A mechanism for eliminating thermal drift in CD SEM sample stages

Abstract

This invention discloses a mechanism for eliminating thermal drift in a CD SEM sample stage, comprising the following steps: simultaneously measuring the sample stage position using an integrated laser interferometer to obtain the relative thermal drift of the sample stage; modifying the structure of the operating platform to ensure that the heat energy generated by the heat source of the X platform propagates approximately axisymmetrically within the sample stage; and adding water cooling to allow the heat energy generated by the heat source of the Y platform to be rapidly conducted away, without affecting the sample stage. This invention improves the structure of the CD SEM stage, suppressing thermal drift and enhancing in-situ stability. The excitation coil of the X platform is arranged at the center of the mover, ensuring uniform heat conduction with axisymmetric symmetry; the coil of the Y platform is located in the middle of the stator, and the stator is mounted on the bottom plate of a vacuum cavity with water-cooling pipelines, which can quickly remove Joule heat from the coil and isolate thermal interference. High-resolution laser interferometers are configured on three sides of the cavity to monitor the sample stage position and thermal drift in real time, enabling real-time correction of thermal drift and significantly improving the positioning accuracy of the equipment and the operational stability of the sample stage.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically a mechanism for eliminating thermal drift in a CD SEM sample stage. Background Technology

[0002] CD SEM is a metrology device used in semiconductor manufacturing processes. It monitors the stability and accuracy of chip manufacturing processes by measuring the critical dimensions (CD) of lithographic or etched patterns, thereby ensuring chip production yield. CD SEM requires extremely high repeatability in CD measurements; the international standard is 3σ less than 1% of the average measured dimension. Many factors affect the repeatability of CD SEM measurements, one important factor being the vibration frequency and relative amplitude between the electron gun and the sample stage when the system is subjected to external factors (such as ground vibration, air pressure fluctuations, and temperature changes). The smaller the relative amplitude, the higher the imaging resolution, and naturally, the higher the repeatability.

[0003] When the influence of external factors on the system is sufficiently small, reducing the relative amplitude of the electron gun and the sample stage is equivalent to improving the in-situ stability of the sample stage. For low- to mid-range CD SEMs, the in-situ stability of the sample stage should be less than 1 nanometer; for high-end CD SEMs, the in-situ stability of the sample stage should be less than 0.1 nanometers.

[0004] The movement of the sample stage in modern CD SEM (especially high-end CD SEM) is generally driven by a precision linear motor—a precision XY vacuum linear motion platform. The main advantages of the linear motion platform are: (1) high speed and large stroke; (2) ultra-high position resolution (theoretically only related to the ripple coefficient of the coil excitation current); (3) easy vacuum sealing, with the excitation coil and the mover or stator installed as a whole in the vacuum chamber.

[0005] Because the excitation coil is placed directly inside the vacuum chamber, the Joule heat generated by the excitation coil during platform operation becomes one of the main factors affecting the in-situ stability of the sample stage. It is difficult to find a suitable and rapid way to dissipate heat from a moving heat source in a vacuum; the heat will be conducted to the sample stage, causing creep in its dimensions due to thermal expansion and contraction. The coefficient of linear expansion for typical metals or ceramics is approximately 10 PPM (approximately 5 PPM for ceramics and 17 PPM for aluminum). With a sample stage size of approximately 400 mm, a temperature change of 0.1 millidegrees will result in a dimensional change approaching the nanometer scale. Thermal oscillations of the sample stage prevent the laser interferometer from obtaining an accurate position. Since the in-situ stability of the sample stage is controlled by the controller based on position feedback measured by the laser interferometer, thermal oscillations of the sample stage are a critical factor affecting the in-situ stability of the sample. Summary of the Invention

[0006] The purpose of this invention is to provide a mechanism for eliminating thermal drift in a CD SEM sample stage, thereby solving the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a mechanism for eliminating thermal drift in a CD SEM sample stage, comprising the following three steps: S1: The relative thermal drift of the sample stage is obtained while measuring the position of the sample stage using an integrated laser interferometer; S2: Change the structure of the operating platform so that the heat generated by the heat source of the X platform propagates approximately axisymmetrically in the sample stage; S3: Adding water cooling allows the heat generated by the heat source of the Y platform to be quickly conducted out without affecting the sample stage.

[0008] Preferably, the operating platform is installed inside a vacuum chamber, which is a cuboid enclosed by a base plate, surrounding walls, and a top cover. The surrounding walls, in top view, are rectangles, with their four walls designated as the left wall, right wall, top wall, and bottom wall. The left wall is used for wafer entry and exit. The precision linear operating platform is installed on the base plate of the chamber, located at the right center of the chamber; that is, when the sample stage is at the origin, the midpoint of the sample stage is equidistant from the top wall, bottom wall, and right wall.

