Electron beam application device and electron beam generation method

Abstract

In order to increase the luminance of an electron beam emitted from a photocathode, the present invention irradiates a photocathode comprising a photoelectric film with a first pulsed light and additionally with a second pulsed light having a longer wavelength than the first pulsed light. In one aspect, an applied electron beam device is provided with: a photocathode (1); a first pulsed light source (7) which emits a first pulsed light; a second pulsed light source (8) which emits a second pulsed light having a ring-band shaped optical intensity distribution; an excitation light optical system (21) which outputs the first pulsed light and the second pulsed light with the optical paths thereof combined; a condensing lens (2) that condenses the first pulsed light and the second pulsed light output from the excitation light optical system at the photocathode; and an electronic optical system which irradiates a sample (20) with an electronic beam emitted from the photocathode.

Description

Technical Field

[0001] The present invention relates to an electron beam application device and an electron beam generation method.

Background Art

[0002] In a high-resolution electron microscope, an electron source with high brightness and a narrow energy width of the emitted electron beam is essential. Furthermore, in recent years, in addition to high spatial resolution, for the purpose of capturing fast temporal changes, technology for pulsing the electron beam serving as a probe has been progressing. A photocathode that generates pulsed electrons by photo pulse incidence is extremely useful for capturing rapidly changing phenomena. Among them, a semiconductor surface containing GaAs or the like is made to have a low work function with Cs, O, etc. and a photoexcitation electron source utilizing negative electron affinity (NEA) that emits electrons excited into the conduction band by interband transition into a vacuum by light incidence has the characteristic of high brightness. It is a planar electron source, and although the focal size of the excitation light that becomes the light source size is about 1 μm which is relatively large, the emitted electrons have good straightness.

[0003] Patent Document 1 discloses an electron gun using this photocathode. As the photocathode, a transparent substrate, specifically, one with a photoelectric film attached on a transparent substrate is used, and excitation is performed with a condenser lens having a large numerical aperture (NA) of about 0.5 placed close to the transparent substrate. By focusing the excitation light on the photoelectric film up to the diffraction limit, a small electron light source is formed, and an electron gun structure that utilizes the electron beam emitted into a vacuum from here is shown.

[0004] Patent Document 2 discloses that the excitation light incident on the photocathode is two laser lights, one of which is radially polarized, and the effective work function of the photocathode is lowered by making the electric field vector formed at the focusing position perpendicular to the photocathode surface.

Prior Art Documents

Patent Document

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] For further high-resolution improvement of an electron microscope using a photoexcited electron source as an electron source, it is desirable to increase the brightness of the photoexcited electron source. However, even if the intensity of the excitation light is increased, the probe current obtained with continuous light does not increase from a certain point. This is because the amount of heat generated and the amount of charge accumulated on the surface increase with the increase in the emitted current, and these limit the emitted current. One solution is to pulse the excitation light. This can alleviate the accumulation of heat and surface charge. In the case of a photocathode mainly composed of GaAs, laser light with a wavelength of 760 to 800 nm is suitable as the excitation light source, and examples of this light source include semiconductor diodes and solid-state lasers using light-emitting crystals. These can extract pulsed light and are useful for various measurements because the pulse width and interval can be controlled by electrical signals or optical signals.

[0007] However, there is still a limit to the brightness even when the excitation light is pulsed. The brightness of the electron source increases as the area of the emission source becomes smaller. In the case of a photoexcited electron source, the size of the cathode region where electrons are emitted is the size of the spot where the excitation light is focused on the photocathode, and this size is determined by the wavelength of the excitation light and it is difficult to make it smaller.

[0008] Also, when using wide-gap semiconductors such as GaN and InGaN as the photoelectric film of the photocathode to enable higher current density electrical conduction, the acceptors doped inside Since it had a high ionization energy and a low hole density at room temperature, it had a high electrical resistance, which limited the emission current density.

[0009] Also, as shown in Patent Document 2, even when a radially polarized beam is irradiated to generate an electric field vector perpendicular to the photocathode surface, since the electric field by light vibrates temporally, the electron emission increases and decreases, which limits the improvement of luminance.

Means for Solving the Problems

[0010] An electron beam application device according to an embodiment of the present invention includes a photocathode including a photoelectric film, a first pulse light source that emits first pulse light, a second pulse light source that emits second pulse light having a wavelength longer than that of the first pulse light and having an annular light intensity distribution, an excitation light optical system that combines and emits the optical path of the first pulse light and the optical path of the second pulse light, a condenser lens that condenses the first pulse light and the second pulse light emitted from the excitation light optical system onto the photocathode, and an electron optical system that irradiates a sample with an electron beam emitted from the photocathode.

Advantages of the Invention

[0011] It becomes possible to increase the luminance of the electron beam emitted from the photocathode. Other problems and novel features will become apparent from the description of this specification and the accompanying drawings.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2A

