timing in accordance with the number 11, the number of molecules of S ₁ state produced upon irradiation of the pump light by the time t _p n ₁ (t
Assuming that p), the number of molecules in the S ₁ state when the erase light is irradiated for the time t _e is given by the following differential equation.

[0078]

(Equation 13)

In Equation 13, a ₂ (S ₁ → S ₂ ) and a _2
2 (f) is S ₁ → S due to the non-resonant two-photon excitation process.
The transition probability and stimulated emission probability of ₂ . Further, the cross sections of σ ₁₂ (2λ ₂ ) and σ _f (2λ ₂ ) are defined for the transition probability and the stimulated emission probability.

[0080]

[Equation 14]

Is obtained. In Equation 14, n₁(T_e)
Is the fluorescence suppression area A₁In unit volume at₁Firefly in state
The number of light labeler molecules. The fluorescent labeler molecule
₁(T _e) In proportion to the fluorescence suppression area A₁Emits fluorescence from
I will. Therefore, in order to increase super-resolution,
Area A₁N₁(T_e) Must be as small as possible. And
For example, the fluorescent region A₀Fluorescent labeler at (= observation area)
-Number of molecules n₁(T_eIt is better to keep it below 10% of
Good. In this case, from Equation 14,

[0082]

(Equation 15)

Is obtained and further organized,

[0084]

(Equation 16)

Is obtained. In the case of Rhodamine 6G, at a wavelength of 599 nm, σ ₁₂ (2λ ₂ ) + σ _f (2λ ₂ ) is 10 ^−50.
cm ⁴ in the order of sec / photon, the pulse width t _e of the erase light is calculated from the number 16 as 30 psec, font emission flux I (2 [lambda] ₂₎ required for the erase light likewise the pump light, 10 ³⁰ photon / cm ² / sec.

[0086] Thus, generation of photon energy E _01/2 · Wavelength 2 [lambda] ₁ pump light and the photon energy E _12/2 · Wavelength 2 [lambda] ₂ erase light having a required strength, a single of N
d: It becomes possible with the YAG laser (1). In the example of FIG. 14, the fundamental wave from one Nd: YAG laser (1) is split by a half mirror (2), the transmitted light is used as pump light, and the reflected light is Raman shifter (3) as described above. )
, And is configured to be used as erase light after wavelength conversion.

The erase light thus obtained is incident on a dichroic mirror (7) as a beam combiner by reflection mirrors (4) and (5) provided behind the Raman shifter (3). Reflection mirror (4)
As (5), by using the one whose surface is coated with an interference film, an unnecessary fundamental wave of the Nd: YAG laser (1) which is not wavelength-converted by the Raman shifter (3) may be cut. Good.

<3> Superposition of Pump Light and Erase Light In this embodiment, on the optical path to the dichroic mirror (7), a phase plate (6) for converting the erase light into a hollow beam is provided. I have. As the phase plate (6), for example, a phase plate having a structure as shown in FIG. 16 can be used. The phase plate (6) illustrated in FIG.
FIG. 8 showing a function of rotating the phase around the optical axis and shifting the phase by 2π.
Is divided into four equal parts around the optical axis, each area is shifted by π / 2, and the phase is shifted by π at a point symmetrical with respect to the optical axis. . By passing through the phase plate (6), the erase light becomes a hollow beam having zero intensity at the center and having an axially symmetric intensity distribution.

The erase light in the form of a hollow beam is combined by the dichroic mirror (7) so as to travel on the same optical axis as the pump light. As a result, the hollow beam erase light overlaps a part of the pump light, and the fluorescent region A ₀ and the fluorescent suppression region A ₁ are generated.

Therefore, in this embodiment, the overlapping means for partially overlapping the irradiation areas of the pump light and the erase light is a phase plate (6) or a combination of the phase plate (6) and the dichroic mirror (7). I can say.