[0009] Preferably, the operating platform consists of an X platform and a Y platform, arranged in a stacked structure, with the X platform on top and the Y platform below. The stator of the X platform is the mover of the Y platform. The sample stage is mounted on the X platform, located in its center. In reality, there should also be a Z platform and a θ platform on the X platform, but since they are almost irrelevant to the in-situ stability of the sample stage at the XY position, they are ignored here.

[0010] Preferably, the top view of the sample stage is a square (the side length should be greater than the diameter of the wafer), wherein the three surfaces directly opposite the upper wall, lower wall and right wall are fitted with reflectors (ultra-high flatness reflectors - quartz glass with aluminum coating) to reflect the laser beam to the corresponding interferometer.

[0011] Preferably, the excitation coil of the X platform is installed at the center of the mover, and the heat generated by the excitation coil can always propagate approximately axisilymetrically on the square mover. The excitation coil of the Y platform is installed on the stator of the Y platform, and the heat generated by the drive coil is mainly dissipated quickly through water cooling, having little impact on the mover of the X platform.

[0012] Preferably, a precision laser interferometer (accuracy below 0.1 nanometers) is installed on the upper wall, lower wall, and right wall of the vacuum cavity to measure the position of the sample stage. The readings of the interferometers on the upper and lower walls can determine the relative thermal expansion and contraction of the sample stage in various directions (because heat conduction on the sample stage is approximately axisymmetric), thereby eliminating measurement errors caused by thermal drift in the X and Y directions.

[0013] Preferably, the stator of the Y-platform is mounted on the bottom plate of the vacuum chamber, located at the right center of the chamber. A water-cooling pipe is installed below the center of the Y-platform to allow the heat generated by the Y-platform drive coil to be quickly and evenly dissipated.

[0014] Compared with existing technologies, the advantages of this invention are as follows: This invention eliminates thermal drift of the sample stage by improving the structure of the CD SEM, thereby improving the in-situ stability of the sample stage; the excitation coil of the X platform is located at the center of the mover, allowing the Joule heat generated by the excitation coil to be conducted approximately axisilytically through the sample stage; the excitation coil of the Y platform is located at the center of the stator, which is mounted on the bottom plate of the vacuum chamber, and the bottom plate of the chamber is equipped with water-cooling pipes; the Joule heat energy of the excitation coil is rapidly conducted out of the vacuum chamber through water cooling, so that it has almost no impact on the sample stage; three high-resolution laser interferometers are mounted on the three walls of the vacuum chamber. These interferometers can measure the relative position of the sample stage while simultaneously obtaining the relative thermal drift of the sample stage, thus eliminating thermal drift in real time and improving position measurement accuracy and in-situ stability. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the linear motion platform of the present invention.

[0016] Figure 2 This is a structural schematic diagram of the linear motion platform of the present invention from another perspective.

[0017] Figure 3 This is a schematic diagram of the XY linear motion platform of the present invention in the X direction.

[0018] Figure 4 This is a schematic diagram of the XY linear motion platform of the present invention in the Y direction.

[0019] Figure 5 This is a schematic diagram of the water cooling system installation of the present invention.

[0020] Figure 6 This is a schematic diagram of the structure and optical path of the integrated laser interferometer of the present invention.

[0021] Figure 7 This is a schematic diagram of the installation of the laser emitter of the present invention.

[0022] Figure 8 This is a schematic diagram illustrating the principle of obtaining thermal drift using a laser interferometer in this invention.

[0023] In the diagram: 11. Mover; 12. Stator; 13. Moving component; 131. Guide groove; 132. Slider; 14. Excitation coil; 15. Permanent magnet guide groove; 21. Mover of X platform; 22. Stator of X platform; 23. Stator of Y platform; 24. Drive coil of X platform; 25. Permanent magnet guide groove of X platform; 26. Drive coil of Y platform; 27. Permanent magnet guide groove of Y platform; 31. Vacuum cavity; 32. XY motion platform; 33. Wall; 34. Fine water 35. Tube; 36. Copper and stainless steel welding; 37. Flexible thermal pad; 38. Heat dissipation square copper plate; 49. Base plate; 40. Cavity wall; 41. Laser interferometer I; 42. Vacuum window; 43. Reflector I; 44. Reflector II; 45. Laser interferometer II; 46. Beam splitter I; 47. Beam splitter II; 48. Beam splitter II; 49. Reflector III; 410. Laser interferometer III; 411. Sample stage; 412. Reflector; 51. Laser reflector; 52. Reflector IV; 53. Reflector V. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Please see Figures 1-8 This invention provides a technical solution: a mechanism for eliminating thermal drift in a CD SEM sample stage, comprising the following three steps: S1: The relative thermal drift of the sample stage is obtained while measuring the position of the sample stage using an integrated laser interferometer; S2: Change the structure of the operating platform so that the heat generated by the heat source of the X platform propagates approximately axisymmetrically in the sample stage; S3: Adding water cooling allows the heat generated by the heat source of the Y platform to be quickly conducted out without affecting the sample stage.