Figure 2B

Figure 3

Figure 4A

Figure 4B

Figure 4C

Figure 5A

Figure 5B

Figure 6

Figure 7A

Figure 7B

Figure 8A

Figure 8B

Figure 8C

Figure 9

Figure 10

Figure 11A

Figure 11B

Figure 11C

Figure 12A

Figure 12B

Figure 13

Figure 14

Figure 15A

Figure 15B

Embodiments for Carrying Out the Invention

[0013] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

Example

[0014] FIG. 1 is a schematic diagram of an electron beam application apparatus. The photocathode 1 has pulsed excitation light 12 incident from above, and generates a pulsed electron beam from the lower surface of the portion where the pulsed excitation light 12 is incident. The electrons emitted from the photocathode 1 are accelerated by the electric field between the opposing extraction electrodes 3 and are guided as a pulsed electron beam 13 from the opening 14 of the vacuum chamber 9 into the electron optical system. As an example of the electron optical system, here, an example of a scanning electron microscope (SEM) is shown. The pulsed electron beam 13 extracted from the opening 14 is reduced by two electron lenses 17 and the objective lens 29 and focused on the sample 20. An SEM image of the sample surface is obtained by the deflector 18 for scanning the focal position and the electron detector 19 for measuring the signal electrons generated from the sample by the irradiation of the pulsed electron beam 13, and it becomes possible to observe the fine structure. Note that an aperture 38 for restricting the opening angle of the electron beam is provided in the orbit of the electron beam and is adjusted to have an optimal orbit shape in conjunction with the electron lens 17. system, here, an example of a scanning electron microscope (SEM) is shown. The pulsed electron beam 13 extracted from the opening 14 is reduced by two electron lenses 17 and the objective lens 29 and focused on the sample 20. An SEM image of the sample surface is obtained by the deflector 18 for scanning the focal position and the electron detector 19 for measuring the signal electrons generated from the sample by the irradiation of the pulsed electron beam 13, and it becomes possible to observe the fine structure. Note that an aperture 38 for restricting the opening angle of the electron beam is provided in the orbit of the electron beam and is adjusted to have an optimal orbit shape in conjunction with the electron lens 17. In the orbit of the electron beam, an aperture 38 for restricting the opening angle of the electron beam is provided and is adjusted to have an optimal orbit shape in conjunction with the electron lens 17.

[0015] The photocathode 1 is an electron source that utilizes a phenomenon known as negative electron affinity. The photoelectric film 10 of the photocathode 1 is a p-type semiconductor, and typically GaAs is used. In order to reduce the work function, for example, Cs and oxygen are adsorbed on its surface.

[0016] Although not shown, a preparation chamber for this purpose is provided adjacent to the electron gun 15, and the surface of the photocathode 1 moved to the preparation chamber is adjusted. The adjustment of the cathode surface is carried out as follows. Perform treatment. Remove oxides and carbides on the semiconductor surface using a heating heater. Generating atomic hydrogen for cleaning treatment is even more effective. Then, by adsorbing Cs vaporized in an oxygen atmosphere onto the semiconductor surface, the work function of the semiconductor surface decreases, and a state is obtained where the vacuum level is lower than the conductor inside the cathode film. As a result, for the inter-band transition of the p-type semiconductor, When electrons excited from the valence band to the conduction band by the incidence of excitation light 12 with sufficient energy for the inter-band transition of the p-type semiconductor are in a state where their potential energy is already higher than the vacuum level, they are emitted into the vacuum by the electric field on the film surface.

[0017] In the case of GaAs, the wavelength of the excitation light needs to be 850 nm or less, and more preferably 780 nm to 660 nm. If the conduction band electrons spread in the photoelectric film 10, the size of the light source increases. As a result, the luminance decreases, which is not preferable for an electron gun for a microscope. Therefore, the thickness of the photoelectric film 10 needs to be about 1 μm or less, and a sufficient concentration of p-type impurities is also required. The photoelectric film 10 is attached to a transparent substrate 11 that does not absorb the excitation light to form a photocathode 1. As the transparent substrate 11 of the photoelectric film 10 made of GaAs, a semiconductor with a wider bandgap such as glass or GaP can be used. The photoelectric film 10 may be directly attached to the transparent substrate 11 or attached via an intermediate layer. Also, when a semiconductor is used as the transparent substrate 11, the photoelectric film 10 may be formed by epitaxial growth. In that case, the photoelectric film 10 may be formed on top of a buffer layer that serves as an intermediate layer.

[0018] The acceleration voltage to the electron source is applied by an acceleration power supply 5 through a cathode holder 4 that is in electrical contact with the photoelectric film 10, and the pulsed electron beam 13 generated from the photocathode 1 is accelerated between the extraction electrode 3. Note that the vacuum chamber 9 of the electron gun 15 is evacuated by a vacuum evacuation device (not shown), preferably an ion pump and a non-evaporable getter pump. The pulsed electron beam 13 generated from the photocathode 1 is accelerated between the extraction electrode 3. The vacuum chamber 9 of the electron gun 15 is evacuated by a vacuum evacuation device (not shown), preferably an ion pump and a non-evaporable getter pump.

[0019] The condenser lens 2 that condenses the pulsed excitation light 12 onto the photocathode 1 is a lens corrected for spherical aberration so that the focus formed on the photoelectric film 10 through the transparent substrate 11 is minimized. In this example, the diameter of the parallel light is 4 to 4.2 mm, and the NA (Numerical Aperture) is 0.5. As a result, in the case of excitation light (laser light) with a wavelength of 660 nm, as shown by the solid line A in Fig. 4A, on the photoelectric film 10 a diameter of about 1 μm (full width at half maximum) can be obtained. The parallel pulsed excitation light 12 is emitted from the first pulse light source 7 outside the vacuum vessel 9 and introduced into the vacuum vessel 9 through the excitation light optical system 21 and the window 6 . On the other hand, a second pulse light source 8 with a longer wavelength (for example, 850 nm) is provided in the excitation light optical system 21.

[0020] The configuration of the excitation light optical system 21 is shown in Fig. 2A. The first pulse light 27 emitted from the first pulse light source 7 passes through the optical coupler 24 and travels toward the photocathode 1 as the pulsed excitation light 12. The second pulse light 28 emitted from the second pulse light source 8 also passes through the orbit of the pulsed excitation light 12 and is irradiated onto the photoelectric film 10 by the condenser lens 2. The second pulse light source 8 is connected to the excitation light optical system 21 via the fine adjustment mount 51. By the fine adjustment mount 51, the second pulse light 28 is reflected by the optical coupler 24 toward the photocathode 1, and thus is adjusted to travel toward the photocathode in coincidence with the optical path of the pulsed excitation light 12. The second pulse light 28 is laser light obtained by modulating the long-wavelength pulsed light 52 emitted from the long-wavelength pulse light source 22 with a wavelength of 850 nm by the optical modulator 23.

[0021] ​​​​As an example of modulation by the optical modulator 23, an example of using a phase mask that creates a phase difference between the center and the outer periphery of the laser light is shown in FIG. 3. FIG. 3 shows a plan view of the phase mask used as the optical modulator 23 and a cross-sectional structure of the central portion. The optical modulator 23 is a plate with a refractive index n that is transparent to the wavelength of the second pulsed light 28, and a circular phase shift region 40 is provided at the center. There is a step between the phase shift region 40 and the peripheral region. As a result, the obtained second pulsed light 28 has a phase shift of π between the central portion and the outer peripheral portion, and the light intensity distribution brought to the photoelectric film 10 through the condenser lens 2 becomes an annular shape with a dark center as shown in FIG. 4B.