[B] Sample Observation System The pump light and the erase light generated and adjusted coaxially as described above are reflected by the half mirror (8) and passed through the objective lens (9) of the sample observation system. (10) Light is collected on the upper surface. Sample (10) contains rhodamine 6 as described above.
Living cells stained with G. This sample (10)
Mounted on a three-dimensional movement stage (11), two-dimensional movement on the light-collecting surface and one-dimensional movement along the optical axis are possible. As the three-dimensional moving stage (11), for example, a stage driven by an inch worm type PZT (piezoelectric element) and capable of positioning with a high positional resolution of several nm can be used.

[C] Fluorescence Detection System The fluorescence detection system comprises a photomultiplier tube (14), an eyepiece (1)
3) An extremely simple structure of an infrared cut filter (12). By using the non-resonant two-photon excitation process, it is possible to extremely effectively prevent the pump light and the erase light from being mixed into the fluorescence signal as background light, but it is possible to further completely eliminate the stray light of the pump light and the erase light. In order to achieve the above, an infrared cut filter (12), which is one of the sharp cut filters, may be provided on the fluorescent light path so as to be freely inserted and removed.

As the infrared cut filter (12), for example, a filter capable of cutting light having a wavelength of 1 μm can be used to completely cut off stray light of pump light and erase light. The fluorescent band of rhodamine 6G is 570n.
m and a distance of 500 nm from the wavelength of the original light source, so that a stray light removing effect can be sufficiently obtained even with a rough infrared cut filter having not so high performance.

Further, since the wavelengths 2λ ₁ and 2λ ₂ of the pump light and the erase light do not overlap with the fluorescence band, it is not necessary to discard the fluorescence signals having the overlapping wavelengths as in the conventional double resonance absorption microscope, and to eliminate the fluorescence. The signals in all the fluorescent bands can be measured without waste without reducing the signal amount. Further, there is an advantage that the detector itself can be an inexpensive general commercial product such as a photomultiplier tube.

For example, in the double resonance absorption microscope disclosed in Japanese Patent Application No. 10-97924, expensive and precise optical elements such as a spectroscope and a notch filter are often used to separate excitation light and fluorescence signal. A complicated operation of measuring an entire fluorescence spectrum by combining an ICCD camera and a spectroscope and removing an erase light and a pump light on a computer was also required. Moreover, the system of the ICCD camera and the spectroscope is extremely expensive, and the burden on the user is too large.

On the other hand, if a time-resolved measurement method as illustrated in FIG. 15 is used as a measurement method, even if excitation light is mixed in, it can be completely cut off during fluorescence measurement. S / N can be remarkably improved.

As described above, according to the double resonance absorption microscope of the present invention, an extremely excellent S / N fluorescence image with a high signal intensity can be obtained with a simple configuration.

[Embodiment 2] Here, the depth resolution in the optical axis direction in the double resonance absorption microscope of the present invention illustrated in FIG. 14 will be described.

First, the non-resonant 2
We discuss quantitatively what kind of fluorescence intensity distribution is generated by the photon excitation process. First, it is assumed that the fluorescent labeler molecules are spatially and uniformly distributed in the sample. Table 1 shows specific calculation conditions.

[0100]

[Table 1]

FIGS. 17 and 18 show the horizontal direction (X-axis direction) in the focal plane from the sample by pump light excitation.
And the fluorescence intensity distribution in the optical axis direction (Z-axis direction) with the focal point as the origin.
The solid curve is the distribution using the non-resonant two-photon excitation process,
The dotted curve shows the distribution when the one-photon excitation process is used. Further, each fluorescence intensity distribution was calculated using Equations 5 and 6.

According to FIGS. 17 and 18, the fluorescence intensity distribution in the one-photon excitation process is shorter than that in the non-resonant two-photon excitation process because the excitation wavelength of the pump light is short. Focus on strength. In contrast, the fluorescence intensity distribution in the case of the non-resonant two-photon excitation process converges rapidly compared to that in the one-photon excitation process, and the intensity is zero at a distance from the focal point.
That is, in the non-resonant two-photon excitation process, the excited rhodamine 6G molecule is localized near the focal point, and emits fluorescence only from a specific place in the three-dimensional sample. This enables three-dimensional spatial resolution in a double resonance absorption microscope.

