Electron microscope and method for observing a sample

By using laser light source and energy slit technology in electron microscope, the non-destructive problem of observation of oxide layer in resistance-changing elements is solved, and selective observation of oxide layer in resistance-changing elements is achieved, especially the filament electrode Non-destructive observation.

CN111758026BActive Publication Date: 2024-11-12THE UNIV OF TOKYO
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201980014792.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2018-02-23
Filing Date
2019-02-18
Publication Date
2024-11-12
Estimated Expiration
2039-02-18

AI Technical Summary

Technical Problem

In the multi-layer structure of the resistance change element, it is impossible to observe the oxide layer formed at the lower part of the upper electrode without destroying the element. In particular, the transmission electron microscope requires cutting the element, which makes it difficult to observe.

Method used

Using an electron microscope including a laser light source, an irradiation lens system, an energy analyzer, an energy slit and an electron beam detector, the sample is measured by CW laser irradiation, the photoelectrons are converged and separated, and the photoelectrons of specific energy are selected through the energy slit for Detection to achieve non-destructive observation of specific substances.

Benefits of technology

It is possible to selectively observe specific substances located at the bottom of the surface layer of the measurement sample, especially the filament electrodes in the oxide layer, without destroying the measurement sample, providing a non-destructive observation method.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111758026B_ABST
    Figure CN111758026B_ABST
Patent Text Reader

Abstract

The present invention provides an electron microscope capable of non-destructive observation and a method for observing a measurement sample. The electron microscope (1) comprises: a laser light source (2) for generating CW laser light (7); an irradiation lens system (condenser lens (4) and objective lens (6)) for irradiating the CW laser light (7) onto a measurement sample (30); an energy analyzer (22) for separating photoelectrons released from the measurement sample (30) by the CW laser light (7) according to their energy; an energy slit (23) for allowing photoelectrons with a predetermined energy to pass through; an electron beam detector (25) for detecting photoelectrons passing through the energy slit (23); a first electron lens system (21) for converging the photoelectrons released from the measurement sample (30) onto the energy analyzer (22); and a second electron lens system (24) for projecting the photoelectrons passing through the energy slit (23) onto the electron beam detector (25).
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an electron microscope and a method for observing a measurement sample. Background Art

[0002] As a next-generation memory, resistance change memory has attracted attention (see Patent Document 1). The resistance change element used in the resistance change memory has a structure in which an oxide layer such as a transition metal oxide is sandwiched between a lower electrode and an upper electrode. In the setting process, a conductive path (filament) is formed in the oxide layer by applying a voltage, and the state becomes low resistance. In addition, in the reset process, the conductive path in the resistance change layer is broken by applying an adjusted voltage, and the state becomes high resistance.

[0003] It can be considered that in the setting process, the voltage applied to the oxide layer causes a reduction effect, and the metal oxide becomes a metal. The metal becomes a filament, forms a current path, and reduces the resistance value. In the reset process, it can be considered that when the current increases further, the metal reduced by Joule heat begins to oxidize, the filament disappears, and the resistance increases again. For example, the resistance variable element makes the low resistance state correspond to 0 and the high resistance state corresponds to 1. By repeating this cycle, 0 and 1 are repeatedly stored, and it can be used as a memory.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2011-096714 Summary of the invention

[0007] Problems to be solved by the invention

[0008] In a multi-layer structure element such as a resistance variable element, when observing the oxide layer formed below the upper electrode, it is necessary to remove the upper electrode, or cut the element and observe the oxide layer exposed on the cut surface using a transmission electron microscope (TEM). Therefore, there is a problem that observation cannot be performed without destroying the element.

[0009] Therefore, an object of the present invention is to provide an electron microscope and a method for observing a measurement sample that enable non-destructive observation.

[0010] Means for solving problems

[0011] The electron microscope of the present invention includes: a laser light source for generating CW laser light; an irradiation lens system for irradiating the CW laser light to a measurement sample; an energy analyzer for separating photoelectrons released from the measurement sample by the CW laser light according to their energy; an energy slit for allowing the photoelectrons with a predetermined energy to pass through; an electron beam detector for detecting the photoelectrons passing through the energy slit; a first electron lens system for converging the photoelectrons released from the measurement sample onto the energy analyzer; and a second electron lens system for projecting the photoelectrons passing through the energy slit onto the electron beam detector.

[0012] The observation method of the measuring sample of the present invention includes: an irradiation step, irradiating the measuring sample with a CW laser generated by a laser light source; a focusing step, focusing the photoelectrons released from the measuring sample using the CW laser to an energy analyzer; a separation step, separating the photoelectrons according to energy using the energy analyzer; a selection step, irradiating the separated photoelectrons to an energy slit, and selecting the photoelectrons with the predetermined energy; a projection step, projecting the photoelectrons passing through the energy slit to an electron beam detector; and a detection step, detecting the photoelectrons projected to the electron beam detector.