[0022] In order to shift the phase of the light by π by the phase shift region 40, the height h of the step of the phase shift region 40 (see FIG. 3) needs to satisfy (Equation 1).

[0023]

Equation

[0024] When the optical modulator 23 is made of a quartz plate (n = 1.46) and the wavelength of the second pulsed light 28 is 850 nm , h = 924 nm. Also, to minimize the light intensity at the center portion, that is, to make the light intensity distribution annular , the diameter d of the phase shift region 40 (see FIG. 3) should be set to the size determined by (Equation 2).

[0025]

Equation

[0026] Here, D (see FIG. 3) is the diameter of the second pulsed light 28. Note that (Equation 2) is the condition when the light intensity of the second pulsed light 28 is uniform. For example, when the second pulsed light 28 has an intensity distribution such as a Gaussian beam, the diameter d needs to be set accordingly.

[0027] Note that 850 nm for the wavelength of the second pulsed light 28 is an example, and the wavelength of the second pulsed light 28 is, for the light It is determined to be substantially consistent with the bandgap energy of the semiconductor GaAs that constitutes the photoelectric film 10. Any wavelength is acceptable as long as it generates stimulated emission and causes the electrons excited in the photoelectric film 10 to recombine. Therefore, if the photoelectric film 10 is made of a crystal other than GaAs, the wavelength is determined by its physical properties. When a superlattice structure is provided, the wavelength corresponds to the bandgap energy of the formed miniband. In this case as well, the condition is that it is the wavelength for generating stimulated emission.

[0028] Also, as the optical coupler 24 that combines the optical path of the first pulsed light 27 and the optical path of the second pulsed light 28, a half mirror, a polarization beam splitter, a dichroic mirror, etc. can be used. By using a dichroic mirror for the optical coupler 24 such that the wavelength of the first pulsed light 27 is transmitted and the second pulsed light 28 is reflected and enters the orbit of the pulsed excitation light 12 respectively, a system with less optical loss can be achieved.

[0029] The effect of irradiating the photoelectric film 10 with the first pulsed light 27 and the spatially modulated second pulsed light 28 will be described. In Fig. 5A, due to the incidence of the first pulsed light 27, electrons in the valence band in the photoelectric film 10 are excited across the energy gap Eg into the conduction band and are shown in a state of temporarily accumulating at the bottom of the conduction band after losing energy due to phonon scattering or the like. The spatial distribution of the electrons in the conduction band at this time is the broken line B in Fig. 4A. Due to the scattering and diffusion of the electrons from the intensity distribution of the first pulsed light 27 shown by the solid line A in the figure, it spreads by about 1.5 μm in the full width at half maximum. Since these electrons are emitted into the vacuum, the source size of the electron beam is wider than the intensity distribution of the excitation light and is about 1.5 μm in the full width at half maximum.

[0030] When a second pulsed light 28 is irradiated in the state shown in FIG. 5A with its pulse intensity (pulse height), pulse width and timing optimized, stimulated emission occurs in which electrons in the conduction band and holes in the valence band immediately recombine and emit light at the location where the second pulsed light 28 hits, as shown in FIG. 5B. Since the light intensity distribution of the second pulsed light 28 is annular and weak in the central portion as shown in FIG. 4B, the distribution of electrons in the conduction band changes from dashed line B to solid line C as shown in FIG. 4C, and the half-width can be narrowed to 1 μm or less. Since electrons are emitted into the vacuum according to the spread of solid line C, it becomes possible to narrow the source size of the electron beam to about 0.8 μm at half-width. do.

[0031] In electron sources that utilize negative electron affinity, making the electron-emitting area smaller is extremely effective in increasing the brightness of the electron beam. Even if the area of ​​the emission area is large, it is in principle possible to increase the brightness if the current density of the emitted electrons is the same. However, under high-brightness conditions, the current density is only several A / mm. 2 That's a big A current density is required. For example, the size of the emission area is set to 1 mm 2 Then, several A It is necessary to pass a large current through it. In this case, side effects such as heat generation causing the GaAs photoconductive film to melt, Cs to evaporate, or surface charges to accumulate on the photoconductive film surface, making it difficult to sustain electron emission. To avoid such side effects, it is necessary to reduce the area of ​​the electron emission region as much as possible and reduce the total amount of current.

[0032] Conventionally, an aspheric lens is used as the condenser lens 2, and the excitation light is focused to the diffraction limit. It was possible to obtain an intensity distribution with a diameter of about 1.2 to 1.5 μm. In contrast, by using the second pulsed light 28, it is possible to reduce the electron emission region by 67 to 53%, and the area by about 44 to 28%, so that for the same total current, the brightness can be increased by 2 to 3 times.

[0033] Thus, in this embodiment, by eliminating the outer portion of the electrons in the conduction band generated in the optoelectronic film 10 with the second pulsed light 28 from the first pulsed light 27, a high-intensity electron beam is generated. The second pulsed light 28 has the effect of causing the excited electrons to recombine with holes by stimulated emission, while also having the effect of exciting the electrons in the valence band to the conduction band. Therefore, it is necessary to optimize the intensity and timing of both pulsed lights so that the second pulsed light 28 is used almost for stimulated emission.

[0034] The adjustment means for both pulsed lights will be described. FIG. 2B shows the time variation of the intensities of both pulsed lights, where the vertical axis is the intensity I and the horizontal axis is the time. The pulse of the first pulsed light 27 has an intensity Ip 1 and a pulse width Tp 1 . The pulse width Tp 1 is a time that can be selected according to the purpose of the experiment, and as a characteristic value, it can be 1 picosecond or more and on the order of nanoseconds or microseconds. The intensity Ip 1 is set according to the required brightness as an electron gun . In contrast, the pulse of the second pulsed light 28 has an intensity Ip 2 and a pulse width Tp 2 . The start delay time Tr of the second pulsed light 28 from the start of the first pulsed light 27 and the end delay time Td of the second pulsed light 28 from the end of the first pulsed light 27, which determine the pulse width Tp 2 , are optimized by preliminary experiments.