FIG. 19 illustrates the S ₁ → S ₂ excitation probability in the focal plane by the erase light in the case of the one-photon excitation process and the non-resonant two-photon excitation process. Assuming that the erase light is converted into a hollow beam to form a first-order Bessel beam, the excitation probability of S ₁ → S ₂ is given by the following equation (1).
7 can be calculated. At this time, the calculation is performed assuming that m = 1 in Expression 17. The S ₁ → S ₂ excitation probability is proportional to the first power of the laser light intensity in the case of the one-photon excitation process, and is proportional to the square of the laser light intensity in the case of the two-photon excitation process. Equation 6 is to be followed.

[0104]

[Equation 17]

According to the calculation results, it is understood that the fluorescence is suppressed according to the spatial distribution illustrated in FIG. Since it is assumed that the erase light is converted into a hollow beam, its shape affects the depth resolution in the optical axis direction of the microscope in both the one-photon excitation process and the non-resonant two-photon excitation process. Absent. However, in the profile of a plane perpendicular to the optical axis, the non-resonant two-photon excitation process is slightly broader and the hollow region is wider, so that it affects the planar resolution.

Next, the fluorescence intensity distribution when the fluorescence suppression effect is developed is obtained by using the fluorescence suppression characteristic curve represented by Expression 14 and the beam profiles exemplified in FIGS. FIGS. 20 and 21 exemplify the fluorescence intensity distributions in the horizontal direction (X-axis direction) and the optical axis direction (Z-axis direction) in the focal plane near the focal point when the non-resonant two-photon excitation process is used, respectively. It was done. Equation 17 was used for the calculation. Assuming that the sample was stained with rhodamine 6G, the calculation was performed assuming that the erase light had a wavelength of 1197 nm and a photon flux of 10 ³² photons / cm2 / sec.

For example, assuming a laser beam having a pulse width of 30 psec and condensing with an aberration-free lens having an NA of 0.75, the specification is approximately 10 μJ / pulse, the light intensity is extremely weak, and a commercially available laser is sufficient. It is a realistic value that can be covered. As is clear from FIG. 20, in the double resonance absorption microscope of the present invention using the non-resonant two-photon excitation process,
When the wavelength of the pump light is 1064 nm, about λ / 20
It can be seen that the fluorescent size of has been realized. In other words, it indicates that a spatial resolution of 50 nm can be obtained using infrared light. On the other hand, FIG. 21 shows the depth resolution in the optical axis direction, and a depth resolution of about 4λ is obtained.

FIG. 22 illustrates the spatial resolution when the fluorescence intensity distribution calculated in FIGS. 20 and 21 is Fourier-transformed into the spatial frequency domain. As is apparent from FIG. 22, according to the present invention, there is a depth resolution in the optical axis direction that cannot be obtained by the conventional double resonance absorption microscope. That is, this corresponds to having a frequency component in the k1z direction which is the optical axis.

As described above, the double resonance absorption microscope of the present invention illustrated in FIG. 14 has a lateral super-resolution by the spatial superposition of the pump light and the erase light, and a vertical direction by the non-resonant two-photon excitation process. It can be seen that spatial resolution in the (depth direction) can also be obtained. By setting the phase plate (6) to be removable from the optical path of the erase light,
The phase plate (6) is removed when only the depth resolution is required in the measurement of the fluorescence signal, and the phase plate (6) is inserted when the planar resolution is also required at the same time, so that either one can be selected. You can also.

Example 3 The double resonance absorption microscope of FIG. 14 described above shows that the S ₀ → S ₁ excitation (pump light) and the S
₁ → S ₂ excitation but those in the case of using the non-resonant two-photon excitation process to both (erase light), in the present invention, S ₀ →
It is also possible to use a non-resonant two-photon excitation process for either the S ₁ excitation or the S ₁ → S ₂ excitation.

