Plasma measuring device

Abstract

This plasma measurement device comprises: a laser light source 20 that emits laser light LA toward plasma PL; a spectrum measurement unit 30 that receives plasma scattered light LP from the plasma PL and measures the wavelength spectrum of such light; and an arithmetic unit 40 that calculates the electron temperature and the electron density of the plasma PL on the basis of the measured wavelength spectrum. The spectrum measurement unit 30 comprises a spectrometer having an incidence slit S1. The incidence slit S1 includes: two blades 31, 32 that are movable in an X direction within an XY plane perpendicular to the traveling direction of the plasma scattered light LP; and at least one blade 33 that is movable in a Y direction perpendicular to the X direction. The slit shape is configured using the edges of at least the three blades 31–33. This configuration makes it possible to perform more accurate measurements while easing structural constraints on plasma measurement.

Description

[Technical Field]

[0001] This invention relates to a plasma measuring device for measuring the electron temperature and electron density of a plasma. [Background technology]

[0002] The electron density n is a fundamental physical quantity of plasma. e and electron temperature T e Laser Thomson scattering (LTS) is a method for non-contact, time- and spatially resolved measurements of [the substance] (for example, see reference (K. Muraoka and A Kono: J. Phys. D: Appl. Phys. 44 043001 (2011))).

[0003] The LTS method forces free electrons in a plasma to vibrate using laser light, and then analyzes the scattered light, which is a secondary electromagnetic wave from that vibration. Because it is a "scattering" measurement observed from a direction different from the optical axis of the laser light, local information can be obtained, and a time resolution of a few nanoseconds can be easily achieved. Since the scattered light undergoes a Doppler shift that reflects the thermal velocity of electrons, the spectral broadening can be determined from the T, similar to Rayleigh scattering from neutral particles. e This allows us to determine the thermal fluctuation component of electrons, and therefore the scattered light intensity and n e It is proportional, n e It is obtained as an absolute value.

[0004] In order to implement the LTS method, since the scattering cross-section of free electrons is low, firstly, a high-intensity pulsed laser beam (output > 10) is required. 11 W / cm 2 Firstly, it is necessary to focus the laser beam onto the plasma. Secondly, a mechanism is needed to minimize stray light (a portion of the laser light that is scattered by the plasma electrodes or window materials) and separate it from the weak LTS light.

[0005] For obtaining a high spatial resolution, the scattered light angle often has a light receiving window provided in a direction close to 90° with respect to the incident direction of the laser beam. In some cases, an element (aperture) is provided to limit the optical path for the input and output of the laser beam so that the reflected light of the laser at the surface of the window does not enter the spectrometer through the light receiving lens. In some cases, incident and exit windows with a Brewster angle are adopted for the purpose of suppressing reflection on the window material.

[0006] On the other hand, actual plasma devices often do not have a structure suitable for LTS measurement. For example, as a light source in the soft X-ray or extreme ultraviolet (EUV) region, a plasma with a high electron density (n e >10 23 m -3 ) and a high electron temperature (Te > 20 eV) may be used. As the plasma generation methods, there are mainly a method using a laser, a method using pinch discharge by a pulsed current, and a method using both of them.

[0007] For the structure for LTS measurement, (i) a window for incident measurement laser beam, (ii) a window for receiving Thomson scattered light, and (iii) a window for discharging the measurement laser beam are each required. Here, the windows of (i) and (ii) can often use the windows already installed for the purpose of observing the plasma state, but the window of (iii) may not be usually prepared. In such a case, it is assumed that the plasma device is modified so that LTS measurement can be applied, but there is a possibility that the characteristics of the plasma, which are crucial, may change and the data that was originally desired cannot be obtained. Therefore, in order to expand the applicable range of LTS measurement, it is necessary to relax the structural constraints required for the device.