[0035] To achieve this, as shown in FIG. 1, a control signal 33 indicating the height and timing of each pulse is sent from the main control device 32 to the excitation light control device 31, and the excitation light control device 31 issues laser control signals 37 to the first pulsed light source 7 and the second pulsed light source 8 respectively to adjust the timing and intensity.

[0036] Regarding the magnitudes of the start delay time Tr and the end delay time Td, in principle, the electrons generated in the conduction band by the first pulsed light 27 lose energy due to phonon scattering on the order of femtoseconds and accumulate at the bottom of the conduction band. However, they then recombine with holes and disappear in about 1 picosecond. Therefore, the start delay time Tr is selected from 0 to 0.1 picosecond, and the end delay time Td is selected from about 0 to 1 picosecond. To optimize the start delay time Tr and the end delay time Td, the brightness of the pulsed electron beam is monitored and adjusted to be the highest. The brightness of the pulsed electron beam 13 can be grasped from the resolution and brightness of the SEM image obtained by the electron optical system. Also, by inserting the Faraday cup 35 into the central axis 34 of the electron optical system and measuring the probe current with the current detector 36, the brightness of the pulsed electron beam 13 can be grasped more simply. However, as will be described later, since the electron beam reduced by the plurality of electron lenses 17 passes through the aperture 38 with a reduced current amount due to the opening angle limitation, the insertion position of the Faraday cup 35 needs to be closer to the sample 20 than the aperture 38. The brightness information obtained in this way is fed back to the main control device 32 for optimizing the conditions of the electron gun 15.

[0037] By using the pulsed electron beam 13 with increased brightness as described above in a scanning electron microscope, higher-resolution observation becomes possible. Alternatively, even with the same resolution as the conventional device, the probe current can be increased, enabling faster observation. In particular, the energy distribution of the electron source in this embodiment is well aligned compared to 0.3 eV of the cold field emission (CFE) electron source using a tungsten filament and 0.8 eV of the Schottky emission (SE) electron source, and in particular, at low acceleration, the resolution can be improved by reducing chromatic aberration. For example, when the acceleration power supply 5 is set to -2 kV, the electron beam 13 emitted from the electron gun can obtain 39 mA / sr or more in terms of the angular current density of radiation.

[0038] With the configuration of Example 1, a small electron light source diameter of 800 nm is obtained, enabling observation with a resolution of 1 nm or less. Therefore, when the size of the electron beam probe imaged on the sample is 0.8 nm or less, the electron The total reduction ratio by the electron lens 17 and the objective lens 29 of the optical system housing 16 may be set to about 1 / 1000. For example, when adjusting the aperture 38 and the optical conditions such that the half opening angle αo of the objective lens 29 onto the sample 20 is 10 mrad, the electron beam that can pass through the aperture 38 has a very narrow range with the half beam opening angle αs on the light source side of the electron gun 15 being 5 μrad or less. Even in this case, a probe current of about 10 pA can be obtained for the electrons irradiated onto the sample 20, and observation can be performed with high sensitivity in a short time. The depth of focus is about 100 nm as spot diameter / αo. Moreover, since the electron beam 13 is a pulsed electron beam, it is also useful in time-resolved measurement.

[0039]

[0040] FIG. 6 shows a modified example of the excitation light optical system 21. The excitation light optical system 21 in FIG. 6 is provided with a monitor optical system 46 for measuring the reflected light 47 from the photocathode 1 in order to facilitate the optimization of the excitation light optical system 21. A reflection light separator 25 for separating the reflected light 47 in the lateral direction (x direction) is provided in the optical path. Here, as an example, a polarized beam splitter (PBS) is used as the reflection light separator 25, and a quarter-wave plate 42 is provided thereunder. The pulsed excitation light 12 transmitted through the reflection light separator 25 is linearly polarized light with an electric field component parallel to the plane of the paper, and becomes circularly polarized light by passing through the quarter-wave plate 42. The circularly polarized light reflected from the photocathode 1 after passing through the condenser lens 2 is converted into linearly polarized light having an electric field component perpendicular to the plane of the paper when passing through the quarter-wave plate from bottom to top. Therefore, it is reflected by the reflection light separator (PBS) 25, and only the reflected light 47 from the photocathode 1 is introduced into the monitor optical system 46. By imaging the reflected light 47 onto the imaging element 41 by an imaging optical system including the imaging lens 44, the condensing position and the focal shape of the excitation light on the photocathode 1 are enlarged. ​​​​​​It can be observed. From the condensing position and focal shape of the observed excitation light, the optical paths of the first pulsed light 27 and the second pulsed light 28 are adjusted so that their focal positions coincide. In this configuration example, by adjusting the position and angle of the second pulsed light source 8 with the fine adjustment mount 51, the optical path of the second pulsed light 28 is aligned with the optical path of the first pulsed light 27. Also, in the monitor optical system 46, light attenuation means such as an ND filter 45 is provided to optimize the observed light amount for the imaging device 41.

[0041] Furthermore, a detachable polarizer 43 may be provided in the monitor optical system 46. By inserting the polarizer 43, the reflected light 47, which is linearly polarized, cannot pass through. On the other hand, since the induced emission light obtained by the incidence of the first pulsed light 27 and the second pulsed light 28 is circularly polarized, it can be observed on the imaging device 41. The incident positions of the first pulsed light 27 and the second pulsed light 28 can be adjusted according to the intensity and shape of the induced emission light. Note that it is not limited to being detachable, and for example, a polarizer capable of switching the polarization direction of the transmissible light can also be used.

[0042] Hereinafter, detailed modification examples will be described.

[0043] In the laser control signal 37 from the excitation light control device 31 to the first pulsed light source 7 and the second pulsed light source 8 shown in FIG. 1, an optical signal may be used for timing adjustment. In the case of an optical signal, timing adjustment can be performed using a device called an optical delay that sets the delay time of light propagation by adjusting the optical path length, which is advantageous for adjusting a very short time.