[0026] The operating platform is installed inside a vacuum chamber, which is a cuboid enclosed by a base plate, surrounding walls, and a top cover. The top view of the surrounding walls is a rectangle, with its four walls designated as the left wall, right wall, top wall, and bottom wall. The left wall is used for wafer entry and exit. The precision linear operating platform is installed on the base plate of the chamber, located at the right center of the chamber; that is, when the sample stage is at the origin, the midpoint of the sample stage is equidistant from the top, bottom, and right walls. The operating platform consists of an X platform and a Y platform, arranged in a stacked structure, with the X platform on top and the Y platform below. The stator of the X platform is the mover of the Y platform. The sample stage is installed on the X platform, located at its center. In reality, the X platform should also have a Z platform and a θ platform, as they are almost irrelevant to the in-situ stability of the sample stage at the XY position and are ignored here. The top view of the sample stage is a square (the side length should be greater than the diameter of the wafer), which is separated from the top, bottom, and right walls. Three opposing surfaces are fitted with reflectors (ultra-high flatness reflectors—quartz glass with aluminum coating) to reflect the laser beam onto the corresponding interferometers. The excitation coil of the X-platform is installed in the center of the mover, and the heat generated by the excitation coil always propagates approximately axisymmetrically on the square mover. The excitation coil of the Y-platform is installed on the stator of the Y-platform, and the heat generated by the drive coil is mainly quickly dissipated through water cooling, having little impact on the mover of the X-platform. A precision laser interferometer (accuracy below 0.1 nanometers) is installed on the upper, lower, and right walls of the vacuum cavity to measure the position of the sample stage. The readings of the interferometers on the upper and lower walls can determine the relative thermal expansion and contraction of the sample stage in various directions (because heat conduction on the sample stage is approximately axisymmetric), thereby eliminating measurement errors caused by thermal drift in the X and Y directions. The stator of the Y-platform is installed on the bottom plate of the vacuum cavity, located in the right center of the cavity. A water-cooling pipe is installed below the center of the Y-platform so that the heat generated by the drive coil of the Y-platform can be quickly and evenly dissipated.