[0013] Effects of the Invention

[0014] According to the present invention, since an energy slit is included for allowing photoelectrons having a predetermined energy to pass through, and an electron beam detector is used to detect the photoelectrons passing through the energy slit, the energy of the electrons detected by the electron beam detector can be selected, and specific substances can be selectively observed. Therefore, specific substances existing in the lowermost layer of the measurement sample can be observed without destroying the measurement sample, thereby realizing non-destructive observation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram showing the overall structure of an electron microscope according to an embodiment of the present invention.

[0016] Figure 2A is a schematic diagram showing the overall structure of a measurement sample according to an embodiment of the present invention, Figure 2B is a schematic diagram showing a cross section of region (i), Figure 2C is a schematic diagram showing a cross section of region (ii), Figure 2D is a schematic diagram showing a cross section of region (iii), Figure 2E It is a schematic diagram showing a cross section of the region (iv).

[0017] Figure 3A This is an image of the junction when the STV is 0.3V. Figure 3B This is the image of the joint when STV is 0V. Figure 3CThis is an image of the junction when the STV is -0.3V.

[0018] Figure 3D This is an image of the junction when the STV is -0.5V. Figure 3E This is an image of the junction when the STV is -1.0V.

[0019] Figure 4 It is a schematic diagram showing the overall structure of an electron microscope according to another embodiment of the present invention. DETAILED DESCRIPTION

[0020] (1) Overall structure of electron microscope according to an embodiment of the present invention

[0021] The electron microscope of this embodiment is a laser photoelectron microscope that detects photoelectrons released from a measurement sample and observes the measurement sample. Figure 1 As shown, the electron microscope 1 of this embodiment includes: a laser light source 2, a wave plate 3, an irradiation lens system composed of a condenser 4 and an objective lens 6, a beam splitter 5, a chamber 10, an energy adjustment mechanism 13, a power supply 14, a first electron lens system 21, an energy analyzer 22, an energy slit 23, a second electron lens system 24, and an electron beam detector 25.

[0022] The laser light source 2 is a laser oscillator that generates CW (Continuous Wave) laser light 7. The wavelength of the CW laser light 7 is selected so that the energy hv of the CW laser light 7 is higher than the work function of the measurement sample 30. This ensures that photoelectrons can be emitted from the measurement sample 30 by irradiation with the CW laser light 7. More specifically, the energy hv of the CW laser light 7 is set to be higher than the work function of the material of the outermost layer constituting the observation area of ​​the measurement sample 30. In this embodiment, the laser light source 2 generates a CW laser 7 having a wavelength of 266 nm (energy hv=4.66 eV). The wave plate 3 is an element for switching the polarization of the CW laser 7 into linear polarization and left-right circular polarization. The wave plate 3 is usually used to set the CW laser 7 to linear polarization, but when the magnetic properties of the sample 30 are measured by magnetic circular dichroism, the wave plate 3 is used to set the CW laser 7 to left-right circular polarization.

[0023] The irradiation lens system converges the CW laser 7 onto the objective lens 6 using the condenser lens 4, converges the CW laser 7 onto the surface of the measurement sample 30 using the objective lens 6, and irradiates the CW laser 7 onto the measurement sample 30. The objective lens 6 is configured so that the focal position is near the surface of the measurement sample 30. The condenser lens 4 and the objective lens 6 are well-known lenses and can be appropriately selected according to the size of the irradiation area of ​​the CW laser 7, that is, the observation area of ​​the measurement sample 30.

[0024] The chamber 10 has a highly airtight structure and is connected to a vacuum pump such as a turbomolecular pump (not shown). The chamber 10 is vacuumed to a predetermined degree (1.0×10 -5 ~10 -8 In the chamber 10, a stage 11 and an objective lens 6 for placing the measurement sample 30 are arranged. In the present embodiment, the chamber 10 is connected to the beam splitter 5, and the objective lens 6 is fixed to the beam splitter 5. Figure 1 For the sake of convenience, the beam splitter 5 and the objective lens 6 are shown as separate structures. A driving mechanism (not shown) is connected to the stage 11, and the stage 11 can be moved in three mutually orthogonal axial directions by the driving mechanism. In the case of this embodiment, the stage 11 is configured so that the mounting surface 11a on which the measurement sample 30 is mounted is orthogonal to the optical axis of the CW laser 7.

[0025] The measurement sample 30 is placed on the mounting surface 11a of the stage 11 in the chamber 10, and the CW laser beam 7 is irradiated perpendicularly to the surface of the measurement sample 30. Here, the measurement sample 30 of this embodiment is described using FIG. 2. Figure 2A As shown, the measurement sample 30 is a resistance variable element composed of a lower electrode 32 formed on a Si substrate 31, an oxide layer 34 formed on the lower electrode 32, and an upper electrode 33 formed on the oxide layer 34. The measurement sample 30 is arranged so that the upper electrode 33 and the lower electrode 32 formed in a rectangular shape are orthogonal to each other in the long axis direction, and a quadrilateral bonding portion 35 is formed.