[0044] Also, the optical modulator 23 is not limited to that shown in FIG. 3, and a spatial phase modulator such as a liquid crystal element may be used. Furthermore, a donut-shaped focal shape with a dark central portion is formed on the photocathode 1 As the phase modulator, a spiral phase plate shown in FIG. 7A may be used. The spiral phase plate 70 is circular. In this example, it is divided into eight parts in the circumferential direction, and the height changes counterclockwise from the divided region 71-0 to the divided region 71-7, and the phase changes by π / 4 step by step. The height h of the step between adjacent divided regions in the counterclockwise direction is set to be (Equation 3).

[0045]

Equation

[0046] The polarization of the second pulsed light 28 that has passed through the spiral phase plate becomes helical, and the intensity pattern of the light focused and formed on the photocathode 1 is as shown in FIG. 7B, with a dark central part whose width can be made narrower than that in the case of the phase plate having the structure of FIG. 3. When using a spiral phase plate, by adjusting it with the first pulsed light 27, an electron emission source with a narrower diameter of about 0.5 μm can be obtained, so that the brightness of the electron beam can be increased by four times or more. The spiral phase plate may have a structure with finer divisions of eight or more or a continuously changing height, and the same effect can be obtained if the phase difference of 2π is achieved after one revolution.

[0047] Also, the photoelectric film 10 is not limited to GaAs, and materials that emit electrons due to other negative electron affinities can be used. For example, GaP, mixed crystals such as GaAs-GaP and GaAs-AlAs, and superlattices can be used. As described above, the wavelength of the second pulsed light 28 needs to be selected according to the material and structure of the photoelectric film 10.

[0048] Also, the pulse widths of the first pulsed light 27 and the second pulsed light 28 in this embodiment are typically selected between 1 ps and 1 μs. However, by adjusting the light intensity, it is possible to widen the usable pulse width. For example, it may be continuous light with a time width from milliseconds to several tens of seconds or more according to the observation time of the electron microscope.

[0049] In addition, there may be a case where the light source size of the electron beam is reduced only during the time of use by an electron beam application apparatus to increase the brightness. In this case, either the first pulsed light 27 or the second pulsed light 28 may be used as pulsed light, and the other may be used as continuous light.

Embodiment

[0050] In this embodiment, in the configuration of the electron beam application apparatus of FIG. 1, infrared light is used for the second pulsed light source 8. The wavelength thereof does not cause an interband transition in the photoelectric film 10, but has an energy that excites electrons from the valence band to the acceptor level and generates holes (positive holes).

[0051] Wide-gap semiconductors such as GaN and diamond have a much larger breakdown electric field than semiconductors such as Si and GaAs of the semiconductor. As a result, since a large amount of electrons can be transported in the bulk, it is extremely promising as an electron source that requires a large current density. However, when trying to emit electrons from a superlattice or mixed crystal of GaN or GaInN-based by negative electron affinity, the current density cannot be increased. The main cause of this is that p-type doping does not work well. For example, when an impurity that becomes an acceptor is doped into a semiconductor such as Si, the energy Ea from the valence band of the acceptor level generated in the forbidden band is as small as 0.05 eV. Even at room temperature, electrons enter the acceptor due to thermal energy and become negatively charged, and holes from which electrons have escaped are generated in the valence band. Since this hole can move around as a charge carrier, it can increase the electrical conductivity or redistribute itself to relax the external electric field and minimize the depletion layer width formed near the surface. On the other hand, in the case of a wide-gap material, the energy Ea of the acceptor level is about 0.1 eV for GaN and about 0.5 eV for diamond. At room temperature, heat

[0052] energy is not sufficient to cause electrons to enter the acceptor, so that the conductivity cannot be increased, and the depletion layer width cannot be minimized. It is not possible to increase the hole density by excitation. Although there is no clear definition for the bandgap energy Eg that classifies a semiconductor as a wide-gap semiconductor, here, those with a bandgap energy Eg of 3 eV or more are considered wide-gap semiconductors.

[0053] Models of the energy structure of a p-type impurity-doped GaN-based wide-gap semiconductor used as the photoelectric film 10 are shown in FIGS. 8A to 8C. In a vacuum, with Cs or oxygen adsorbed on the surface of the photoelectric film 10 to lower the work function, FIG. 8A shows the model when no light is irradiated. Since the acceptors are hardly charged and the hole density is low, there are no charge carriers, and due to the difference between the internal potential and the surface potential, the vicinity of the surface is depleted. Even if excitation light is irradiated here to generate electrons and holes by interband transition, since there are hardly any charge carriers, the emission current density does not become sufficiently large due to the attractive force between electrons and holes or the potential rise due to the charges accumulated on the surface, and the electron source has an extremely low luminance.

[0054] In this state, infrared second pulsed light is irradiated. The energy hν 2 of the second pulsed light is equal to or greater than the energy Ea of the acceptor level, and as shown in FIG. 8B, holes are generated when electrons in the valence band charge the acceptors. Here, h is Planck's constant, and ν 2 is the frequency of the second pulsed light. As a result, due to the high-density holes, the internal electric field becomes small, and a very narrow depletion layer is formed on the surface. In this state, when the first pulsed light 27 with energy hν 1 (ν 1 is the frequency of the first pulsed light 27) is incident, as shown in FIG. 8C, the electrons excited to the conduction band by interband transition are emitted into the vacuum at high density.

[0055] The wavelength of the second pulsed light should be sufficiently lower than the bandgap energy and equal to or greater than the energy Ea of the acceptor level, with a wavelength of 1 μm or more. The long-wavelength limit is 12 μm for GaN and 2.45 μm for diamond, which can be determined for each bandgap energy. That's fine.

[0056] If the infrared light that generates holes is continuously irradiated, it will cause the temperature of the photoelectric film 10 to rise, leading to Cs evaporation, crystal deterioration, etc., and stable electron emission as an electron source will not be possible. Therefore, it is effective to irradiate in synchronization with the first pulsed light 27, and as for the timing, it is simultaneous with or slightly prior to the irradiation of the first pulsed light 27. This relationship is shown in FIG. 9. The vertical axis represents the intensity I, and the horizontal axis represents time, showing the time change of the intensities of both pulsed lights. The pulse of the first pulsed light 27 has an intensity Ip , a pulse width Tp 1 , the pulse of the second pulsed light has an intensity Ip 1 , and a pulse width Tp 2 . The leading time -Tr of the rise of the second pulsed light with respect to the rise of the first pulsed light 27, and the leading time -Td of the fall of the second pulsed light with respect to the fall of the first pulsed light 27 are optimized within the range of about 0 to several picoseconds, 2 and together with the intensities Ip , Ip 1 , Ip 2 , similar to Example 1, information such as the probe current amount and resolution in the electron optical system is fed back and determined so as to obtain the maximum luminance.