Prior Art Documents

Patent Documents

[0008]

Patent Document 1

Non-Patent Document 1

Non-Patent Document 2

Non-Patent Document 3

Non-Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0010] An object of the present invention is to provide a plasma measurement device capable of performing more accurate measurement while relaxing structural constraints on plasma measurement.

Means for Solving the Problems

[0011] A plasma measurement device according to an aspect of the present invention includes a laser light source that irradiates laser light toward a plasma, a spectral measurement unit that receives scattered light from the plasma and measures its wavelength spectrum, and an arithmetic unit that calculates the electron temperature and electron density of the plasma based on the measured wavelength spectrum. The spectral measurement unit includes a spectroscope having an entrance slit, the entrance slit includes two blades movable in a first direction within a plane perpendicular to the traveling direction of the scattered light, and at least one blade movable in a second direction perpendicular to the first direction, and the slit shape is set by the edges of at least three blades.

Effects of the Invention

[0012] According to the present invention, it is possible to perform measurement with higher accuracy while relaxing the structural constraints of plasma measurement.

Brief Description of the Drawings

[0013] [Figure 1] It is a configuration diagram showing an example of a plasma measurement apparatus according to the present invention. [Figure 2] It is a configuration diagram showing another example of a plasma measurement apparatus according to the present invention. [Figure 3] It is a configuration diagram showing an example of a spectrum measurement unit. [Figure 4] It is a graph showing an example of a wavelength spectrum signal of a photodetector. [Figure 5] It shows a partially enlarged view of FIG. 2. [Figure 6] It is an arrangement diagram showing an example of an incident slit of a spectroscope. [Figure 7] FIG. 7(A) is a front view showing a specific example of a slit, and FIG. 7(B) is a side view thereof. [Figure 8] It is a graph showing an example of a wavelength spectrum of plasma scattered light. [Figure 9] It is a graph showing the result of actually measuring the influence of noise scattered light in the apparatus of FIG. 2. [Figure 10] It is a perspective view showing an example of the arrangement of LTS measurement. [Figure 11] FIGS. 11(A) to (D) show shadow graphs of a Sn target. FIG. 11(A) shows the state before irradiation with laser light, FIG. 11(B) shows the state at a delay time Δt = 1.3 μs, FIG. 11(C) shows the state at a delay time Δt = 2.0 μs, and FIG. 11(D) shows the state at a delay time Δt = 2.5 μs. FIGS. 11(E) to (G) show EUV emission distributions. [Figure 12]Figure 12(A) shows the EUV emission distribution corresponding to Figure 11(E). Figure 12(B) shows the distribution of electron density ne (upper graph) and electron temperature Te (lower graph) obtained by LTS measurement. Figure 12(C) shows the emissivity distribution calculated based on the electron density ne and electron temperature Te in Figure 12(B). Figure 12(D) is a reconstructed graph of the graph in Figure 12(C). Figure 12(E) is a graph showing the EUV emission. [Modes for carrying out the invention]

[0014] Figure 1 is a diagram showing an example of a plasma measurement device according to the present invention. The plasma measurement device comprises a vacuum vessel 10, a laser light source 20, a spectrum measurement unit 30, a calculation unit 40, and the like.

[0015] The vacuum vessel 10 has the function of containing plasma PL inside and is maintained at a vacuum level of, for example, 10 Torr or less. Plasma PL can be generated by methods such as 1) heating the target by irradiating it with pulsed laser light, or 2) heating the target by current drive by pulsed discharge.

[0016] The laser light source 20 has the function of generating laser light LA ​​for measurement toward the plasma PL, and for example, it generates the second harmonic (532 nm) of a YAG laser.

[0017] The spectrum measurement unit 30 has the function of receiving plasma scattered light LP generated by plasma PL and measuring its wavelength spectrum.

[0018] The calculation unit 40 is composed of, for example, a computer, and is connected to the laser light source 20, the spectrum measurement unit 30, etc. in a communicative manner. Based on the wavelength spectrum measured by the spectrum measurement unit 30, it calculates the electron temperature and electron density of the plasma PL.