[0026] like Figure 2B As shown, the joint 35 ( Figure 2A The region (i) shown in FIG. 1 has a three-layer structure of a lower electrode 32, an oxide layer 34, and an upper electrode 33 on a Si substrate 31. The lower electrode 32 is made of TiN (titanium nitride), the oxide layer 34 is made of Ta2O5 (tantalum pentoxide), and the upper electrode 33 is made of Pt (platinum). In the measurement sample 30, when a voltage is applied between the upper electrode 33 and the lower electrode 32, the Ta2O5 of the oxide layer 34 of the joint 35 is reduced to form a Ta2O x The filament is formed. The filament becomes a conductive path between the upper electrode 33 and the lower electrode 32, and the measurement sample 30 becomes a low resistance state. In the case of this embodiment, the thickness of the lower electrode 32, the oxide layer 34 and the upper electrode 33 are 20nm, 5nm, and 10nm respectively, and the area of ​​the joint 35 is approximately 25μm 2 .

[0027] Such a resistance change element can be manufactured, for example, according to the following method. First, a Si substrate 31 is prepared, and a rectangular lower electrode 32 is formed on the Si substrate 31 by sputtering using a mask formed by photolithography. Next, the mask is removed, and an oxide layer 34 is formed on the entire surface of the Si substrate 31 on which the lower electrode 32 is formed by sputtering. Finally, a mask is formed on the oxide layer 34 by photolithography, and a rectangular upper electrode 33 is formed by sputtering so that its long axis is orthogonal to the long axis of the lower electrode 32.

[0028] Therefore, in the region of the upper electrode 33 other than the bonding portion 35 ( Figure 2A In the region (ii) shown in Figure 2C As shown, a two-layer structure of an oxide layer 34 and an upper electrode 33 is formed on a Si substrate 31. In the region of the lower electrode 32 other than the bonding portion 35 ( Figure 2A In the region (iii) shown in Figure 2D As shown in FIG. 1 , a two-layer structure of a lower electrode 32 and an oxide layer 34 is formed on a Si substrate 31. Figure 2A In the region (iii) shown in FIG. 1 , the oxide layer 34 formed on the lower electrode 32 is not shown for convenience. The region ( Figure 2A The area shown in (iv) is as follows Figure 2E The structure shown in FIG. 1 is that an oxide layer 34 is formed on a Si substrate 31 .

[0029] The CW laser light 7 is irradiated onto the surface of the measurement sample 30. In the case of the present embodiment, the irradiation optical system is adjusted so that the CW laser light 7 is irradiated onto the entire joint 35, and the entire joint 35 can be observed at once. As described above, the wavelength of the CW laser light 7 is selected so that the energy hv of the CW laser light 7 is higher than the work function of the outermost layer (Pt in the present embodiment) of the measurement sample 30. Therefore, when the CW laser 7 is irradiated to the measurement sample 30, a photoelectric effect occurs, and the electrons of the measurement sample 30, i.e., the upper electrode 33, the oxide layer 34, and the lower electrode 32 are excited, and photoelectrons are released from the measurement sample 30. A large number of photoelectrons are released from the area irradiated by the CW laser 7, and the photoelectrons are incident on the beam splitter 5. In this specification, the large number of released photoelectrons are collectively referred to as electron beams 27.

[0030] Here, as described above, the wavelength of the CW laser light 7 can be determined based on the work function of the outermost layer of the measurement sample 30. The wavelength is selected, but preferably below 266 nm, more preferably below 213 nm. By selecting the wavelength in this way, the work function can be measured. In addition, if the energy hν of the CW laser 7 is compared with the work function of the outermost layer of the measurement sample 30, The difference is set to It is preferred to select the wavelength of the CW laser 7 so that ΔE is 0 eV to 0.5 eV. In this way, by selecting the wavelength of the CW laser 7, even materials at a deeper position (for example, the oxide layer 34 present under the upper electrode 33) can be observed with a higher resolution. This is because the smaller the energy of the photoelectron, the greater the influence of the energy barrier that the photoelectron experiences when escaping from the surface of the sample. Due to this influence, the obliquely emitted photoelectrons can easily be totally reflected on the surface and only the photoelectrons emitted nearly vertically can form an image. For example, this is beneficial when observing a filament formed on the oxide layer 34 below the upper electrode 33, which is the outermost layer of the measurement sample 30 having a multilayer structure. In addition, ΔE is measured using a laser photoelectron microscope to measure the desired measurement position within the measurement sample 30 and based on the electron energy distribution of the measurement position obtained using an energy analyzer. ΔE is related to the width of the band structure within the electron energy distribution (from the cutoff point to the Fermi level E described later). F )quite.

[0031] In addition, the electron microscope 1 is provided with a power supply 14, the negative side of the power supply 14 is connected to the stage 11, and the positive side is provided at the ground G, so that a negative voltage can be applied to the measurement sample 30. The power supply 14 is a general power supply that can output a high voltage. In the case of the present embodiment, a voltage of -20 kV is applied to the measurement sample 30 by the power supply 14. Therefore, an electric field is generated between the measurement sample 30 and the beam splitter 5 to which no voltage is applied, and photoelectrons are easily released from the measurement sample 30 by the electric field, and the released photoelectrons are accelerated toward the beam splitter 5, and the electron beam 27 is introduced into the beam splitter 5.