[0057] The wide-gap material of the photoelectric film 10 is not limited to GaN and diamond, and mixed crystals such as AlGaN and GaInN or superlattices of these and GaN etc. may be used.

[0058] Here, instead of directly irradiating the light that generates holes with infrared light of a long wavelength to the condensing position of the first pulsed light 27 on the photocathode (in this case, since it is not necessary to combine the optical paths of the first pulsed light 27 and the second pulsed light, the excitation light optical system 21 can be made unnecessary), two wavelengths of light may be combined and used. FIG. 10 shows a configuration example of the excitation light optical system 21 in this case. By the optical coupler 24b, the optical path of the third pulsed light 54 from the third pulsed light source 53 is combined with the second pulsed The second pulse light 52 and the third pulse light 54 are irradiated onto the photocathode 1 so as to coincide with the optical path of the second pulse light 52 emitted from the light source 22. The respective wavelengths of the second pulse light 52 and the third pulse light 54 are long wavelengths that do not cause interband transitions in the photoelectric film 10, and the energy difference between the two is set to be equal to or greater than the energy Ea of the acceptor level. In this configuration, since visible light or near-infrared light can be used as the second pulse light 52 and the third pulse light 54, the condenser lens 2 can be used under conditions where the refractive index is ideally close. Therefore, the temperature rise range caused by light irradiation is minimized only in the vicinity of the electron emission source rather than directly irradiating infrared light, and the hole density can be increased. In addition, since resonant charging of the acceptor and generation of holes are possible, there is an advantage that the reaction cross section (scattering cross section) becomes large and the efficiency becomes high.

[0059] Note that if the second pulse light 52 and the third pulse light 54 are simultaneously irradiated, there is a possibility that interband transitions may occur due to two-photon absorption. To prevent this interband transition, the respective energies may be set to 1 / 2 or less of the bandgap energy Eg. Also, in order to obtain the maximum luminance the intensities of the second pulse light 52 and the third pulse light 54 and the irradiation timing of the second pulse light 52 and the third pulse light 54 with respect to the irradiation timing of the first pulse light 27 are determined by feedback of information such as the probe current amount and resolution in the electron optical system, as in Example 1.

[0060] Note that infrared light irradiation has the effect of increasing the luminance not only on the photoelectric film including the wide-gap semiconductor but also on the photoelectric film 10 such as GaAs having a bandgap energy Eg of around 1 eV .

[0061] When the GaAs or the photoelectric film 10 mainly composed of GaAs is used as the electron source with negative electron affinity, at the initial stage of electron emission, as shown in Fig. 11A, the energy hν by the first pulse light 27 1The electrons excited to the conduction band by photoexcitation are emitted into the vacuum beyond the lowered vacuum level. However, as time passes, the electrons scattered inside the surface are trapped in a level such as the surface state that is difficult to escape. FIG. 11B shows the state in which electrons are trapped in the surface level, which increases the electron potential energy on the surface of the photoelectric film 10, thereby reducing the electron emission amount. This is one of the causes of the luminance limitation.

[0062] Therefore, by irradiating with a second pulsed light that is infrared light and exciting and removing the electrons trapped in the surface level to the conduction band or the vacuum level as shown in FIG. 11C, it is possible to prevent the luminance from decreasing. The potential energy of the surface level is near the center of the forbidden band, and in the case of GaAs, it is about 0.7 eV, which is half of the bandgap energy Eg. Therefore when the energy hν 2 of the second pulsed light is 0.7 eV or more and less than 1.4 eV, the electrons trapped in the surface level are excited, and the electron emission continues by being removed by drift and diffusion in the solid or emitted into the vacuum, achieving high luminance.

[0063] In this case, the timing of the second pulsed light with respect to the first pulsed light 27 may be adjusted to irradiate simultaneously or to delay the irradiation timing of the second pulsed light slightly as shown in FIG. 2B.

Example

[0064] In this example, in the configuration of the electron beam application apparatus of FIG. 1, radially polarized long-wavelength light is used as the second pulsed light. Therefore, a light source that generates a long-wavelength radially polarized beam is used as the second pulsed light source 8. In the radially polarized beam, as shown in FIG. 12A, the electric field direction 55 of the light in the cross-section of the parallel light is radial, and at the focus formed on the photocathode 1 by the condenser lens 2, as shown in FIG. 12B, the electric field direction is perpendicular to the surface of the photoelectric film 10 to become.

[0065] As such methods for generating radial polarization, various techniques have been developed. For example, a linearly polarized light can be converted using a radially / azimuthally transforming plate (Spatially-variant waveplate) or a conical Brew ster prism. In the second pulse light source 8, these optical elements may be applied instead of the optical modulator 23. Additionally, radial polarization can also be generated by using techniques such as applying a pulse voltage to an antenna with a fine structure pattern.

[0066] In a state where the first pulse light 27 irradiates the photoelectric film 10 to cause electron emission, an energy model when a radial polarization beam is irradiated is shown in FIG. 13. When the second pulse light is not irradiated, the vacuum level 60 changes uniformly with respect to the distance z. This is because an electric field is applied between the photocathode 1 and the extraction electrode 3 due to the potential difference therebetween (see FIG. 1 ). When electrons are emitted from the photoelectric film 10 in this state, the velocity of the emitted electrons is small in the immediate vicinity of the surface of the photoelectric film 10 shown as region S in FIG. 13. Therefore, when increasing the emitted electron density, due to the slow velocity of the electrons, in region S, the electrons that came out earlier inhibit the emission of the electrons that come later, or due to the Coulomb force between the two, phenomena such as the energy width spreading or the orbit spreading in the lateral direction occur. These are called the space charge effect, which limits the brightness of the electron source that emits electrons into the vacuum, or broadens the energy distribution, leading to deterioration of the quality of the electron beam for an electron microscope.