[0019] The walls of the vacuum vessel 10 are provided with a number of windows 11 to 13 made of transparent material. For example, window 11 can be used to introduce a laser beam LA for measurement into the vacuum vessel 10. Window 12 can be used to discharge the laser beam that has passed through the plasma PL to the outside of the vacuum vessel 10. Window 13 can be used to introduce plasma scattered light LP from the plasma PL into the spectrum measurement unit 30. When the laser beam LA passes through windows 11 and 12, noise scattered light is generated due to surface roughness, dirt, etc., of windows 11 and 12.

[0020] Figure 2 is a configuration diagram showing another example of a plasma measurement device according to the present invention. The plasma measurement device has a configuration similar to that of Figure 1 and includes a vacuum vessel 10, a laser light source 20, a spectrum measurement unit 30, a calculation unit 40, and the like.

[0021] The walls of the vacuum vessel 10 are provided with multiple windows 11 and 13 made of transparent material. For example, window 11 can be used to introduce laser light LA ​​for measurement into the vacuum vessel 10. Window 13 can be used to introduce plasma scattered light LP from plasma PL into the spectrum measurement unit 30. On the other hand, window 12 shown in Figure 1 is not present.

[0022] Furthermore, an object 15 (for example, an electrode holder) is placed behind the plasma PL. The laser light that has passed through the plasma PL collides with object 15. At that time, a large amount of noise scattered light is generated from object 15 toward the surroundings.

[0023] Figure 3 is a configuration diagram showing an example of the spectral measurement unit 30. The spectral measurement unit 30 comprises a spectrometer including multiple diffraction gratings G1 to G5, multiple slits S1 to S3, multiple lenses L1 to L6 and multiple mirrors M1 to M3, and a photodetector DT. Slit S1 functions as an incident slit that allows plasma scattered light LP from the plasma PL to pass through. The light that passes through slit S1 is collimated by lens L1, diffracted by diffraction gratings G1 and G2, focused by lens L2, and incident on slit S2.

[0024] Slit S2 is configured as a so-called inverted slit, with a metal wire at its center. The metal wire has the function of blocking the zeroth-order diffracted light from the diffraction gratings G1 and G2. By matching the wavelength of the zeroth-order diffracted light with the wavelength of the laser light LA, only the plasma scattered light LP, which has a different wavelength from the laser light LA, can be extracted.

[0025] Light passing through slit S2 is collimated by lens L3, diffracted by diffraction gratings G3 and G4, focused by lens L4, reflected by mirror M1, and incident on slit S3. Light passing through slit S3 is reflected by mirror M2, collimated by lens L5, diffracted by diffraction gratings G5 and G6, focused by lens L6, reflected by mirror M3, and incident on photodetector DT. Photodetector DT is composed of an image sensor, such as an ICCD camera, and outputs a wavelength spectral signal showing the spatial distribution according to the wavelength of light.

[0026] When such a spectrometer is used, the stray light removal performance in the wavelength range more than 15 pm away from the wavelength of the laser light LA ​​is 10 -5 In summary, the spectrometer can achieve a signal transmittance of 10% or more and a solid angle of received scattered light of 5 mSr (millisteradian) or more, demonstrating its superior spectral performance.

[0027] Figure 4 is a graph showing an example of the wavelength spectral signal of the photodetector DT. The wavelength spectrum of laser Thomson scattering is distributed symmetrically around the wavelength λ0 of the laser light LA. In particular, in cooperative scattering, a double peak appears in the plasma scattered light LP based on the ion term. The distance between these peaks changes depending on the electron temperature of the plasma. The area of ​​the double peak changes depending on the electron density of the plasma.