[0032] The electron microscope 1 has an energy adjustment mechanism 13 between the power supply 14 and the stage 11. The energy adjustment mechanism 13 is a power supply that outputs a predetermined voltage STV. The energy adjustment mechanism 13 is connected in series with the power supply 14, and can apply the total voltage of STV and the output voltage of the power supply 14 to the measurement sample 30. The energy adjustment mechanism 13 can adjust the energy Ep of the photoelectrons released from the measurement sample 30 by adjusting the value of STV. In addition, in the case of the present embodiment, if the kinetic energy of the photoelectrons is Ek, the energy Ep of the photoelectrons is Ep=20kV+Ek-STV. The kinetic energy Ek of the photoelectrons is the value of the kinetic energy generated by the electrons in the measurement sample 30 being excited by the CW laser 7, and varies according to the energy E of the electrons in the substance. Therefore, the energy Ep of the photoelectrons also depends on the energy E of the electrons in the substance.

[0033] When the electron beam 27 is incident, the beam splitter 5 deflects the direction of the electron beam 27, so that the path of the CW laser 7 and the path of the electron beam 27 are separated. The beam splitter 5 causes the electron beam 27 to be incident on the first electron lens system 21 connected to the exit port of the deflected electron beam 27. The first electron lens system 21 is composed of a plurality of electron lenses, and converges the incident electron beam 27. One end of the first electron lens system 21 is connected to the beam splitter 5, and the other end is connected to the energy analyzer 22, so that the electron beam 27 is converged and incident on the energy analyzer 22.

[0034] The energy analyzer 22 is a well-known energy analyzer, which separates the incident electron beam 27 according to the energy Ep of the photoelectron, and emits the electron beam 27 separated according to the energy Ep. The energy analyzer 22 has a hemispherical shape, and the entrance and exit of the beam are arranged in the flat part of the hemisphere. The first electron lens system 21 is connected to the entrance of the energy analyzer 22, and the second electron lens system 24 is connected to the exit of the energy analyzer 22. The energy analyzer 22 separates the electron beam 27 incident from the first electron lens system 21 according to the energy Ep of the photoelectron and emits it to the second electron lens system 24.

[0035] An energy slit 23 is provided at the exit port of the energy analyzer 22. The energy slit 23 is a conventional slit in which a through groove is formed linearly on the surface of a plate-like member. The energy slit 23 allows the electron beam 27 irradiated to the groove to pass through, and blocks the electron beam 27 irradiated to the plate-like member. In addition, in reality, the electron beam 27 irradiated to the plate-like member is not completely blocked, and a part of these electron beams 27 also passes through the energy slit 23. Therefore, the intensity of the electron beams 27 other than the electron beam 27 irradiated to the groove is reduced by the energy slit 23. In the present embodiment, the width of the groove of the energy slit 23 is 40 μm.

[0036] Since such an energy slit 23 is arranged at the exit port of the energy analyzer 22, the electron beam 27 that passes through the energy slit 23 among the electron beams 27 separated by the energy analyzer 22 enters the second electron lens system 24. At this time, since the electron beam 27 is separated by the energy analyzer 22 according to the energy Ep of the photoelectron, the passing position of the electron beam 27 in the exit port is also determined according to the energy Ep of the photoelectron. Therefore, by adjusting the position of the energy slit 23, the energy Ep of the photoelectron detected by the electron beam detector 25 can be selected. Since the energy Ep of the photoelectron depends on the energy E of the electron in the substance (in the measurement sample 30), by changing the position of the energy slit 23, the energy E in the measurement sample 30 can be selected, and the electrons in the measurement sample 30 to be detected can be selected.

[0037] On the other hand, since the energy Ep of the photoelectrons is 20 kV + Ek - STV, the energy Ep of the photoelectrons can be changed by changing the value of STV. Therefore, by changing the value of STV, the energy Ep of the photoelectrons detected by the electron beam detector 25 can be selected, and the electrons in the measurement sample 30 to be detected can be selected.

[0038] The second electron lens system 24 is composed of a plurality of electron lenses, and projects the incident electron beam 27 onto the electron beam detector 25. The electron beam detector 25 is a two-dimensional photoelectron detector, which detects the photoelectrons of the projected electron beam 27, and generates an image of the measurement sample 30 based on the intensity of the detected photoelectrons. Since the electron beam detector 25 generates images based on the projected electron beam 27, images can be continuously generated, and in addition to still images, moving images can also be generated. The electron beam detector 25 is connected to a PC (not shown), and the generated image can be sent to the PC, and the image can be saved in the storage device of the PC or displayed on the monitor of the PC for confirmation by the operator of the electron microscope 1.