[0067] ​On the other hand, when the radially polarized second pulsed light is focused and irradiated, as shown in Fig. 13, a state in which it vibrates temporally between the vacuum level 61 (dashed line) with a negative electric field superimposed on the surface and the vacuum level 62 (dotted line) with a positive electric field superimposed on the surface is obtained. To reduce the space charge effect, it is preferable to accelerate the electrons in the region S in a shorter time. That is, the first pulsed light 27 may be irradiated at the timing when the vacuum level 62 with a positive electric field superimposed thereon, and electrons may be emitted from the photoelectric film 10.

[0068] Fig. 14 shows the irradiation timings of the first pulsed light 27 and the second pulsed light with time on the horizontal axis. The waveform 81 is the intensity I of the first pulsed light 27, and the waveform 82 shows the time change of the electric field F on the surface of the photoelectric film 10 due to the second pulsed light. As described above, by irradiating the first pulsed light 27 in the time zone when the electric field F due to the second pulsed light is positive, the space charge effect can be reduced, and the high luminance of the photoelectric film 10 can be achieved. When the pulse width Tp of the first pulsed light 1 is 1 ps, since the time of the positive part of the vibrating electric field of the second pulsed light needs to be >1 ps, for example, when the vibration period is 2 ps, the frequency is 0.5 THz and the wavelength is 600 μm. The pulse width Tp of the second pulsed light 2 is several ps, and the leading time -Tr of the rise of the second pulsed light with respect to the rise of the first pulsed light 27 is also about several ps. Similar to Example 1, the magnitude of the leading time -Tr is optimized based on the luminance measurement results in the electron optical system.

[0069] As the time of the pulse width of the first pulsed light 27 is shortened, the frequency of the second pulsed light is selected in the terahertz order. For example, when the pulse width of the first pulsed light 27 is shorter, at 0.1 ps in this case, the wavelength of the second pulsed light is 60 μm or more.

[0070] Thus, in this embodiment, electrons are emitted at the timing when a positive electric field is superimposed. As shown in FIG. 13, the electrons are accelerated by the vacuum level 62 with the superimposed positive electric field, and since the energy of the electrons is conserved in vacuum, even if the acceleration varies somewhat due to an electric field that changes with time, the electrons emitted from the same potential energy will have a determined velocity at a location far enough from the photocathode film 10. This means that the spread of the energy of the electron beam emitted from the electron source is narrow and is guided to the electron optical system, which has the advantages of improving the resolution by reducing chromatic aberration and enabling experiments with short pulses.

[0071] FIGS. 15A and 15B show examples of a suitable focusing optical system for the second pulsed light in this embodiment. The first pulsed light 27 is focused by the condenser lens 2 toward the photocathode 1. On the other hand, since the refractive index depends on the wavelength, if the wavelength of the second pulsed light is too long for the condenser lens 2 it may not be possible to sufficiently focus the second pulsed light with the condenser lens 2. In such a case it is effective to use a mirror with low wavelength dependence of light. For example, as shown in FIG. 15A, a reflecting objective lens 56 using a mirror may be used. The second pulsed light passes through a hole in the upper part of the concave mirror 57, is reflected by the convex mirror 58, and is focused on the photocathode 1 by the concave mirror 57. A dichroic mirror is used for the convex mirror 58 so that the first pulsed light 27 can pass through it.

[0072] Alternatively, as shown in FIG. 15B, the second pulsed light may be introduced from a direction different from that of the first pulsed light 27 and focused by a parabolic mirror 59 or the like (in this case, since it is not necessary to combine the optical paths of the first pulsed light 27 and the second pulsed light, the excitation light optical system 21 can be made unnecessary). As shown in FIG. 15B, when the second pulsed light is a parallel beam, a parabolic mirror 59 is used, whereas when the second pulsed light is a divergent beam from a small point source, an ellipsoidal mirror may be used.

[0073] The focusing of the second pulsed light by these mirrors is performed on the lower surface of the photocathode 1, that is, the photo The mirror may be arranged on the opposite side of the condenser lens 2 across the photocathode 1. Further, a reflecting objective lens 56 or a parabolic mirror 59 may be used for focusing the first pulsed light 27.

[0074] The photoelectric film 10 of the photocathode 1 is not limited to GaAs, and may be GaP, AlGaAs, GaN, InGaN, etc., and a metal film such as gold can also reduce the space charge effect during electron emission.

[0075] As described above, the present invention has been specifically described based on the embodiments, but it is not limited thereto, and various modifications can be made without departing from the gist thereof. Each embodiment can be applied alone or in combination with others to achieve an effect. In addition, although a scanning electron microscope has been described as an example of an electron beam application device, the present invention can be applied to various electron beam application devices such as a transmission electron microscope and a scanning transmission electron microscope. Further, the present invention is not limited to an electron beam application device provided with an electron detector for detecting electrons (secondary electrons, reflected electrons, etc.) generated by irradiation with an electron beam, and may be provided with other detectors such as a detector for detecting characteristic X-rays. In any case, by obtaining a high-intensity pulsed electron beam, it becomes possible to further improve the resolution of a microscope image, speed up measurement, and enhance functionality.

Description of Reference Numerals

[0076] 1: Photocathode, 2: Condenser lens, 3: Extraction electrode, 4: Cathode holder, 5: Accelerating Source, 6: Window, 7: First pulsed light source, 8: Second pulsed light source, 9: Vacuum chamber, 10: Photoelectric film, 11: Transparent substrate, 12: Pulsed excitation light, 13: Pulsed electron beam, 14: Opening, 15: Electron gun, 16: Electron optical system housing, 17: Electron lens, 18: Deflector, 19: Electron detector, 20: Sample, 21: Excitation light optical system, 22: Long-wavelength pulsed light source, 23: Optical modulator, 24, 24b: Optical coupler, 25: Reflection light separator , 27: First pulsed light, 28: Second pulsed light, 29: Objective lens, 31: Excitation light control device, 32: Main body control device, 33: Control signal, 34: Central axis, 35: Faraday cup, 36: Current detector, 37: Laser control signal, 38: Aperture, 40: Phase shift region, 41: Image sensor, 42: 1 / 4 Waveplate, 43: Polarizer, 44: Imaging lens, 45: ND filter, 46: Monitor optical system, 47: Reflected light, 51: Fine adjustment mount, 52: Long wavelength pulsed light, 53: Third pulsed light source, 54: Third pulsed light, 55: Electric field orientation, 56: Reflective objective lens, 57: Concave mirror, 58: Convex mirror, 59: Parabolic mirror, 60, 61, 62: Vacuum level, 70: Spiral phase plate, 71 - 0~7: Division region, 81, 82 : Waveform.