[0028] Figure 4 illustrates the case where the peak-to-peak distance is approximately 0.2 nm. By employing the inverted slit S2 described above, it becomes possible to remove light within a range of ±14 pm (picometers) centered on the wavelength λ0. This significantly improves the signal-to-noise ratio of the plasma scattered light.

[0029] Figure 5 shows a magnified portion of Figure 2. An object 15 is placed behind the plasma PL. The laser light that has passed through the plasma PL collides with the object 15, generating a large amount of noise scattered light. Let's consider the case where the plasma scattered light LP, which propagates in the direction of angle θ with respect to the optical axis of the laser light LA, is measured. If the distance from the center of the plasma PL to the surface of the object 15 along the direction of laser light propagation is D, then the interval d between the plasma scattered light LP and the noise scattered light LN, which propagate in the direction of angle θ, is d = D × cos(90°-θ).

[0030] Figure 6 is a diagram showing an example of the entrance slit S1 of a spectrometer. The entrance slit S1 has two blades 31 and 32 that are movable in the X direction in the XY plane (parallel to the plane of the paper) perpendicular to the direction of propagation of the plasma scattered light LP. The edges of the blades 31 and 32 set the slit shape, and the slit shape is adjusted so that the beam of plasma scattered light LP can pass through, while blocking as much light other than the plasma scattered light LP as possible.

[0031] On the other hand, the beam of noise scattered light LN is shifted by an interval d in the YZ plane from the beam of plasma scattered light LP. Therefore, the two blades 31 and 32 cannot block the incidence of noise scattered light LN.

[0032] As a countermeasure, the entrance slit S1 further has a third blade 33 that is movable in the Y direction within the XY plane. The position of the blade 33 is adjusted to block the beam of noise scattered light LN. By employing an entrance slit S1 with these three blades, it is possible to efficiently prevent noise scattered light generated from object 15 from entering the spectrometer.

[0033] Furthermore, if a lens with magnification M is interposed between the plasma PL and the entrance slit S1, the distance d at the entrance slit S1 becomes d = D × M × cos(90° - θ).

[0034] Figure 7(A) is a front view showing a specific example of slit S1, and Figure 7(B) is a side view thereof. This is commercially available as a variable slit (product SLX-1, manufactured by Sigma Koki). The slit body 35 has a rectangular parallelepiped shape with an opening in the center and is equipped with a total of four blades: a pair of blades that can move left and right, and a pair of blades that can move up and down. The positions of the left and right blades can be adjusted by rotating knobs M31 and M32, and the positions of the up and down blades can be adjusted by rotating knobs M33 and M34.

[0035] Figure 8 is a graph showing an example of the wavelength spectrum of plasma scattered light. The vertical axis represents the signal intensity of the plasma scattered light (arbitrary units, linear), and the horizontal axis represents the shift amount from the central wavelength λ0. Plasma electron temperature T e It can be seen that as increases, the distance between peaks also increases. For example, electron temperature T e At 10 eV (approximately 100,000 K), the peak appears at λ0 ± 0.025 nm. e At 50 eV (approximately 500,000 K), the peak appears at λ0 ± 0.055 nm. e At 200 eV (approximately 2 million K), the peak appears at λ0 ± 0.15 nm.

[0036] Figure 9 is a graph showing the results of experimental measurements of the effect of noise scattered light using the apparatus shown in Figure 2. The vertical axis represents the signal intensity of the noise scattered light (arbitrary units, logarithmic scale), and the horizontal axis represents the amount of shift from the central wavelength λ0. The experimental conditions were θ=135° and magnification M=1 in all cases. The dashed line graph is for reference only. e It exhibits plasma scattering light with a double peak of =50 eV.

[0037] Curve Q1 shows the case where the distance D=1mm (d=0.7mm) from the center of the plasma PL to the surface of object 15 is used, with only the left and right blades 31 and 32 being used. Note that the dip near the top of curve Q1 is due to saturation of the measurement system, and in reality, it is a single-peak Gaussian shape. Because object 15 is extremely close to the plasma PL, the signal intensity of the noise scattered light is the largest, indicating that a large amount of noise scattered light LN is incident on the spectrometer from object 15. As a result, the double peak of the plasma scattered light is obscured by the noise scattered light and cannot be measured.