FIG. 23 illustrates a double resonance absorption microscope of the present invention in which a one-photon excitation process is used for S ₀ → S ₁ excitation and a non-resonant two-photon excitation process is used for S ₁ → S ₂ excitation. It was done. The double resonance absorption microscope of FIG. 23 constitutes a laser scanning fluorescence microscope similarly to FIG. 14, and is an embodiment in which the irradiation area of the pump light and the erase light can be partially polymerized. It is assumed that the resonance multiphoton excitation process uses a non-resonant two-photon excitation process, and the sample (114) is a biological sample stained with rhodamine 6G as a fluorescent labeler molecule.

Therefore, the pump light has a photon energy E ₀₁ = E ₀₁ = necessary for S ₀ → S ₁ excitation in the ordinary one-photon excitation process.
About 2.33 eV · wavelength lambda ₁ = about 532nm, and the erase light-resonant two-photon excitation S in the process ₁ → S ₂ photon energy required for excitation E _12/2 = about 1.036EV · Wavelength 2
λ ₂ = about 1197 nm.

First, the pump light due to the one-photon excitation process is:
Nd: YAG separated by half mirror (102)
A fundamental wave having a wavelength of 1064 nm and a pulse width of 30 psec from the laser (101) passes through the KTP crystal (103) and is converted into a second harmonic (wavelength of 532 nm) to be generated. The pump light is enlarged by a telescope (104) and is incident on a dichroic mirror (111) as a beam combiner by a reflection mirror (105).

On the other hand, as for the erase light by the non-resonant two-photon excitation process, the primary Stokes light having a wavelength of 1197.4 nm is generated when the other fundamental wave separated by the half mirror (102) passes through the Raman shifter (106). do it,
Generated. As in the case of FIG. 14, the Raman shifter (106) is made of Ba (NO ₃ ) ₂ crystal and has an integral multiple of the resonance wave number of the crystal photon of 1047 cm ⁻¹ and has a longer wavelength side (lower energy side in terms of photon energy). The wavelength of the incident light is converted. The erase light is expanded by a telescope (107) and then converted into a hollow beam by a phase plate (108).

FIG. 24 (b) shows an example of a phase plate (108) for a wavelength of 1197 nm. FIG.
The phase plate (108) illustrated in (b) is obtained by dividing a region on a glass substrate into four parts and depositing magnesium fluoride in each region as illustrated in the cross section in FIG. 24 (a). In addition, the thickness of the magnesium fluoride is changed in each region so that the phases of the erase lights that have passed through the regions that are point-symmetrical to the optical axis with respect to the wavelength of 1197 nm are shifted from each other by π. As a result, erase light having an ideal hollow beam shape with zero intensity on the optical axis can be generated. It is preferable that the phase plate (108) can be inserted and removed, and the hollow beam of the erase light can be arbitrarily selected. Here, a description will be given assuming that the erase light is formed into a hollow beam so as to provide the horizontal super-resolution, and the hollow beam erase light is superimposed only on a part of the pump light.

The erase beam converted into a hollow beam is converted into a dichroic mirror (111) by a reflection mirror (109).
Incident on A notch filter (1) is provided in the optical path between the reflection mirror (109) and the dichroic mirror (111).
10) is inserted, and when the erase light passes through the notch filter (110), the light having a wavelength of 1064 nm, which is the fundamental wave of the Nd: YAG laser (101) which cannot be converted by the Raman shifter (106), is cut. Thus, only the erase light having a wavelength of 1197 nm is completely obtained.

Then, the photon energy E ₀₁ = about 2.33
eV · Wavelength lambda ₁ = erase light of the pump light and the photon energy E _12/2 = about 1.036EV · Wavelength 2 [lambda] ₂ = about 1197nm to about 532nm is a coaxial light by the dichroic mirror (111), a sample observation optical system It is led to.

In FIG. 23, the KTP crystal (103) is provided on the side of the pump light path of the half mirror (102), but the erase light can be generated even if it is provided on the incident side of the half mirror (102). . That is, since there is a fundamental wave that is not converted into a second harmonic even after passing through the KTP crystal (103), if the laser intensity is sufficient, the unconverted fundamental wave is incident on the Raman shifter (106), and S ₁ → S ₍₂ ) Erase light having the intensity required for excitation can be generated.