[0039] An example of observing the measurement sample 30 using such an electron microscope 1 is shown in FIG. FIG. 3A to FIG. 3C , FIG. 3D to FIG. 3E In addition, the resistance variable element is in a low resistance state in which a filament is formed by applying a predetermined voltage between the upper electrode 33 and the lower electrode 32 in advance. FIG. 3A to FIG. 3C , FIG. 3D to FIG. 3E The upper image in the figure is an image of the joint 35 taken by an electron microscope 1, and the lower figure of the image is a schematic diagram showing the state density of the material constituting the measurement sample 30, wherein the horizontal axis represents the energy E of the electrons of the measurement sample 30, and the vertical axis represents the intensity of the electrons (i.e., the number of states). FIG. 3A to FIG. 3C , FIG. 3D to FIG. 3E E in the diagram representing the density of states F is the Fermi level, and the two solid lines represent the band EB of energy E of electrons that can pass through the energy slit 23 as photoelectrons. FIG. 3A to FIG. 3C , FIG. 3D to FIG. 3E The band EB shown means that more electrons with energy E within the band EB can pass through the energy slit 23, and a part of the electrons with energy E outside the band EB can also pass through the energy slit 23. In addition, outside the band EB, electrons with energy E close to the lower limit or upper limit of the band EB can easily pass. In addition, the cutoff point (Cut off) of the state density curve is the energy at which the kinetic energy of the released photoelectrons is zero.

[0040] Figure 3A This is an image of the junction 35 when the STV is 0.3 V. Figure 3AIn the image, the boundary between the upper electrode 33 made of Pt and the joint 35 located in the upper left of the image can be observed, and it can be confirmed that there is a certain structure in the lower right of the image. Moreover, the image becomes a low brightness value as a whole. In this case, Figure 3A As shown in the state density curve at the bottom, the energy E of the electrons that can pass through the energy slit 23 as photoelectrons is in the frequency band EB that is lower than the Fermi level E F In the energy E in the band EB, any substance constituting the measurement sample 30 does not exist, and electrons having the energy E in the band EB do not exist in the measurement sample 30. Therefore, no photoelectrons easily pass through the energy slit 23, and the number of photoelectrons detected by the electron beam detector 25 is small, resulting in an image with a low overall brightness value.

[0041] Figure 3B This is an image of the joint 35 when STV is 0V. Figure 3B , you can Figure 3B The part with higher brightness value is found in the position indicated by the arrow in the figure. This can be considered that the brightness value of the structure observed when STV is 0.3V becomes higher. In this case, the band EB is located at the Fermi level E F Nearby, there are Pt and Ta2O in band EB x (filament) state, in Pt and Ta2O x There are electrons with energy E in band EB in Ta2O5. On the other hand, there is no state in Ta2O5 within band EB, and there are no electrons with energy E in band EB in Ta2O5. Therefore, due to the x The released photoelectrons can easily pass through the energy slit 23, so in the electron beam detector 25, the Pt and Ta2O x The detected amount of photoelectrons released from Pt is large, and the detected amount of photoelectrons released from Ta2O5 is small. As a result, in the image, Pt and Ta2O x The brightness value of Ta2O5 is high, while the brightness value of Ta2O5 is low. F Nearby, Ta2O x The state density is greater than that of Pt, so the number of photoelectrons released is also greater. Therefore, in the electron beam detector 25, Ta2O x The released photoelectrons are detected more frequently and in the image Ta2O x The brightness value of the part is the highest. Figure 3B The structure observed in the figure has the highest brightness value, so it can be considered to be composed of Ta2O x The one formed can be considered as the filament.

[0042] Thus, since the electron microscope 1 has the energy slit 23, the energy E of the electrons to be detected can be selected, and thus the oxide layer 34 composed of Ta2O5 can be selectively observed. x The constructed filament can further be observed in the oxide layer 34 formed under the upper electrode 33, and specific substances can be selectively and non-destructively observed.

[0043] In addition, compared with the image when STV is 0.3V, Figure 3B In the image, the brightness value of the entire image becomes higher. It can be considered that this is because the photoelectrons released from Pt easily pass through the energy slit 23, which makes the brightness value of the upper electrode 33 in the image higher. Moreover, the boundary between the upper electrode 33 and the joint 35 can be confirmed in the upper left and lower right of the image. In addition, the lower electrode 32 should exist in the lower left and upper right of the image, but because there are fewer electrons passing through the energy slit 23, it becomes a dark part in the image and cannot be confirmed. Since the joint 35 and the upper electrode 33 are brighter, the boundary between the lower electrode 32 and the joint 35 can be confirmed.

[0044] Figure 3C This is an image of the junction 35 when the STV is -0.3 V. Figure 3C When the image is taken, the brightness value of the upper electrode 33 portion becomes higher. The boundary between the upper electrode 33 and the joint 35 can be confirmed more clearly. In addition, it can be confirmed that the brightness value of the filament has decreased. In addition, a plurality of granular portions can be observed in the joint 35. In this case, the band EB is on the lower energy side than when the STV is 0 V, and Pt and Ta2O exist in the band EB. x state, in Pt and Ta2O x There are electrons with energy E in the band EB. Compared with the case where STV is 0 V, the state in the band EB is dominated by Pt. Compared with Pt, Ta2O x Therefore, electrons with energy E in the band EB are more present on the Pt side, and as a result, more photoelectrons released from Pt are detected by the electron beam detector 25. In the image, the brightness value of the upper electrode 33 composed of Pt becomes higher, while the brightness value of the filament becomes lower. In addition, since the granular part is observed due to the increase in photoelectrons released from Pt, it can be considered that it is Pt.