[0038] Curve Q2 shows the case when using the left and right blades 31, 32 and the upper blade 33 at a distance D=1 mm (d=0.7 mm). Although object 15 is very close to the plasma PL, the use of the upper blade 33 prevents noise scattered light from entering the spectrometer. Therefore, it can be seen that the noise scattered light is attenuated by about two orders of magnitude (about 20 dB) compared to curve Q1. As a result, the double peak of the plasma scattered light is slightly larger than that of the noise scattered light. It becomes possible to measure the distance between peaks, for example, in a medium-density plasma (10 23 m -3 ) can be measured.

[0039] Curve Q3 shows the case where the left and right blades 31, 32 and the upper blade 33 are used at a distance D=3mm (d=2.1mm). Since the object 15 is somewhat far from the plasma PL, the distance d is large, and the upper blade 33 can block more noise scattered light. Therefore, it can be seen that the noise scattered light is attenuated by about 1.5 orders of magnitude (about 15dB) compared to curve Q2. As a result, the double peak of the plasma scattered light is considerably larger than that of the noise scattered light. It becomes possible to measure the distance between peaks, for example, in low-density plasma (10 22 m -3 ) can be measured.

[0040] Curve Q4 shows the case where only the left and right blades are used at a distance D = 80 mm. Since object 15 is quite far from the plasma PL, the amount of noise scattered light incident on the spectrometer from object 15 is the smallest.

[0041] Thus, according to the present invention, 10 22 ~10 26 m -3 Plasmas with electron densities and electron temperatures of 10-200 eV can be measured. However, if the electron density is too low, the plasma scattered light becomes too weak. If the electron density is too high, reflection by free electrons increases, preventing the laser light from penetrating the plasma.

[0042] Next, LTS measurement of the plasma for the EUV light source will be described. Figure 10 is a perspective view showing an example of the LTS measurement setup. The droplet generator 50 continuously supplies the Sn (tin) target 51 downwards (-z direction). The picosecond pre-laser light, CO2 laser light, and measurement laser light (YAG laser second harmonic, 532 nm) travel along a direction 60 that is coaxial with the x-axis. EUV light intensity measurement is performed in the xy plane at a direction 61 150° to the x-axis, Thomson scattered light measurement is performed at a direction 62 120° to the x-axis, and EUV emission image and shadow measurement is performed at a direction 63 90° to the x-axis.

[0043] The Sn target 51 is, for example, a droplet with a diameter of 26 μm. When picosecond pre-laser light is shone onto this Sn target 51 to expand it into a mist, and then multiple pulses of light (for example, pulse widths of about 20 ns) are shone from a CO2 laser source at three different timings, the Sn target 51 generates EUV light.

[0044] Figures 11(A) to (D) show the shadow graphs of the Sn target. Figure 11(A) shows the state before irradiation with picosecond pulsed laser light, Figure 11(B) shows the state after irradiation with CO2 laser light with a delay time Δt = 1.3 μs (seconds) after irradiation with picosecond pulsed laser light, Figure 11(C) shows the state after irradiation with CO2 laser light with a delay time Δt = 2.0 μs after irradiation with picosecond pulsed laser light, and Figure 11(D) shows the state after irradiation with CO2 laser light with a delay time Δt = 2.5 μs after irradiation with picosecond pulsed laser light.

[0045] Figures 11(E) to (G) show the EUV (13.5 nm ± 1% width) emission distribution, with Figure 11(E) corresponding to Figure 11(B), Figure 11(F) to Figure 11(C), and Figure 11(G) to Figure 11(D). The EUV conversion efficiency η can be calculated from Figures 11(E) to (G), yielding η = 3.1%, 4.0%, and 2.8%, respectively. The results show that Δt = 2.0 μs was the most efficient, with a conversion efficiency of η reaching 4%.