The pump light and the hollow beam erase light, which have been converted into coaxial light, ride on the optical axis of the sample observation optical system by the half mirror (112) and are focused on the sample surface by the objective lens (113).

In this embodiment, a microscope optical system for observation is also provided on the rear side of the sample (114).
The transmitted light of 4) is transmitted light lenses (116a) and (116).
Observed by a CCD camera (119) via b).

On the other hand, the fluorescent light emitted from the sample surface passes through the objective lens (113) and the half mirror (112), and passes through the notch filter (118). The notch filter (118) cuts the stray light of the pump light of wavelength lambda ₁ 532 nm. Erase light, because not mixed into the fluorescent signal since the wavelength 2 [lambda] _2, need not be removed from the fluorescent signal by using a notch filter.

The fluorescent light from which the stray light of the pump light has been removed is condensed on the diameter of the pinhole (120) placed at the confocal position by the imaging lens (119), and is then focused on the pinhole (1).
After passing through 23) and being spectrally separated by the spectroscope (121), the fluorescence spectrum is directly imaged by the ICCD camera (122).

The fluorescence signal has an excellent plane resolution due to the superposition of the pump light and the hollow beam erase light, and also has an excellent depth resolution due to the non-resonant two-photon excitation process for the erase light. . Of course, since neither the pump light nor the erase light is mixed into the fluorescent light signal as background light, the S / N becomes extremely high.

The three-dimensional imaging of such a fluorescence signal can be performed, for example, using a computer (not shown) as follows.

The three-dimensional moving stage (115) on which the sample is mounted is moved one-dimensionally in the optical axis direction while performing two-dimensional scanning on a plane orthogonal to the optical axes of the pump light and the erase. The fluorescent signal from each scanning point is ICCD
Input from the camera (122) to the computer. If each fluorescence signal data is visualized, the sample (114)
Is obtained. More specifically, the computer uses an Nd: YAG laser (10
By driving the controller (not shown) of 1),
At the same time as controlling the laser oscillation timing, the three-dimensional movement stage (115) is controlled, and the sample (114) is three-dimensionally scanned in synchronization with the laser oscillation timing. Similarly, ICCD is synchronized with laser oscillation.
The data of the fluorescence spectrum signal of the sample (114) is input to the computer from the camera (122).

Here, the pulse width of the pump light and the erase light of 30 psec corresponds to the fluorescence lifetime of the rhodamine 6G molecule of 3 nsec.
Therefore, the measurement time gate of the ICCD camera (122) and N
d: The optical path length of the YAG laser (101) can be adjusted. For example, say at a timing illustrated in FIG. 15 described above, t _p is the pulse width 30psec of the pump light, t _e is the pulse width 30psec of the erase light, t _g is the measurement gate of ICCD camera (122). As a result, even if a very small amount of pump or erase light is mixed in the fluorescence signal, it can be controlled so as not to be mixed in the ICCD camera (122) at the time of fluorescence measurement, and a further improvement in S / N is realized.

When this time-resolved measurement method is used, the structure may be simplified by removing the notch filter (118). Of course, if the S / N is more important than the optical system configuration, the time-resolved measurement method can be used together with the notch filter (118) remaining.

In the example of FIG. 23, the mechanical scan of the three-dimensional moving stage (115) is premised. However, it is needless to say that the laser beam itself is scanned by the galvanomirror that swings the half mirror (111). Is also possible.

Incidentally, the double resonance absorption microscope of this embodiment described above has a configuration in which a one-photon excitation process is used for S ₀ → S ₁ excitation and a non-resonant two-photon excitation process is used for S ₁ → S ₂ excitation. On the contrary, the same effect as above is realized by using a non-resonant two-photon excitation process for S ₀ → S ₁ excitation and a _one- photon excitation process for S ₁ → S ₂ excitation. It is of course possible to do so. In this case, with respect to the pump light, the KTP crystal (103) on the pump light path is removed, and the fundamental wave of 1064 nm (= 2λ ₁ ) of the Nd: YAG laser (101) is used as the pump light as it is. As for the erase light, the KTP crystal (103) is converted into a Raman shifter (106).
And the KTP crystal (103) is used to insert the fundamental wave 1064 nm into the second harmonic 532n.
m, and a secondary Stokes light having a second harmonic of 532 nm to 599 nm (= λ ₂ ) is generated by the Raman shifter (106), and is used as erase light. A phase plate (108) corresponding to a wavelength of 599 nm as shown in FIG.
The 9-nm erase light becomes an ideal hollow beam. As the notch filter (118), a filter capable of cutting off stray light of 599 nm erase light is used.