[0045] Figure 3D This is an image of the junction 35 when the STV is -0.5 V. Figure 3DWhen the image is taken, it can be confirmed that the brightness values ​​of the upper electrode 33 (right side in the image) and the lower electrode 32 (upper part in the image) in the image become higher. In addition, no filament is observed. In this case, the band EB is at a lower energy side than when the STV is -0.3 V, and Pt and TiN exist in the band EB, and Ta2O does not exist. x In other words, there are electrons with energy E within the band EB in Pt and TiN, and the photoelectrons released from Pt and TiN easily pass through the energy slit 23. The number of electrons with energy E within the band EB is roughly the same in Pt and TiN. Therefore, it can be considered that the brightness values ​​of the upper electrode 33 and the lower electrode 32 in the image become higher. In addition, it can be considered that since there is no Ta2O x state, so from Ta2O x The detected amount of released photoelectrons decreases, and the filament cannot be observed.

[0046] Figure 3E This is an image of the junction 35 when the STV is -1.0 V. Figure 3E When an image is taken, it can be confirmed that the brightness value of the lower electrode 32 (upper part in the image) in the image becomes higher. In this case, the band EB is at a lower energy side than when STV is -0.5V, and TiN and Si exist in the band EB. In other words, electrons with energy E in the band EB exist in TiN and Si, and photoelectrons released from TiN and Si easily pass through the energy slit 23. From the energy density point of view, TiN dominates among the electrons with energy E in the band EB. Therefore, it can be considered that the brightness value of the lower electrode 32 in the image becomes higher.

[0047] In this way, by determining STV based on the state density so that the frequency band EB of the energy E of electrons that can pass through the energy slit 23 as photoelectrons becomes such an energy band in which the state density of the specific substance to be detected is high, the amount of photoelectrons released from the specific substance is increased, the photoelectrons can be easily detected, and the specific substance can be selectively observed.

[0048] In the above structure, the electron microscope 1 is configured to include: a laser light source 2, which generates a CW laser 7; an irradiation lens system (a condenser 4 and an objective lens 6), which irradiates the CW laser 7 to the measurement sample 30; an energy analyzer 22, which separates the photoelectrons released from the measurement sample 30 by the CW laser 7 according to the energy Ep; an energy slit 23, which allows photoelectrons with a predetermined energy Ep to pass through; an electron beam detector 25, which detects the photoelectrons passing through the energy slit 23; a first electron lens system 21, which converges the photoelectrons released from the measurement sample 30 to the energy analyzer 22; and a second electron lens system 24, which projects the photoelectrons passing through the energy slit 23 to the electron beam detector 25.

[0049] Therefore, the electron microscope 1 includes an energy slit 23 that allows photoelectrons having a predetermined energy Ep to pass through, and the photoelectrons that pass through the energy slit 23 are detected by the electron beam detector 25, so that the energy E of the electrons detected by the electron beam detector 25 can be selected, and a specific substance can be selectively observed. Therefore, the electron microscope 1 can observe the filament of the oxide layer 34 that exists under the upper electrode 33 of the outermost layer of the measurement sample 30 without destroying the upper electrode 33, and non-destructive observation can be achieved.

[0050] Furthermore, since the electron microscope 1 projects the photoelectrons that have passed through the energy slit 23 to the electron beam detector 25 through the second electron lens system 24 and detects them, the entire measurement area can be observed at once without scanning, and the measurement sample 30 can be continuously observed. Therefore, when the measurement sample 30 is the above-mentioned variable resistance element, for the variable resistance element in which a filament is not formed, by observing the oxide layer 34 with the electron microscope 1 while applying a voltage between the upper electrode 33 and the lower electrode 32, the formation of the filament can be observed.

[0051] (2) Other implementation methods.

[0052] In the above-mentioned embodiment, the case where the CW laser light 7 is irradiated perpendicularly to the measurement sample 30 and the path of the electron beam 27 formed by the photoelectrons released from the measurement sample 30 is separated from the path of the CW laser light 7 by the beam splitter 5 is described, but the present invention is not limited thereto, and the electron microscope 1 may not include the beam splitter 5. For example, in the case where the photoelectrons are released perpendicularly to the measurement sample 30 to detect the photoelectrons, the laser light source 2 is configured so that the path of the CW laser light 7 is inclined at a predetermined angle (for example, 45 degrees) relative to the perpendicular direction of the measurement sample 30, and the CW laser light 7 is irradiated to the measurement sample 30.

[0053] In the above embodiment, the power source 14 is connected to the stage 11, a negative voltage is applied to the measurement sample 30, and an electric field is generated between the measurement sample 30 and the beam splitter 5 to facilitate the emission of photoelectrons. However, the present invention is not limited to this. Figure 1 The same structure is assigned the same symbol Figure 4As shown in the electron microscope 40, the positive side of the power supply 14 is connected to the first electron lens system 21, and the negative side is grounded to the ground G, and a positive voltage is applied to the first electron lens system 21, so that the first electron lens system 21 is charged with a positive voltage, and an electric field is generated between the measurement sample 30 and the beam splitter 5 so that photoelectrons are easily released from the measurement sample 30. In this case, the energy adjustment mechanism 13 is also grounded to the ground G. Since the voltage is applied to the first electron lens system 21 side instead of the measurement sample 30, it is no longer necessary to apply the voltage to the measurement sample 30, which is advantageous for industrialization. In addition, the power supply 14 can also be connected to the beam splitter 5 to apply the voltage to the beam splitter 5.