Claims

1. A photocathode having a photoelectric film, A first pulse light source that emits a first pulse light, A second pulse light source that emits a second pulse light having a longer wavelength than the first pulse light and having an annular light intensity distribution, An excitation light optical system that combines and emits the optical paths of the first pulse light and the second pulse light, A condenser lens that condenses the first pulse light and the second pulse light emitted from the excitation light optical system onto the photocathode, An electron beam application device comprising an electron optical system that irradiates a sample with an electron beam emitted from the photocathode.

2. In claim 1, The electron beam application device, wherein the wavelength of the second pulse light is a wavelength at which electrons excited in the photoelectric film are recombined with holes by induced emission.

3. In claim 1, The second pulse light source includes an optical modulator that makes the light intensity distribution of the second pulse light annular, The electron beam application device, wherein the optical modulator is a phase mask, a spatial phase modulator, or a spiral phase plate.

4. In claim 1, The excitation light optical system includes a monitor optical system that measures reflected light from the photocathode, The electron beam application device, wherein the monitor optical system includes a quarter-wave plate inserted into the optical paths of the first pulse light and the second pulse light, a reflected light separator inserted into the optical paths of the first pulse light and the second pulse light that separates the reflected light that has passed through the quarter-wave plate from the optical paths of the first pulse light and the second pulse light, and an imaging optical system that forms an image of the reflected light from the reflected light separator on an imaging device.

5. In claim 4, The electron beam application device, wherein the monitor optical system is capable of inserting a polarizer that prevents light having the same polarization direction as the reflected light from the reflected light separator from passing between the imaging optical system and the reflected light separator.

6. In claim 1, A control device that controls the light intensities of the first pulse light and the second pulse light and the irradiation timing of the second pulse light to the photocathode with respect to the irradiation timing of the first pulse light to the photocathode, The electron optical system includes an aperture that limits the opening angle of the electron beam irradiated onto the sample. An electron beam application apparatus in which the light intensity of the second pulsed light and the irradiation timing of the second pulsed light to the photocathode with respect to the irradiation timing of the first pulsed light to the photocathode are optimized based on an image of the sample obtained by irradiating the sample with the electron beam by the electron optical system or the amount of current of the electron beam that has passed through the aperture.

7. A photocathode including a photoelectric film, A first pulsed light source that emits first pulsed light, A second pulsed light source that emits second pulsed light that is infrared light, A condenser lens that condenses the first pulsed light onto the photocathode, An electron beam application apparatus including an electron optical system that irradiates a sample with an electron beam emitted from the photocathode.

8. In claim 7, The photoelectric film includes a wide-gap semiconductor having a bandgap energy of 3 eV or more, An electron beam application apparatus in which the energy of the second pulsed light is lower than the bandgap energy of the wide-gap semiconductor and equal to or higher than the energy of the acceptor level of the wide-gap semiconductor.

9. In claim 7, An electron beam application apparatus in which the wavelength of the second pulsed light is a wavelength capable of exciting electrons trapped at the surface level of the photoelectric film to the conduction band or the vacuum level.

10. A photocathode including a photoelectric film, A first pulsed light source that emits first pulsed light, A second pulsed light source that emits second pulsed light, A third pulsed light source that emits third pulsed light, An excitation light optical system having a first optical coupler that combines the optical path of the second pulsed light and the optical path of the third pulsed light, and a second optical coupler that combines the pulsed light from the first optical coupler and the first pulsed light, A condenser lens that condenses the first pulsed light, the second pulsed light, and the third pulsed light emitted from the excitation light optical system onto the photocathode, An electron beam application apparatus including an electron optical system that irradiates a sample with an electron beam emitted from the photocathode, The photoelectric film includes a wide-gap semiconductor having a bandgap energy of 3 eV or more, An electron beam application apparatus in which the difference between the energy of the second pulsed light and the energy of the third pulsed light is equal to or higher than the energy of the acceptor level of the wide-gap semiconductor.

11. In claim 10, An electron beam application device in which the energy of the second pulsed light and the energy of the third pulsed light are each 1 / 2 or less of the band gap energy of the wide-gap semiconductor.

12. A photocathode provided with a photoelectric film, A first pulsed light source that emits first pulsed light, A second pulsed light source that emits second pulsed light having a longer wavelength than the first pulsed light and radially polarized, A condenser lens that condenses the first pulsed light onto the photocathode, A focusing optical system that condenses the second pulsed light onto the condensation position of the first pulsed light on the photocathode, A control device that controls the irradiation timing of the second pulsed light to the photocathode with respect to the irradiation timing of the first pulsed light to the photocathode, An electron optical system that irradiates a sample with an electron beam emitted from the photocathode, The control device is an electron beam application device that controls to irradiate the first pulsed light in a time zone when the electric field on the surface of the photoelectric film is positive by irradiation of the second pulsed light.

13. In claim 12, An excitation light optical system that combines and emits the optical paths of the first pulsed light and the second pulsed light, The electron beam application device in which the focusing optical system includes a reflective objective lens.

14. In claim 12, The second pulsed light is incident from a direction different from that of the first pulsed light and is condensed by a mirror onto the condensation position of the first pulsed light on the photocathode.

15. An electron beam generation method for generating an electron beam from a photocathode provided with a photoelectric film, The first pulsed light is condensed onto the photocathode by a condenser lens, An electron beam generation method in which second pulsed light having a longer wavelength than the first pulsed light and having an annular light intensity distribution is condensed onto the condensation position of the first pulsed light on the photocathode.

Images