The present invention is not limited to the above-described example, and it goes without saying that various embodiments are possible in detail. For example, in the above examples, a sample is stained using rhodamine 6G as a typical example of a fluorescent labeler molecule. 3D spatial resolution can be realized. Wavelength-tunable OPA, OPO, OPG using optical nonlinear effect according to fluorescent labeler molecule
A laser system or the like may be used to generate pump light and erase light of appropriate photon energy and wavelength.

[0131]

As described above in detail, according to the invention of this application, the excitation light is prevented from being mixed into the fluorescent signal,
A new dual resonance absorption microscope capable of achieving extremely excellent S / N and also achieving excellent three-dimensional spatial resolution is provided.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the electron configuration of a sample molecule in a ground state.

FIG. 2 is a diagram exemplifying an electron configuration of a sample molecule excited to an S ₁ state.

FIG. 3 is a diagram exemplifying an electron arrangement of a sample molecule excited to an S ₂ state.

FIG. 4 is a diagram illustrating an electron configuration of a sample molecule that has been deexcited.

FIG. 5 is a conceptual diagram illustrating a double resonance absorption process.

FIG. 6 is a conceptual diagram spatially illustrating a double resonance absorption process.

FIG. 7 is a conceptual diagram exemplifying partial polymerization of pump light and erase light and suppression of fluorescence by the polymerization.

FIG. 8 is a conceptual diagram illustrating a phase plate for forming a hollow beam.

FIG. 9 is a diagram illustrating the absorption characteristics (solid line) and the fluorescence characteristics (dotted line) of Rhodamine 6G.

FIG. 10 is a conceptual diagram illustrating a non-resonant two-photon excitation process.

FIG. 11 is a conceptual diagram illustrating a double resonance absorption process using a non-resonant two-photon excitation process in the invention of this application.

FIG. 12 is a diagram illustrating a relationship between excitation light, an absorption band, and a fluorescence band.

FIGS. 13A and 13B are conceptual diagrams illustrating the spatial resolution in the case of a one-photon excitation process and a case of a non-resonant two-photon excitation process, respectively.

FIG. 14 is a schematic diagram showing an embodiment of the double resonance absorption microscope of the invention of this application.

FIG. 15 is a diagram showing an example of irradiation / measurement timing in the time-resolved measurement method.

FIG. 16 is a schematic view showing an example of a phase plate for forming a hollow beam.

FIG. 17 shows X in the focal plane when a one-photon excitation process (dotted line) and a non-resonant two-photon excitation process (solid line) are used.
It is the figure which illustrated the fluorescence intensity distribution of the axial direction.

FIG. 18 is a diagram illustrating a fluorescence intensity distribution in the Z-axis direction with the focal point as the origin when a one-photon excitation process (dotted line) and a non-resonant two-photon excitation process (solid line) are used.

FIG. 19 shows X in the focal plane when a one-photon excitation process (dotted line) and a non-resonant two-photon excitation process (solid line) are used.
It is the figure which illustrated S1 → S2 excitation probability of an axial direction.

FIG. 20 is a diagram illustrating a fluorescence intensity distribution in the X-axis direction in the focal plane when a non-resonant two-photon excitation process is used.

FIG. 21 is a diagram illustrating a fluorescence intensity distribution in the Z-axis direction when a non-resonant two-photon excitation process is used.

FIG. 22 is a conceptual diagram illustrating a spatial resolution when a non-resonant two-photon excitation process is used.

FIG. 23 is a schematic view showing another embodiment of the double resonance absorption microscope of the invention of this application.