[0054] In addition, in the above embodiment, the energy slit 23 is described as being fixed, but the present invention is not limited to this. Figure 4 As in the electron microscope 40 shown in the figure, the slit moving mechanism 29 is provided at the energy slit 23 , and the position of the energy slit 23 is moved to change the energy band EB of the electrons passing through the energy slit 23 .

[0055] (3) Uses of electron microscopes.

[0056] In the above description, the resistance variable element as the measurement sample 30 is observed using the electron microscope 1, and the filament formed in the oxide layer 34 between the upper electrode 33 and the lower electrode 32 can be observed, and the resistance state of the resistance variable element can be visualized. This is because the energy of the electron is near the Fermi level, and the Ta2O5 constituting the filament formed in the oxide layer 34 is higher than the Pt constituting the upper electrode and the Ta2O5 constituting the oxide layer 34. x The state density is higher, and the intensity of the photoelectrons detected by the electron beam detector 25 is higher. Therefore, by setting the energy slit 23 that mainly allows electrons with energy near the Fermi level to pass, the electrons from Ta2O can be selectively detected. x Thus, the electron microscope 1 can selectively observe a substance having a higher state density than other substances according to a predetermined electron energy, and thus, this property can be used for other purposes as well.

[0057] For example, the resistance state of a phase change element constituting a phase change memory (PCM) can be visualized. A phase change element is an element that stores information by utilizing the characteristic that a phase change material such as a chalcogenide alloy has different resistance values ​​in a crystalline state (low resistance state) and an amorphous state (high resistance state). In a phase change element, a layer made of a phase change material is sandwiched between an upper electrode and a lower electrode, and the resistance state is changed by the Joule heat generated by the flow of current between the upper electrode and the lower electrode. In a phase change material, the work function and the state density near the Fermi level are different in the crystalline state and the amorphous state, and the amount of photoelectrons released is different. Therefore, the energy slit 23 and the STV are set so that electrons with energy near the Fermi level mainly pass through. When observing the resistance change element using an electron microscope 1, the crystalline state can be selectively observed, and the change between the crystalline state and the amorphous state can be visualized. In addition, a voltage can be applied between the upper electrode and the lower electrode to continuously observe the change between the crystalline state and the amorphous state.

[0058] In addition, the electron microscope 1 can be used to evaluate the magnetoresistive element constituting the magnetic random access memory. The magnetoresistive element has a three-layer structure of a ferromagnetic layer, an insulating layer and a ferromagnetic layer. The magnetoresistive element is formed so that one ferromagnetic layer is used as a fixed layer and the magnetization direction is fixed, and the other ferromagnetic layer is used as a free layer and the magnetization direction changes due to an external magnetic field and spin torque. The magnetoresistive element is an element that stores information by utilizing the characteristics that the resistance is low when the magnetization directions of the fixed layer and the free layer are parallel and the resistance is high when the magnetization directions of the fixed layer and the free layer are antiparallel.

[0059] The electron microscope 1 can visualize the magnetic domain structure of the free layer (e.g., Fe epitaxial film) of such a magnetoresistive element. In this case, in the electron microscope 1, the magnetic domain structure is visualized by utilizing the different characteristics of the amount of photoelectrons released when irradiating a laser with left circularly polarized light and the amount of photoelectrons released when irradiating a laser with right circularly polarized light according to the magnetization direction. Specifically, the CW laser 7 irradiated to the magnetoresistive element as the measurement sample 30 is switched to left-handed circularly polarized light and right-handed circularly polarized light using a wave plate 3 and observed separately to obtain the intensity difference of the two images obtained, thereby extracting only the magnetic information to realize the visualization of the magnetic domain structure. In this case, since the electron microscope 1 has an energy slit 23, the chromatic aberration of the electron lens can be reduced and the measurement can be performed with high resolution. In this case, even if the free layer exists in the lower part of the fixed layer or the insulating layer, it can be observed.

[0060] In addition, the electron microscope 1 can detect defects in the insulating layer of the magnetoresistive element. The defective portion of the insulating layer has a larger state density near the Fermi level than other portions of the insulating layer. Therefore, by setting the energy slit 23 and STV to mainly allow electrons with energy near the Fermi level to pass, the detection intensity of the photoelectrons released from the defective portion can be made to exceed the detection intensity of the photoelectrons released from the insulating layer, and the metal portion in the insulator can be selectively observed. Therefore, the electron microscope 1 can detect defects in the insulating layer of the magnetoresistive change element. In addition, by observing the magnetoresistive element while applying voltage to the fixed layer and free layer of the magnetoresistive element, the formation of defects caused by insulation breakdown can also be observed.