[0027] Working principle: Figure 1 and Figure 2 This is the structure of a linear motion platform; a linear motion platform in one direction consists of a mover 11 and a stator 12; the moving component 13 includes a guide groove 131 or a slider 132, and an excitation coil 14 or a permanent magnet guide groove 15; in the left figure, the mover 11 includes a permanent magnet guide groove 15, and the stator 12 includes an excitation coil 14; in the right figure, the opposite is true, the mover 11 includes an excitation coil 14, and the stator 12 includes a permanent magnet guide groove 15; in order to ensure that the mover 11 moves smoothly and accurately relative to the stator 12 (causing minimal vibration during movement), the motion platform generally includes a combination of multiple guide grooves 131 and sliders 132 symmetrically arranged on both sides; the magnetic field generated by the excitation current passing through the excitation coil 14 and the magnetic field of the permanent magnet guide groove 15 interact (attract or repel each other) to drive the mover 11 to move relative to the stator 12 along the permanent magnet guide groove; Figure 3 and Figure 4 It is an XY linear motion platform in the X direction ( Figure 3 ) and Y direction ( Figure 4 The X platform is stacked on the Y platform, and the stator 22 of the X platform is the mover of the Y platform; the excitation coil 24 of the X platform is located in the center of the mover 21 of the X platform, and the permanent magnet guide slot 25 of the X platform is located in the stator 22 of the X platform (which is also the mover of the Y platform); the drive coil 26 of the Y platform is located in the stator 23 of the Y platform, and the permanent magnet guide slot 27 of the Y platform is located in the stator 22 of the X platform (which is also the mover of the Y platform); if the sample stage 411 is installed in the center of the X platform, the Joule heat generated by the drive coil 24 of the X platform is conducted in the XY direction of the sample stage in approximately the same way, and the thermal expansion and contraction of the sample stage 411 in the XY direction is approximately equal at all times; Figure 5 This is a schematic diagram of the water-cooled installation. The XY linear motion platform 32 is mounted on the base plate 38 at the right center of the vacuum chamber 31. An opening, the same size as the heat dissipation square copper plate 37, is made in the center below the motion platform on the base plate 38 for mounting the heat dissipation copper plate 37. The heat dissipation copper plate 37 is welded to the base plate 38 around its perimeter to ensure a vacuum seal. The upper surface of the heat dissipation copper plate 37 should be slightly lower than the vacuum surface of the base plate 38 (approximately 0.1 mm lower) to prevent the heat dissipation plate 38 from pressing against the bottom of the XY motion platform 32 due to thermal expansion and contraction or atmospheric pressure. A piece of material approximately 0.2 mm thick is added on top of the heat dissipation copper plate 37. A flexible thermally conductive pad 36 (such as an indium sheet, with very low hardness, excellent thermal conductivity, and low outgassing rate) is used to ensure that the bottom of the XY motion platform 32 is pressed tightly against the flexible thermally conductive pad 36, and between the flexible thermally conductive pad 36 and the heat dissipation copper plate 37, thereby reducing thermal resistance. A row of small holes (approximately 0.5 mm in diameter recommended) with the same diameter and smooth inner walls is drilled along a certain direction on the heat dissipation copper plate 37 to serve as water-cooling pipes 34, and the ends of each hole are sealed by welding. Small holes of the same diameter are drilled on the bottom surface of the heat dissipation copper plate 37, angled towards each of the thin water pipes 34. The water-cooling equipment is connected to these holes via flexible hoses of the same diameter. The purpose is to allow water to flow uniformly and smoothly through these holes, preventing vibration caused by blockage or air ingress. This water cooling can quickly dissipate the heat generated by the excitation coil of the Y platform, ensuring it has no impact on the sample stage 411. Figure 6 This is the structure and optical path diagram of the integrated laser interferometer (XY cross-sectional view at the midpoint of the Z-axis of the sample stage); the laser beam travels along the Y direction, enters the beam splitter 48, and is split into two beams in the X and Y directions; the beam in the X direction is turned to the Y direction by the reflector 49, passes through the vacuum window 43 and enters the vacuum cavity, and then passes through the laser interferometer 410, which is firmly installed on the cavity wall, and is perpendicularly directed to the reflector 412 on the lower side of the sample stage; the laser beam is reflected back by the reflector and enters the laser interferometer 410, which can measure the relative distance from the lower side of the sample stage 411 to the lower wall of the cavity; The beam in the Y direction split by beam splitter 48 is split into beam in the X direction and beam in the Y direction by beam splitter 47. The beam in the X direction passes through vacuum window 43 and laser interferometer 46 and is directed to the reflector on the right side of the sample stage and reflected back to laser interferometer 46. In this way, the relative distance from the right side of the sample stage 411 to the right wall of the cavity can be measured. The Y-direction light split by beam splitter 47 passes through mirror 45, mirror 44, and laser interferometer 43 and enters the vacuum chamber from the upper wall of the vacuum chamber. The relative distance from the upper side of the sample stage 411 to the upper wall of the vacuum chamber can be measured.

[0028] The three laser interferometers (laser interferometer 1 42, laser interferometer 2 46 and laser interferometer 3 410) must be securely mounted on the three walls of the cavity and must be installed in a vacuum to avoid measurement errors caused by external pressure fluctuations and temperature changes; all the reflectors and beam splitters used are securely mounted on the outside of the vacuum cavity walls.