24 (a), (b) and (c) are schematic cross sections of a phase plate, a schematic plane of a phase plate for light having a wavelength of 1197.4 nm, respectively.
It is the figure which illustrated the schematic plane of the phase plate for wavelength 599nm light.

[Explanation of symbols]

 Reference Signs List 1 Nd: YAG laser 2,8 half mirror 3 Raman shifter 4,5 reflection mirror 6 phase plate 7 dichroic mirror 9 objective lens 10 sample 11 three-dimensional moving stage 12 infrared cut filter 13 eyepiece 14 photomultiplier tube 101 Nd: YAG laser 102, 112 Half mirror 103 KTP crystal 104, 107 Telescope 105, 109 Reflecting mirror 106 Raman shifter 108 Phase plate 110, 118 Notch filter 111 Dichroic mirror 113 Objective lens 114 Sample 115 Three-dimensional moving stage 116a, 116b Lens for transmitted light 117 CCD camera 119 Imaging lens 120 Pinhole 121 Spectroscope 122 ICCD camera

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Masaaki Fujii 2-3-1 Tatsumi Minami, Okazaki City, Aichi Prefecture Building No. 6, No. 303 (72) Inventor Takashi Omatsu 5-10-9 Hirado, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture (72 ) Inventor Taku Sato 4-320-4-1209 Fuka term (reference) 4320-4-1 Tsukagoshi, Saiwai-ku, Kawasaki-shi, Kanagawa 2G043 AA03 DA02 DA05 EA01 EA02 EA13 FA01 FA02 FA03 GA07 GB19 HA02 HA09 HA15 JA01 JA02 KA01 KA02 KA09 LA02 LA03 MA01 NA01 2H052 AA07 AA09 AC04 AC15 AC27 AC29 AC34 AF07 AF14 AF21 AF25

Claims (6)

[Claims] A pump light source for exciting a sample molecule from a ground state to a first electronic excited state, and an erase light for exciting a sample molecule in the first electronic excited state to a second electronic excited state or a higher excited state. In a dual resonance absorption microscope that includes a light source and detects light emission when sample molecules in each excited state are de-excited to the ground state, pump light excites the ground-state sample molecules to the first electronic excited state. A double resonance absorption microscope characterized by having a photon energy that is not more than half of the energy. 2. A pump light source for exciting a sample molecule from a ground state to a first electronic excited state, and an erase light for exciting a sample molecule in the first electronic excited state to a second electronic excited state or a higher excited state. In a double resonance absorption microscope that includes a light source and detects light emission when sample molecules in each excited state are de-excited to the ground state, the erase light changes the sample molecules in the first electronic excited state to the second electronic excited state. A double resonance absorption microscope having a photon energy that is equal to or less than half the excitation energy to be excited. 3. A pump light source for exciting a sample molecule from a ground state to a first electronic excited state, and an erase light for exciting the sample molecule in the first electronic excited state to a second electronic excited state or a higher excited state. In a dual resonance absorption microscope that includes a light source and detects light emission when sample molecules in each excited state are de-excited to the ground state, pump light excites the ground-state sample molecules to the first electronic excited state. The erase light has a photon energy that is less than half the excitation energy that excites the sample molecule in the first electronic excited state to the second electronic excited state. A double resonance absorption microscope. 4. A photon energy band of light emitted from a sample molecule in a first electronically excited state overlaps with an excitation energy band for exciting a sample molecule in a first electronically excited state to a second electronically excited state or a higher excited state. The double resonance absorption microscope according to any one of claims 1 to 3, wherein: 5. An apparatus for irradiating a sample with the pump light and the erase light through the overlapping means, wherein the sample molecules in the first electronically excited state are provided. The double resonance absorption microscope according to any one of claims 1 to 4, wherein a light emission region upon deexcitation to a ground state is partially suppressed. 6. The double resonance according to claim 1, wherein the sample is stained with a fluorescent labeler molecule having at least three electronic states including a ground state, and the sample molecule is the fluorescent labeler molecule. Absorption microscope.