[0061] In addition, the electron microscope 1 can be used for defect inspection of the capacitor layer of NAND-Flash. In NAND-Flash, the state density of the Fermi level is larger in the part with crystal defects than in other parts. Therefore, by setting the energy slit 23 and STV to mainly allow electrons with energy near the Fermi level to pass, the detection intensity of the photoelectrons released from the crystal defects can be made to exceed the detection intensity of the photoelectrons released from the parts other than the crystal defects, and the crystal defects in the capacitor layer can be selectively observed. Therefore, the electron microscope 1 can detect the crystal defects of the capacitor layer of NAND-Flash. In addition, the capacitor layer of NAND-Flash can be observed while the NAND-Flash is powered on to observe the crystal defects of the capacitor layer.

[0062] In addition, the electron microscope 1 can be used to inspect the structure of the metal mask formed in the glass. The state density of the Fermi level is larger in the metal mask portion in the glass than in the glass portion. Therefore, by setting the energy slit 23 and STV to mainly allow electrons with energy near the Fermi level to pass, the detection intensity of the photoelectrons released from the metal mask can be made to exceed the detection intensity of the photoelectrons released from the glass, and the metal mask in the glass can be selectively observed. Therefore, the electron microscope 1 can inspect the structure of the metal mask formed in the glass.

[0063] In addition, the electron microscope 1 can inspect metal wiring (e.g., Cu) in an insulating material (e.g., low-k). The state density of the metal wiring in the insulating material is larger than that of the insulating material at the Fermi level. Therefore, by setting the energy slit 23 and STV to mainly allow electrons with energy near the Fermi level to pass, the detection intensity of photoelectrons released from the metal wiring can be made to exceed the detection intensity of photoelectrons released from the insulating material, and the metal wiring in the insulating material can be selectively observed. Therefore, the electron microscope 1 can inspect the metal wiring in the insulating material.

[0064] Explanation of symbols

[0065] 1: Electron Microscope

[0066] 2: Laser light source

[0067] 7: CW laser

[0068] 13: Energy adjustment mechanism

[0069] 14: Power supply

[0070] 21: The first electron lens system

[0071] 22: Energy Analyzer

[0072] 23: Energy Slit

[0073] 24: Second electron lens system

[0074] 25: Electron beam detector

[0075] 27: Electron beam

Claims

1. Electron microscope, including: A laser light source for generating CW laser; An irradiation lens system irradiates the CW laser to the measurement sample; an energy analyzer for separating photoelectrons emitted from the measurement sample by the CW laser according to energy; An energy slit for allowing the photoelectrons with predetermined energy to pass through; an electron beam detector for detecting the photoelectrons passing through the energy slit and generating an image of the measurement sample based on the intensity of the detected photoelectrons; a first electron lens system for focusing the photoelectrons emitted from the measurement sample onto the energy analyzer; a second electron lens system, which projects the photoelectrons passing through the energy slit onto the electron beam detector; a power source for applying a voltage to the measurement sample or the first electron lens system to accelerate the photoelectrons released from the measurement sample; as well as an energy adjustment mechanism that determines a voltage to be applied to the measurement sample according to a state density of a specific substance constituting the measurement sample, and applies the determined voltage to the measurement sample so that a frequency band of energy of photoelectrons passing through the energy slit becomes a frequency band of energy of photoelectrons released from the specific substance, The energy slit is provided with a slit moving mechanism to move the position of the energy slit so as to change the energy band of the photoelectrons passing through the energy slit. 2 . The electron microscope according to claim 1 , wherein the difference between the energy of the CW laser and the work function of the measurement sample is 0 eV to 0.5 eV. 3 . The electron microscope according to claim 1 , wherein the power supply charges the first electron lens system to a positive voltage, and introduces the photoelectrons generated in the measurement sample into the first electron lens system.

4. Observation methods for measuring samples, including: an irradiation step of irradiating a CW laser generated by a laser light source to the measurement sample; A focusing step of focusing the photoelectrons released from the measurement sample by the CW laser light onto an energy analyzer using a first electron lens system; A separation step, using the energy analyzer to separate the photoelectrons according to energy; A selection step of irradiating the separated photoelectrons to an energy slit to select the photoelectrons having a predetermined energy; a projecting step, using a second electron lens system to project the photoelectrons passing through the energy slit to an electron beam detector; a detection step of detecting the photoelectrons projected onto the electron beam detector and generating an image of the measurement sample based on the intensity of the detected photoelectrons; A step of applying a voltage to the measurement sample or the first electron lens system by a power supply to accelerate the photoelectrons released from the measurement sample; as well as The voltage applied to the measurement sample is determined by an energy adjustment mechanism according to the state density of a specific substance constituting the measurement sample, and the determined voltage is applied to the measurement sample so that the energy band of the photoelectrons passing through the energy slit becomes the energy band of the photoelectrons released from the specific substance. Wherein, in the selection step, the position of the energy slit is moved by a slit moving mechanism provided on the energy slit to change the energy band of the photoelectrons passing through the energy slit. 5 . The method for observing a measurement sample according to claim 4 , wherein a difference between the energy of the CW laser and a work function of the measurement sample is 0 eV to 0.5 eV. 6 . The observation method of a measurement sample according to claim 4 , wherein a positive voltage is applied to the first electron lens system by the power supply to introduce the photoelectrons generated in the measurement sample into the first electron lens system.