[0029] The three sides of the sample stage are made of artificial quartz glass with a surface flatness requirement of less than λ / 20. The reflective surface of the mirror is coated with a 30-nanometer-thick aluminum mold (aluminum has a high reflectivity for helium-neon laser beams with a wavelength of about 600 nanometers). Figure 7 This is a schematic diagram of the laser emitter installation; the laser emitter 51 should be securely mounted on a base next to the vacuum chamber; the laser beam emitted by the laser emitter 51 is directed along the X (or Y) direction towards the reflector 4 52, and after reflection, travels along the Z direction to reach the measurement plane, then is reflected again by the reflector 53, and travels along the Y direction on the measurement plane (see the following optical path diagram). Figure 7 ); Figure 8 This is a schematic diagram illustrating the principle of obtaining thermal drift using a laser interferometer. We assume the sample stage moves downwards by a distance S in the Y direction, and the temperature increases by ΔT. Due to the temperature change, the dimensions of the moving stage increase by ΔD1 in all directions, and the dimensions of the cavity increase by ΔD2 in all directions (in reality, because the cavity has a large heat capacity, the change in cavity dimensions caused by ambient temperature is very slow and small, and can be disregarded); the thermal drift is as follows: L2-L2ΔT=ΔD1-ΔD2,(1) The thermal drift above is: L1-L1ΔT=ΔD1-ΔD2 (2) Because heat is conducted approximately uniformly in all directions of the sample stage, the thermal drift of the sample stage in all four directions is ΔD1 - ΔD2. We need to obtain ΔD1 - ΔD2. L1ΔT and L2ΔT are obtained from real-time measurements using a laser interferometer, but L1 and L2 are unknown. From the figure, we can see... L1 + L2 = LB1 + LB2 LB1 and LB2 are the results measured by the motion table before this motion. Adding (1) and (2) above, we can obtain... 2*(ΔD1-ΔD2)=L1-L1ΔT+L2-L2ΔT=LB1+LB2-(L1ΔT+L2ΔT) so, ΔD1-ΔD2=(LB1+LB2-(L1ΔT+L2ΔT)) / 2 Therefore, by obtaining LB1+LB2 and using it as a reference, we can obtain the thermal drift of the sample stage at any given time in real time. We can obtain this reference using a laser interferometer before the system starts up (when the room temperature does not change much), and then obtain the thermal drift relative to this reference during subsequent real-time measurements and subtract it to eliminate the thermal drift.

[0030] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A mechanism for eliminating thermal drift in a CD SEM sample stage, characterized in that: It includes the following three steps: S1: The relative thermal drift of the sample stage is obtained while measuring the position of the sample stage using an integrated laser interferometer; S2: Change the structure of the operating platform so that the heat generated by the heat source of the X platform propagates approximately axisymmetrically in the sample stage; S3: Adding water cooling allows the heat generated by the heat source of the Y platform to be quickly conducted out without affecting the sample stage.

2. A mechanism for eliminating thermal drift in the CD SEM sample stage according to claim 1, characterized in that: The operating platform is installed inside a vacuum chamber, which is a cuboid enclosed by a base plate, walls, and a top cover. The top view of the walls is a rectangle, and its four walls are referred to as the left wall, right wall, top wall, and bottom wall, respectively. The left wall is used for the entry and exit of wafers. The precision linear operating platform is installed on the base plate of the chamber and is located in the right center of the chamber. That is, when the sample stage is at the origin, the distance from the midpoint of the sample stage to the top wall, bottom wall, and right wall is equal.

3. A mechanism for eliminating thermal drift in the CD SEM sample stage according to claim 2, characterized in that: The operating platform consists of an X platform and a Y platform, which are stacked in a structure with the X platform on top and the Y platform below. The stator of the X platform is the mover of the Y platform. The sample stage is mounted on the X platform and located in its center. In fact, there should also be a Z platform and a θ platform on the X platform, but since they are almost irrelevant to the in-situ stability of the sample stage at the XY position, they are ignored here.

4. A mechanism for eliminating thermal drift in the CD SEM sample stage according to claim 3, characterized in that: The sample stage is square in top view (the side length should be greater than the diameter of the wafer), and three surfaces that are directly opposite the top wall, bottom wall and right wall are covered with reflectors (ultra-high flatness reflectors - quartz glass with aluminum coating) to reflect the laser beam to the corresponding interferometer.

5. A mechanism for eliminating thermal drift in the CD SEM sample stage according to claim 4, characterized in that: The excitation coil of the X platform is installed in the center of the mover, and the heat generated by the excitation coil can always be propagated approximately axially symmetrically on the square mover; the excitation coil of the Y platform is installed on the stator of the Y platform, and the heat generated by the drive coil is mainly discharged quickly through water cooling, which has little impact on the mover of the X platform.

6. A mechanism for eliminating thermal drift in the CD SEM sample stage according to claim 5, characterized in that: Each of the upper, lower, and right walls of the vacuum chamber is equipped with a precision laser interferometer (accuracy below 0.1 nanometers) for measuring the position of the sample stage. The readings of the interferometers on the upper and lower walls can determine the relative thermal expansion and contraction of the sample stage in various directions (because heat conduction on the sample stage is approximately axisymmetric), thereby eliminating measurement errors caused by thermal drift in the X and Y directions.

7. A mechanism for eliminating thermal drift in a CD SEM sample stage according to claim 4, characterized in that: The stator of the Y-platform is mounted on the bottom plate of the vacuum chamber, located at the right center of the chamber. A water-cooling pipe is installed below the center of the Y-platform to allow the heat generated by the Y-platform drive coil to be quickly and evenly dissipated.

Images