[0046] Figure 12(A) shows the EUV emission distribution corresponding to Figure 11(E). The dashed line in the figure shows the outline of the Sn target. Figure 12(B) shows the electron density n obtained by LTS measurement. e (Graph above) and electron temperature T e The graph below shows the distribution. The solid line represents the optimal region (30-40 eV). Figure 12(C) shows the electron density n in Figure 12(B). e and electron temperature T e The emissivity distribution calculated based on this is shown. Figure 12(D) is a reconstructed graph of the graph in Figure 12(C).

[0047] Figure 12(E) is a graph showing the EUV emission process. The plasma (EUV light source) heated by the laser light irradiated from left to right in the figure flows toward the center, and it was found that EUV light is emitted particularly efficiently from the bright region in the figure.

[0048] Thus, LTS measurement makes it possible to measure the electron density and electron temperature of EUV light, and these parameters can be used to optimize the conditions for EUV emission efficiency. Therefore, this method makes it possible to improve the design efficiency and productivity of EUV light sources and maximize the output of EUV light sources. As a result, the present invention can make a significant contribution to the mass production technology of next-generation semiconductors with line widths of 3 nm or less. [Industrial applicability]

[0049] This invention is extremely useful industrially because it allows for more accurate measurements while easing the structural constraints on plasma measurement. [Explanation of Symbols]

[0050] 10 Vacuum container 11-13 windows 15 Object 20 Laser light sources 30. Spectrum measurement section 31, 32, 33 Blades 40 Arithmetic section LA laser light PL Plasma LP plasma scattering light LN Noise Scattered Light S1~S3 Slit G1~G5 Diffraction Grating

Claims

1. A laser light source that irradiates laser light onto the plasma, A spectrum measurement unit that receives scattered light from the plasma and measures its wavelength spectrum, The system includes a calculation unit that calculates the electron temperature and electron density of the plasma based on the measured wavelength spectrum, The spectral measurement unit includes a spectrometer having an entrance slit, The incident slit includes two blades movable in a first direction in a plane perpendicular to the direction of propagation of scattered light, and at least one blade movable in a second direction perpendicular to the first direction, and the slit shape is defined by the edges of at least three blades. The distance from the center of the plasma to the object on the incident optical axis of the measurement laser is 1 to 3 mm. The plasma measuring device is characterized in that the calculation unit compares the area of ​​the spectral optical signal intensity distribution showing a double peak in the plasma scattered light at an electron density Te calculated based on wavelengths λo±Δλ shifted from the central wavelength λo of the plasma scattered light spectrum with the area of ​​the noise scattered light intensity distribution in a shifted wavelength range (λo-Δλ, λo+Δλ) measured at the distance of the object from the center of the plasma determined by adjusting the incident slit, and determines a range of optical signal intensity distribution such that the spectral optical signal intensity showing a double peak in the plasma scattered light is higher than the noise scattered light intensity, thereby determining the electron density Te and electron temperature Ne at which EUV light can be efficiently emitted.

2. The plasma measuring apparatus according to claim 1, wherein the slit shape of the incident slit portion is set to prevent scattered light from an object placed at a predetermined distance from the center of the plasma along the direction of propagation of the laser light from entering the spectrometer, and to allow scattered light from the plasma to enter the spectrometer.

3. The plasma measuring apparatus according to claim 1, wherein the spectrometer includes a plurality of diffraction gratings and a plurality of slits, and at least one of the plurality of slits is an inverse slit that blocks the zeroth order of diffraction.

4. The plasma being measured is 10 ２２ ~10 ２６ I understand －３ The plasma measuring apparatus according to claim 1, having an electron density and an electron temperature of 10 to 200 eV.

Images